osmith
/
wireshark
forked from osmocom/wireshark
1
0
Fork 0
Code Pull Requests Releases Wiki Activity
wireshark/ui/qt/rtp_player_dialog.cpp

1769 lines
60 KiB
C++
Raw Normal View History

Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
/* rtp_player_dialog.cpp
*
* Wireshark - Network traffic analyzer
* By Gerald Combs <gerald@wireshark.org>
* Copyright 1998 Gerald Combs
*
Fix comment end after SPDX identifier Move */ to a separate line below the SPDX identifier. Change-Id: Id1032215449cfccae0933147b45e04b65e0b727f Reviewed-on: https://code.wireshark.org/review/27211 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2018-04-30 07:47:58 +00:00
* SPDX-License-Identifier: GPL-2.0-or-later
*/
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
#include <ui/rtp_media.h>
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
#include "rtp_player_dialog.h"
qt: use #include <file.h> for generated include files make sure that generated include files are picked up only from the directories set by -I (or /I), not from the current directory if we use #include "file.h", Visual Studio searches for file.h in the same diretory as the source file that includes file.h if we do an out-of-tree build with cmake and the source directory contains files from an in-tree build (done with autotools), we might end up including the wrong file Change-Id: Iaaed2626258b6ff0c12485fe3f436bd03bbb5adf Reviewed-on: https://code.wireshark.org/review/15873 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2016-06-12 21:00:21 +00:00
#include <ui_rtp_player_dialog.h>
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
#ifdef QT_MULTIMEDIA_LIB
#include <epan/dissectors/packet-rtp.h>
tap: fix remaining potential memleaks with register_tap_listener Additionally, add an attribute to the tap function to prevent future callers from leaking this memory. Change-Id: Ief6af2bbc74d19153628f09d7b273e85cb2284ab Reviewed-on: https://code.wireshark.org/review/26642 Petri-Dish: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2018-03-25 21:24:59 +00:00
#include <wsutil/report_message.h>
Move utf8_entities.h to wsutil Change-Id: I6298b3de5f0a1cb988014ff16082eaf8c2a3c3c0 Reviewed-on: https://code.wireshark.org/review/10786 Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2015-10-04 17:10:29 +00:00
#include <wsutil/utf8_entities.h>
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
Qt: Move utils to separate utils directory Following the move for widgets directory, moving utils to the utils directory. Guidelines for this directory are: - Generic use but not a widget - Utility functionality used by many classes Note: additionally all includes affected by this move have been changed to absolute path includes, instead of relative ones. Change-Id: I019ae4b6e6f6d06a5745a63ed195edbd36fb936b Reviewed-on: https://code.wireshark.org/review/22602 Petri-Dish: Roland Knall <rknall@gmail.com> Reviewed-by: Roland Knall <rknall@gmail.com>
2017-07-25 14:15:18 +00:00
#include <ui/qt/utils/color_utils.h>
#include <ui/qt/widgets/qcustomplot.h>
#include <ui/qt/utils/qt_ui_utils.h>
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
#include "rtp_audio_stream.h"
Qt: Move utils to separate utils directory Following the move for widgets directory, moving utils to the utils directory. Guidelines for this directory are: - Generic use but not a widget - Utility functionality used by many classes Note: additionally all includes affected by this move have been changed to absolute path includes, instead of relative ones. Change-Id: I019ae4b6e6f6d06a5745a63ed195edbd36fb936b Reviewed-on: https://code.wireshark.org/review/22602 Petri-Dish: Roland Knall <rknall@gmail.com> Reviewed-by: Roland Knall <rknall@gmail.com>
2017-07-25 14:15:18 +00:00
#include <ui/qt/utils/tango_colors.h>
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
#include <widgets/rtp_audio_graph.h>
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
#include <QAudio>
Qt: Make the RTP player output device selectable. Add a combobox for selecting the output device and populate it with our available devices. Let the user know if our output format isn't supported. Ping-Bug: 13105 Change-Id: I299c7d0f191bb66d93896338036000e2c377781f Reviewed-on: https://code.wireshark.org/review/19046 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2016-12-02 23:52:02 +00:00
#include <QAudioDeviceInfo>
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
#include <QFrame>
#include <QMenu>
#include <QVBoxLayout>
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
#include <QAudioFormat>
#include <QAudioOutput>
#include <ui/qt/utils/rtp_audio_silence_generator.h>
RTP player: Always include QPushButton. It looks like QPushButton gets included via ui_rtp_player_dialog.h in Qt 5 but not in Qt 4. Make sure we include it explicitly whether or not QT_MULTIMEDIA_LIB is defined. Change-Id: I8203a1cc6f7b9beef0f749b93836a75885f85edd Reviewed-on: https://code.wireshark.org/review/10962 Reviewed-by: Gerald Combs <gerald@wireshark.org>
2015-10-12 17:15:09 +00:00
#endif // QT_MULTIMEDIA_LIB
Make sure we can compile without QtMultimedia. Change-Id: I8db453a735956435fc6e2e4276961adb1f7ed11a Reviewed-on: https://code.wireshark.org/review/10892 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2015-10-08 21:56:06 +00:00
#include <QPushButton>
Qt: Move utils to separate utils directory Following the move for widgets directory, moving utils to the utils directory. Guidelines for this directory are: - Generic use but not a widget - Utility functionality used by many classes Note: additionally all includes affected by this move have been changed to absolute path includes, instead of relative ones. Change-Id: I019ae4b6e6f6d06a5745a63ed195edbd36fb936b Reviewed-on: https://code.wireshark.org/review/22602 Petri-Dish: Roland Knall <rknall@gmail.com> Reviewed-by: Roland Knall <rknall@gmail.com>
2017-07-25 14:15:18 +00:00
#include <ui/qt/utils/stock_icon.h>
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
#include "wireshark_application.h"
// To do:
// - Fully implement shorcuts (drag, go to packet, etc.)
// - Figure out selection and highlighting.
// - Make streams checkable.
// - Add silence, drop & jitter indicators to the graph.
// - How to handle multiple channels?
// - Threaded decoding?
// - Play MP3s. As per Zawinski's Law we already read emails.
// - RTP audio streams are currently keyed on src addr + src port + dst addr
// + dst port + ssrc. This means that we can have multiple rtp_stream_info
// structs per RtpAudioStream. Should we make them 1:1 instead?
Qt: RTP audio stream fixups. Make sure audio_stream_ is non-NULL before we try to use it. Delete audio_stream_ more gracefully and add a note about mutexes on OS X and Windows. Bug: 12166 Change-Id: I12e76c49e631bc1de813c5c7d82c7d928c71237e Reviewed-on: https://code.wireshark.org/review/15759 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2016-06-06 21:55:10 +00:00
// Current and former RTP player bugs. Many have attachments that can be usef for testing.
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
// Bug 3368 - The timestamp line in a RTP or RTCP packet display's "Not Representable"
// Bug 3952 - VoIP Call RTP Player: audio played is corrupted when RFC2833 packets are present
// Bug 4960 - RTP Player: Audio and visual feedback get rapidly out of sync
// Bug 5527 - Adding arbitrary value to x-axis RTP player
// Bug 7935 - Wrong Timestamps in RTP Player-Decode
// Bug 8007 - UI gets confused on playing decoded audio in rtp_player
// Bug 9007 - Switching SSRC values in RTP stream
// Bug 10613 - RTP audio player crashes
// Bug 11125 - RTP Player does not show progress in selected stream in Window 7
// Bug 11409 - Wireshark crashes when using RTP player
Qt: RTP audio stream fixups. Make sure audio_stream_ is non-NULL before we try to use it. Delete audio_stream_ more gracefully and add a note about mutexes on OS X and Windows. Bug: 12166 Change-Id: I12e76c49e631bc1de813c5c7d82c7d928c71237e Reviewed-on: https://code.wireshark.org/review/15759 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2016-06-06 21:55:10 +00:00
// Bug 12166 - RTP audio player crashes
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
// XXX It looks like we duplicate some functionality here and in the RTP
// analysis code, which has its own routines for writing audio data to a
// file.
// In some places we match by conv/call number, in others we match by first frame.
enum {
rtp_player_dialog: Route audio for a stream to left/right speaker in RTP player Column 'Play' added to player. Double click on a stream in the column changes audio routing for the stream. When soundcard supports only one channel, there are Mute/Play option. When soundcard supports two or more channels, there are Mute/L/L+R/R options. Muted channel is drawn with dotted line. Change-Id: If120c902195da46f98a1663c589f20c6a1da0ba7 Reviewed-on: https://code.wireshark.org/review/35687 Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-07 21:16:42 +00:00
channel_col_,
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
src_addr_col_,
src_port_col_,
dst_addr_col_,
dst_port_col_,
ssrc_col_,
first_pkt_col_,
num_pkts_col_,
time_span_col_,
sample_rate_col_,
RTP Player: Added play rate of decoded audio It shows which rate wireshark really decoded the stream and how it will be played.
2021-03-29 06:49:51 +00:00
play_rate_col_,
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
payload_col_,
stream_data_col_ = src_addr_col_, // RtpAudioStream
RTP dissector: Remove unused functions Patch removes (just comments out) functions not used by the code as found by Martin Mathieson.
2021-02-05 09:19:48 +00:00
graph_audio_data_col_ = src_port_col_, // QCPGraph (wave)
graph_sequence_data_col_ = dst_addr_col_, // QCPGraph (sequence)
graph_jitter_data_col_ = dst_port_col_, // QCPGraph (jitter)
graph_timestamp_data_col_ = ssrc_col_, // QCPGraph (timestamp)
RTP player: It is possible to select SETUP packet related to RTP stream When SETUP (or related signalling packet) exists, it is selected by Shift+G. If RTP stream is "alone", first RTP packet of the stream is selected.
2021-03-09 13:47:07 +00:00
// first_pkt_col_ is skipped, it is used for real data
graph_silence_data_col_ = num_pkts_col_, // QCPGraph (silence)
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
};
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
class RtpPlayerTreeWidgetItem : public QTreeWidgetItem
{
public:
RtpPlayerTreeWidgetItem(QTreeWidget *tree) :
QTreeWidgetItem(tree)
{
}
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
bool operator< (const QTreeWidgetItem &other) const
{
// Handle numeric sorting
switch (treeWidget()->sortColumn()) {
case src_port_col_:
case dst_port_col_:
case num_pkts_col_:
case sample_rate_col_:
RTP player: It is possible to select SETUP packet related to RTP stream When SETUP (or related signalling packet) exists, it is selected by Shift+G. If RTP stream is "alone", first RTP packet of the stream is selected.
2021-03-09 13:47:07 +00:00
return text(treeWidget()->sortColumn()).toInt() < other.text(treeWidget()->sortColumn()).toInt();
RTP Player: Added play rate of decoded audio It shows which rate wireshark really decoded the stream and how it will be played.
2021-03-29 06:49:51 +00:00
case play_rate_col_:
return text(treeWidget()->sortColumn()).toInt() < other.text(treeWidget()->sortColumn()).toInt();
RTP player: It is possible to select SETUP packet related to RTP stream When SETUP (or related signalling packet) exists, it is selected by Shift+G. If RTP stream is "alone", first RTP packet of the stream is selected.
2021-03-09 13:47:07 +00:00
case first_pkt_col_:
int v1;
int v2;
v1 = data(first_pkt_col_, Qt::UserRole).toInt();
v2 = other.data(first_pkt_col_, Qt::UserRole).toInt();
return v1 < v2;
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
default:
// Fall back to string comparison
return QTreeWidgetItem::operator <(other);
break;
}
}
};
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
RtpPlayerDialog::RtpPlayerDialog(QWidget &parent, CaptureFile &cf) :
WiresharkDialog(parent, cf)
#ifdef QT_MULTIMEDIA_LIB
, ui(new Ui::RtpPlayerDialog)
rtp_player: Player is able to set start of audio play by double click Patch adds ability to set start of audio play by double clicking on waveform. Patch fixes unreported issue with placing waveform at incorrect place when switched relative/absolute time mode (check/uncheck Time of Day). Change-Id: Ib8ce24aea870e2443e033afbb6d6e9fbcf222431 Reviewed-on: https://code.wireshark.org/review/35621 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-01 23:38:35 +00:00
, first_stream_rel_start_time_(0.0)
, first_stream_abs_start_time_(0.0)
, first_stream_rel_stop_time_(0.0)
, streams_length_(0.0)
, start_marker_time_(0.0)
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
#endif // QT_MULTIMEDIA_LIB
Update to qcustomplot 2.0.1 Upgrade the internal version of QCustomPlot to 2.0.1 Change-Id: I1eb372d8e6a2f6c1bbdde4c74596785bf2d405c2 Reviewed-on: https://code.wireshark.org/review/17980 Reviewed-by: Roland Knall <rknall@gmail.com> Petri-Dish: Roland Knall <rknall@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2016-09-25 09:57:55 +00:00
, number_ticker_(new QCPAxisTicker)
, datetime_ticker_(new QCPAxisTickerDateTime)
rtp_player_dialog: Route audio for a stream to left/right speaker in RTP player Column 'Play' added to player. Double click on a stream in the column changes audio routing for the stream. When soundcard supports only one channel, there are Mute/Play option. When soundcard supports two or more channels, there are Mute/L/L+R/R options. Muted channel is drawn with dotted line. Change-Id: If120c902195da46f98a1663c589f20c6a1da0ba7 Reviewed-on: https://code.wireshark.org/review/35687 Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-07 21:16:42 +00:00
, stereo_available_(false)
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
, marker_stream_(0)
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
, last_ti_(0)
RTP Player: Dialog is nonmodal now and can be called multiple ways Changes: - refactored main_dialog handling of telephony dialogs - RTP Player dialog is nonmodal now and can be left open - it is possible to issue three actions on RTP Player dialog from other dialogs (other dialog have selected set of RTP streams before action) - replace - removes existing streams from RTP dialog and shows new set - add - adds new set to existing list in RTP dialog - remove - remove streams in set from list in RTP dialog - Sequence Dialog: - was modified to hold rtpstream_info_t for RTP streams - added Play button - VoIP features (RTP Play button, select/deselect RTP stream) are disabled after creation and must be enabled. It handles that RTP Play button is not shown e.g. in TCP sequence show
2021-03-27 20:01:43 +00:00
, listener_removed_(true)
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
{
ui->setupUi(this);
Qt: Add dialog geometry restore Add GeometryStateDialog class to handle load and save dialog geometry. The QDialog class name will be used as window name. For shared classes the UAT name or the statistics title or abbr will be used. Change-Id: I5a019598307fb3861518f41e733de834788184d8 Reviewed-on: https://code.wireshark.org/review/14139 Reviewed-by: Stig Bjørlykke <stig@bjorlykke.org>
2016-02-28 18:23:20 +00:00
loadGeometry(parent.width(), parent.height());
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
setWindowTitle(wsApp->windowTitleString(tr("RTP Player")));
ui->streamTreeWidget->installEventFilter(this);
ui->audioPlot->installEventFilter(this);
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
#ifdef QT_MULTIMEDIA_LIB
ui->splitter->setStretchFactor(0, 3);
ui->splitter->setStretchFactor(1, 1);
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
ui->streamTreeWidget->sortByColumn(first_pkt_col_, Qt::AscendingOrder);
RTP dissector: Remove unused functions Patch removes (just comments out) functions not used by the code as found by Martin Mathieson.
2021-02-05 09:19:48 +00:00
graph_ctx_menu_ = new QMenu(this);
graph_ctx_menu_->addAction(ui->actionZoomIn);
graph_ctx_menu_->addAction(ui->actionZoomOut);
graph_ctx_menu_->addAction(ui->actionReset);
graph_ctx_menu_->addSeparator();
graph_ctx_menu_->addAction(ui->actionMoveRight10);
graph_ctx_menu_->addAction(ui->actionMoveLeft10);
graph_ctx_menu_->addAction(ui->actionMoveRight1);
graph_ctx_menu_->addAction(ui->actionMoveLeft1);
graph_ctx_menu_->addSeparator();
graph_ctx_menu_->addAction(ui->actionGoToPacket);
RTP player: It is possible to select SETUP packet related to RTP stream When SETUP (or related signalling packet) exists, it is selected by Shift+G. If RTP stream is "alone", first RTP packet of the stream is selected.
2021-03-09 13:47:07 +00:00
graph_ctx_menu_->addAction(ui->actionGoToSetupPacketPlot);
RTP dissector: Remove unused functions Patch removes (just comments out) functions not used by the code as found by Martin Mathieson.
2021-02-05 09:19:48 +00:00
set_action_shortcuts_visible_in_context_menu(graph_ctx_menu_->actions());
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
ui->streamTreeWidget->setMouseTracking(true);
connect(ui->streamTreeWidget, SIGNAL(itemEntered(QTreeWidgetItem *, int)),
this, SLOT(itemEntered(QTreeWidgetItem *, int)));
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
connect(ui->audioPlot, SIGNAL(mouseMove(QMouseEvent*)),
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
this, SLOT(mouseMovePlot(QMouseEvent*)));
connect(ui->audioPlot, SIGNAL(mouseMove(QMouseEvent*)),
this, SLOT(mouseMovePlot(QMouseEvent*)));
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
connect(ui->audioPlot, SIGNAL(mousePress(QMouseEvent*)),
this, SLOT(graphClicked(QMouseEvent*)));
rtp_player: Player is able to set start of audio play by double click Patch adds ability to set start of audio play by double clicking on waveform. Patch fixes unreported issue with placing waveform at incorrect place when switched relative/absolute time mode (check/uncheck Time of Day). Change-Id: Ib8ce24aea870e2443e033afbb6d6e9fbcf222431 Reviewed-on: https://code.wireshark.org/review/35621 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-01 23:38:35 +00:00
connect(ui->audioPlot, SIGNAL(mouseDoubleClick(QMouseEvent*)),
this, SLOT(graphDoubleClicked(QMouseEvent*)));
RTP Player: Selected waveform is emphased. Waveform can be selected by click. When there is many waveforms, it is not clear which one is selected in list. It is important for selecting packets from graph. Patch emphasis selected waveform. In addition, it allows to select waveform by clicking on it.
2021-02-04 13:23:43 +00:00
connect(ui->audioPlot, SIGNAL(plottableClick(QCPAbstractPlottable*,int,QMouseEvent*)),
this, SLOT(plotClicked(QCPAbstractPlottable*,int,QMouseEvent*)));
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
cur_play_pos_ = new QCPItemStraightLine(ui->audioPlot);
cur_play_pos_->setVisible(false);
rtp_player: Player is able to set start of audio play by double click Patch adds ability to set start of audio play by double clicking on waveform. Patch fixes unreported issue with placing waveform at incorrect place when switched relative/absolute time mode (check/uncheck Time of Day). Change-Id: Ib8ce24aea870e2443e033afbb6d6e9fbcf222431 Reviewed-on: https://code.wireshark.org/review/35621 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-01 23:38:35 +00:00
start_marker_pos_ = new QCPItemStraightLine(ui->audioPlot);
start_marker_pos_->setPen(QPen(Qt::green,4));
setStartPlayMarker(0);
drawStartPlayMarker();
start_marker_pos_->setVisible(true);
Update to qcustomplot 2.0.1 Upgrade the internal version of QCustomPlot to 2.0.1 Change-Id: I1eb372d8e6a2f6c1bbdde4c74596785bf2d405c2 Reviewed-on: https://code.wireshark.org/review/17980 Reviewed-by: Roland Knall <rknall@gmail.com> Petri-Dish: Roland Knall <rknall@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2016-09-25 09:57:55 +00:00
datetime_ticker_->setDateTimeFormat("yyyy-MM-dd\nhh:mm:ss.zzz");
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
ui->audioPlot->xAxis->setNumberFormat("gb");
ui->audioPlot->xAxis->setNumberPrecision(3);
Update to qcustomplot 2.0.1 Upgrade the internal version of QCustomPlot to 2.0.1 Change-Id: I1eb372d8e6a2f6c1bbdde4c74596785bf2d405c2 Reviewed-on: https://code.wireshark.org/review/17980 Reviewed-by: Roland Knall <rknall@gmail.com> Petri-Dish: Roland Knall <rknall@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2016-09-25 09:57:55 +00:00
ui->audioPlot->xAxis->setTicker(datetime_ticker_);
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
ui->audioPlot->yAxis->setVisible(false);
ui->playButton->setIcon(StockIcon("media-playback-start"));
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
ui->playButton->setEnabled(false);
RTP player: Added pause button Pause button added. Button is visible only when media are played.
2020-12-29 13:41:14 +00:00
ui->pauseButton->setIcon(StockIcon("media-playback-pause"));
ui->pauseButton->setCheckable(true);
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
ui->pauseButton->setVisible(false);
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
ui->stopButton->setIcon(StockIcon("media-playback-stop"));
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
ui->stopButton->setEnabled(false);
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
Qt RTP: Add the default device and adjust sample rates. In the RTP player dialog, list the default audio device first, ensure it's selected by default and ensure that the list items are unique. According to http://code.qt.io/cgit/qt/qtmultimedia.git/tree/src/plugins/windowsaudio/qwindowsaudiodeviceinfo.cpp?h=5.9 the default device on Windows uses the special WAVE_MAPPER id, which appears to support various sample rates even when the underlying hardware doesn't. Ensuring the names are unique fixes an issue I'm seeing on a test machine here. When decoding, check to see if our sample rate is supported by our output device and adjust accordingly. Bug: 13906 Change-Id: Iddc0beb2459bfac42276ff29d227c2619b0a8d90 Reviewed-on: https://code.wireshark.org/review/22756 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2017-07-21 18:03:04 +00:00
// Ordered, unique device names starting with the system default
QMap<QString, bool> out_device_map; // true == default device
out_device_map.insert(QAudioDeviceInfo::defaultOutputDevice().deviceName(), true);
Qt: Make the RTP player output device selectable. Add a combobox for selecting the output device and populate it with our available devices. Let the user know if our output format isn't supported. Ping-Bug: 13105 Change-Id: I299c7d0f191bb66d93896338036000e2c377781f Reviewed-on: https://code.wireshark.org/review/19046 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2016-12-02 23:52:02 +00:00
foreach (QAudioDeviceInfo out_device, QAudioDeviceInfo::availableDevices(QAudio::AudioOutput)) {
Qt RTP: Add the default device and adjust sample rates. In the RTP player dialog, list the default audio device first, ensure it's selected by default and ensure that the list items are unique. According to http://code.qt.io/cgit/qt/qtmultimedia.git/tree/src/plugins/windowsaudio/qwindowsaudiodeviceinfo.cpp?h=5.9 the default device on Windows uses the special WAVE_MAPPER id, which appears to support various sample rates even when the underlying hardware doesn't. Ensuring the names are unique fixes an issue I'm seeing on a test machine here. When decoding, check to see if our sample rate is supported by our output device and adjust accordingly. Bug: 13906 Change-Id: Iddc0beb2459bfac42276ff29d227c2619b0a8d90 Reviewed-on: https://code.wireshark.org/review/22756 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2017-07-21 18:03:04 +00:00
if (!out_device_map.contains(out_device.deviceName())) {
out_device_map.insert(out_device.deviceName(), false);
}
}
foreach (QString out_name, out_device_map.keys()) {
Qt: Make the RTP player output device selectable. Add a combobox for selecting the output device and populate it with our available devices. Let the user know if our output format isn't supported. Ping-Bug: 13105 Change-Id: I299c7d0f191bb66d93896338036000e2c377781f Reviewed-on: https://code.wireshark.org/review/19046 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2016-12-02 23:52:02 +00:00
ui->outputDeviceComboBox->addItem(out_name);
Qt RTP: Add the default device and adjust sample rates. In the RTP player dialog, list the default audio device first, ensure it's selected by default and ensure that the list items are unique. According to http://code.qt.io/cgit/qt/qtmultimedia.git/tree/src/plugins/windowsaudio/qwindowsaudiodeviceinfo.cpp?h=5.9 the default device on Windows uses the special WAVE_MAPPER id, which appears to support various sample rates even when the underlying hardware doesn't. Ensuring the names are unique fixes an issue I'm seeing on a test machine here. When decoding, check to see if our sample rate is supported by our output device and adjust accordingly. Bug: 13906 Change-Id: Iddc0beb2459bfac42276ff29d227c2619b0a8d90 Reviewed-on: https://code.wireshark.org/review/22756 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2017-07-21 18:03:04 +00:00
if (out_device_map.value(out_name)) {
ui->outputDeviceComboBox->setCurrentIndex(ui->outputDeviceComboBox->count() - 1);
Qt: Make the RTP player output device selectable. Add a combobox for selecting the output device and populate it with our available devices. Let the user know if our output format isn't supported. Ping-Bug: 13105 Change-Id: I299c7d0f191bb66d93896338036000e2c377781f Reviewed-on: https://code.wireshark.org/review/19046 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2016-12-02 23:52:02 +00:00
}
}
if (ui->outputDeviceComboBox->count() < 1) {
ui->outputDeviceComboBox->setEnabled(false);
ui->playButton->setEnabled(false);
RTP player: Added pause button Pause button added. Button is visible only when media are played.
2020-12-29 13:41:14 +00:00
ui->pauseButton->setEnabled(false);
Qt: Make the RTP player output device selectable. Add a combobox for selecting the output device and populate it with our available devices. Let the user know if our output format isn't supported. Ping-Bug: 13105 Change-Id: I299c7d0f191bb66d93896338036000e2c377781f Reviewed-on: https://code.wireshark.org/review/19046 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2016-12-02 23:52:02 +00:00
ui->stopButton->setEnabled(false);
ui->outputDeviceComboBox->addItem(tr("No devices available"));
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
} else {
stereo_available_ = isStereoAvailable();
Qt: Make the RTP player output device selectable. Add a combobox for selecting the output device and populate it with our available devices. Let the user know if our output format isn't supported. Ping-Bug: 13105 Change-Id: I299c7d0f191bb66d93896338036000e2c377781f Reviewed-on: https://code.wireshark.org/review/19046 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2016-12-02 23:52:02 +00:00
}
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
ui->audioPlot->setMouseTracking(true);
ui->audioPlot->setEnabled(true);
ui->audioPlot->setInteractions(
QCP::iRangeDrag |
QCP::iRangeZoom
);
RTP dissector: Remove unused functions Patch removes (just comments out) functions not used by the code as found by Martin Mathieson.
2021-02-05 09:19:48 +00:00
graph_ctx_menu_->addSeparator();
list_ctx_menu_ = new QMenu(this);
RTP Player: Added additional shortcuts Added shortcuts: - Mute/Unmute/Invert Mute - Play/Pause/Stop Removed outdated help in tooltip text. It makes no sense to use it, all actions are available in context menu.
2021-03-11 08:22:15 +00:00
list_ctx_menu_->addAction(ui->actionPlay);
graph_ctx_menu_->addAction(ui->actionPlay);
list_ctx_menu_->addAction(ui->actionStop);
graph_ctx_menu_->addAction(ui->actionStop);
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
QMenu *selection_menu1 = list_ctx_menu_->addMenu(tr("Select"));
QMenu *selection_menu2 = graph_ctx_menu_->addMenu(tr("Select"));
selection_menu1->addAction(ui->actionSelectAll);
selection_menu2->addAction(ui->actionSelectAll);
selection_menu1->addAction(ui->actionSelectNone);
selection_menu2->addAction(ui->actionSelectNone);
selection_menu1->addAction(ui->actionSelectInvert);
selection_menu2->addAction(ui->actionSelectInvert);
RTP dissector: Remove unused functions Patch removes (just comments out) functions not used by the code as found by Martin Mathieson.
2021-02-05 09:19:48 +00:00
QMenu *audio_routing_menu1 = list_ctx_menu_->addMenu(tr("Audio Routing"));
QMenu *audio_routing_menu2 = graph_ctx_menu_->addMenu(tr("Audio Routing"));
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
// All AudioRouting actions are in menu, some of them are disabled later
RTP Player: Added additional shortcuts Added shortcuts: - Mute/Unmute/Invert Mute - Play/Pause/Stop Removed outdated help in tooltip text. It makes no sense to use it, all actions are available in context menu.
2021-03-11 08:22:15 +00:00
audio_routing_menu1->addAction(ui->actionAudioRoutingMute);
audio_routing_menu2->addAction(ui->actionAudioRoutingMute);
audio_routing_menu1->addAction(ui->actionAudioRoutingUnmute);
audio_routing_menu2->addAction(ui->actionAudioRoutingUnmute);
audio_routing_menu1->addAction(ui->actionAudioRoutingMuteInvert);
audio_routing_menu2->addAction(ui->actionAudioRoutingMuteInvert);
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
audio_routing_menu1->addAction(ui->actionAudioRoutingP);
audio_routing_menu2->addAction(ui->actionAudioRoutingP);
audio_routing_menu1->addAction(ui->actionAudioRoutingL);
audio_routing_menu2->addAction(ui->actionAudioRoutingL);
audio_routing_menu1->addAction(ui->actionAudioRoutingLR);
audio_routing_menu2->addAction(ui->actionAudioRoutingLR);
audio_routing_menu1->addAction(ui->actionAudioRoutingR);
audio_routing_menu2->addAction(ui->actionAudioRoutingR);
RTP Player: Possibility to remove stream from RTP player When many streams are analyzed, some streams can be found as useless during replay. Patch adds option to remove it from list in RTP player. Once stream is removed, it can't be returned except close/reopen RTP player with new selection.
2021-02-04 14:03:45 +00:00
list_ctx_menu_->addAction(ui->actionRemoveStream);
graph_ctx_menu_->addAction(ui->actionRemoveStream);
RTP player: It is possible to select SETUP packet related to RTP stream When SETUP (or related signalling packet) exists, it is selected by Shift+G. If RTP stream is "alone", first RTP packet of the stream is selected.
2021-03-09 13:47:07 +00:00
list_ctx_menu_->addAction(ui->actionGoToSetupPacketTree);
RTP dissector: Remove unused functions Patch removes (just comments out) functions not used by the code as found by Martin Mathieson.
2021-02-05 09:19:48 +00:00
set_action_shortcuts_visible_in_context_menu(list_ctx_menu_->actions());
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
#endif // QT_MULTIMEDIA_LIB
}
RTP Player: Dialog is nonmodal now and can be called multiple ways Changes: - refactored main_dialog handling of telephony dialogs - RTP Player dialog is nonmodal now and can be left open - it is possible to issue three actions on RTP Player dialog from other dialogs (other dialog have selected set of RTP streams before action) - replace - removes existing streams from RTP dialog and shows new set - add - adds new set to existing list in RTP dialog - remove - remove streams in set from list in RTP dialog - Sequence Dialog: - was modified to hold rtpstream_info_t for RTP streams - added Play button - VoIP features (RTP Play button, select/deselect RTP stream) are disabled after creation and must be enabled. It handles that RTP Play button is not shown e.g. in TCP sequence show
2021-03-27 20:01:43 +00:00
// _U_ is used when no QT_MULTIMEDIA_LIB is available
QPushButton *RtpPlayerDialog::addPlayerButton(QDialogButtonBox *button_box, QDialog *dialog _U_)
Make sure we can compile without QtMultimedia. Change-Id: I8db453a735956435fc6e2e4276961adb1f7ed11a Reviewed-on: https://code.wireshark.org/review/10892 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2015-10-08 21:56:06 +00:00
{
if (!button_box) return NULL;
QPushButton *player_button;
VoIP UI: Added shortcuts for dialog buttons VoIP/SIP Calls, Sequence, RTP Streams, RTP Player and RTP Analysis dialogs have shortcuts assigned. Shortcuts are same over all dialogs.
2021-03-23 10:37:59 +00:00
player_button = button_box->addButton(tr("&Play Streams"), QDialogButtonBox::ActionRole);
player_button->setToolTip(tr("Open RTP player dialog"));
Make sure we can compile without QtMultimedia. Change-Id: I8db453a735956435fc6e2e4276961adb1f7ed11a Reviewed-on: https://code.wireshark.org/review/10892 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2015-10-08 21:56:06 +00:00
player_button->setIcon(StockIcon("media-playback-start"));
RTP Player: Dialog is nonmodal now and can be called multiple ways Changes: - refactored main_dialog handling of telephony dialogs - RTP Player dialog is nonmodal now and can be left open - it is possible to issue three actions on RTP Player dialog from other dialogs (other dialog have selected set of RTP streams before action) - replace - removes existing streams from RTP dialog and shows new set - add - adds new set to existing list in RTP dialog - remove - remove streams in set from list in RTP dialog - Sequence Dialog: - was modified to hold rtpstream_info_t for RTP streams - added Play button - VoIP features (RTP Play button, select/deselect RTP stream) are disabled after creation and must be enabled. It handles that RTP Play button is not shown e.g. in TCP sequence show
2021-03-27 20:01:43 +00:00
#if defined(QT_MULTIMEDIA_LIB)
QMenu *button_menu = new QMenu(player_button);
button_menu->setToolTipsVisible(true);
QAction *ca;
ca = button_menu->addAction(tr("Re&place"));
ca->setToolTip(tr("Replace existing streams in RTP Player with new set"));
connect(ca, SIGNAL(triggered()), dialog, SLOT(rtpPlayerReplace()));
ca = button_menu->addAction(tr("&Add"));
ca->setToolTip(tr("Add new set to existing list of streams in RTP Player"));
connect(ca, SIGNAL(triggered()), dialog, SLOT(rtpPlayerAdd()));
ca = button_menu->addAction(tr("&Remove"));
ca->setToolTip(tr("Remove selected streams from list of streams in RTP Player"));
connect(ca, SIGNAL(triggered()), dialog, SLOT(rtpPlayerRemove()));
player_button->setMenu(button_menu);
#else
player_button->setEnabled(false);
player_button->setText(tr("No Audio"));
#endif
Make sure we can compile without QtMultimedia. Change-Id: I8db453a735956435fc6e2e4276961adb1f7ed11a Reviewed-on: https://code.wireshark.org/review/10892 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2015-10-08 21:56:06 +00:00
return player_button;
}
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
#ifdef QT_MULTIMEDIA_LIB
RtpPlayerDialog::~RtpPlayerDialog()
{
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
if (marker_stream_) {
marker_stream_->stop();
delete marker_stream_;
}
for (int row = 0; row < ui->streamTreeWidget->topLevelItemCount(); row++) {
QTreeWidgetItem *ti = ui->streamTreeWidget->topLevelItem(row);
RtpAudioStream *audio_stream = ti->data(stream_data_col_, Qt::UserRole).value<RtpAudioStream*>();
if (audio_stream)
delete audio_stream;
}
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
delete ui;
}
Fix crash in RTP Player on stop and close When dragging the UI, this somehow causes a great lag. Then by spam-clicking on the Stop button, a double free seems to occur. Fix this by moving the audio cleanup to the outputStateChanged callback as documented at https://doc.qt.io/qt-5/qaudiooutput.html. Note that calling stop() in the IdleState also triggers a change event, resulting in the desired cleanup. Stop streams before the dialog is closed (via accept/reject). This *cannot* be done in the destrutor of RtpPlayerDialog because destructing QAudioOutput processes events from the event queue, resulting in preature destruction of other objects... crash. Change-Id: I6bfb33c9396e9bc1ffd346519d22390a97b6bdaf Reviewed-on: https://code.wireshark.org/review/11894 Petri-Dish: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2015-11-17 17:54:41 +00:00
void RtpPlayerDialog::accept()
{
RTP Player: Fix for #16452 Reason for #16452 is that RTP Player dialog is destroyed before tap finishes when Escape key pressed. The only solution I found is to check whether dialog does exists when tap finishes. If so, data are processed and shown. If dialog was destroyed in meanwhile, no data processing is done.
2021-01-02 00:58:21 +00:00
if (!listener_removed_) {
remove_tap_listener(this);
listener_removed_ = true;
}
Fix crash in RTP Player on stop and close When dragging the UI, this somehow causes a great lag. Then by spam-clicking on the Stop button, a double free seems to occur. Fix this by moving the audio cleanup to the outputStateChanged callback as documented at https://doc.qt.io/qt-5/qaudiooutput.html. Note that calling stop() in the IdleState also triggers a change event, resulting in the desired cleanup. Stop streams before the dialog is closed (via accept/reject). This *cannot* be done in the destrutor of RtpPlayerDialog because destructing QAudioOutput processes events from the event queue, resulting in preature destruction of other objects... crash. Change-Id: I6bfb33c9396e9bc1ffd346519d22390a97b6bdaf Reviewed-on: https://code.wireshark.org/review/11894 Petri-Dish: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2015-11-17 17:54:41 +00:00
int row_count = ui->streamTreeWidget->topLevelItemCount();
// Stop all streams before the dialogs are closed.
for (int row = 0; row < row_count; row++) {
QTreeWidgetItem *ti = ui->streamTreeWidget->topLevelItem(row);
RtpAudioStream *audio_stream = ti->data(stream_data_col_, Qt::UserRole).value<RtpAudioStream*>();
audio_stream->stopPlaying();
}
WiresharkDialog::accept();
}
void RtpPlayerDialog::reject()
{
RtpPlayerDialog::accept();
}
Split RTP player tapping, decoding, and plotting. In RtpAudioStream split tapping+decoding into separate member functions. Store RTP payloads in memory. In RtpPlayerDialog split tapping+plotting. This more closely resembles what we're doing in the GTK+ UI and paves the way for jitter support and other changes. Change-Id: I244c225cec8930545622e6582b7be35ebe45b237 Reviewed-on: https://code.wireshark.org/review/11195 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2015-10-20 15:31:52 +00:00
void RtpPlayerDialog::retapPackets()
{
RTP Player: Dialog is nonmodal now and can be called multiple ways Changes: - refactored main_dialog handling of telephony dialogs - RTP Player dialog is nonmodal now and can be left open - it is possible to issue three actions on RTP Player dialog from other dialogs (other dialog have selected set of RTP streams before action) - replace - removes existing streams from RTP dialog and shows new set - add - adds new set to existing list in RTP dialog - remove - remove streams in set from list in RTP dialog - Sequence Dialog: - was modified to hold rtpstream_info_t for RTP streams - added Play button - VoIP features (RTP Play button, select/deselect RTP stream) are disabled after creation and must be enabled. It handles that RTP Play button is not shown e.g. in TCP sequence show
2021-03-27 20:01:43 +00:00
if (!listener_removed_) {
// Retap is running, nothing better we can do
return;
}
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
ui->hintLabel->setText("<i><small>" + tr("Decoding streams...") + "</i></small>");
wsApp->processEvents();
RTP Player: Dialog is nonmodal now and can be called multiple ways Changes: - refactored main_dialog handling of telephony dialogs - RTP Player dialog is nonmodal now and can be left open - it is possible to issue three actions on RTP Player dialog from other dialogs (other dialog have selected set of RTP streams before action) - replace - removes existing streams from RTP dialog and shows new set - add - adds new set to existing list in RTP dialog - remove - remove streams in set from list in RTP dialog - Sequence Dialog: - was modified to hold rtpstream_info_t for RTP streams - added Play button - VoIP features (RTP Play button, select/deselect RTP stream) are disabled after creation and must be enabled. It handles that RTP Play button is not shown e.g. in TCP sequence show
2021-03-27 20:01:43 +00:00
// Clear packets from existing streams before retap
for (int row = 0; row < ui->streamTreeWidget->topLevelItemCount(); row++) {
QTreeWidgetItem *ti = ui->streamTreeWidget->topLevelItem(row);
RtpAudioStream *row_stream = ti->data(stream_data_col_, Qt::UserRole).value<RtpAudioStream*>();
row_stream->clearPackets();
}
RTP Player: Fix for #16452 Reason for #16452 is that RTP Player dialog is destroyed before tap finishes when Escape key pressed. The only solution I found is to check whether dialog does exists when tap finishes. If so, data are processed and shown. If dialog was destroyed in meanwhile, no data processing is done.
2021-01-02 00:58:21 +00:00
// destroyCheck is protection againts destroying dialog during recap.
// It stores dialog pointer in data() and if dialog destroyed, it
// returns null
QPointer<RtpPlayerDialog> destroyCheck=this;
tap: fix remaining potential memleaks with register_tap_listener Additionally, add an attribute to the tap function to prevent future callers from leaking this memory. Change-Id: Ief6af2bbc74d19153628f09d7b273e85cb2284ab Reviewed-on: https://code.wireshark.org/review/26642 Petri-Dish: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2018-03-25 21:24:59 +00:00
GString *error_string;
RTP Player: Dialog is nonmodal now and can be called multiple ways Changes: - refactored main_dialog handling of telephony dialogs - RTP Player dialog is nonmodal now and can be left open - it is possible to issue three actions on RTP Player dialog from other dialogs (other dialog have selected set of RTP streams before action) - replace - removes existing streams from RTP dialog and shows new set - add - adds new set to existing list in RTP dialog - remove - remove streams in set from list in RTP dialog - Sequence Dialog: - was modified to hold rtpstream_info_t for RTP streams - added Play button - VoIP features (RTP Play button, select/deselect RTP stream) are disabled after creation and must be enabled. It handles that RTP Play button is not shown e.g. in TCP sequence show
2021-03-27 20:01:43 +00:00
listener_removed_ = false;
Add a tap "finish" callback, called when a listener is removed. Change-Id: Ic6c23dbd39d1adf8f730f1c866e409f731947475 Reviewed-on: https://code.wireshark.org/review/28786 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2018-07-21 00:07:19 +00:00
error_string = register_tap_listener("rtp", this, NULL, 0, NULL, tapPacket, NULL, NULL);
tap: fix remaining potential memleaks with register_tap_listener Additionally, add an attribute to the tap function to prevent future callers from leaking this memory. Change-Id: Ief6af2bbc74d19153628f09d7b273e85cb2284ab Reviewed-on: https://code.wireshark.org/review/26642 Petri-Dish: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2018-03-25 21:24:59 +00:00
if (error_string) {
report_failure("RTP Player - tap registration failed: %s", error_string->str);
g_string_free(error_string, TRUE);
return;
}
Split RTP player tapping, decoding, and plotting. In RtpAudioStream split tapping+decoding into separate member functions. Store RTP payloads in memory. In RtpPlayerDialog split tapping+plotting. This more closely resembles what we're doing in the GTK+ UI and paves the way for jitter support and other changes. Change-Id: I244c225cec8930545622e6582b7be35ebe45b237 Reviewed-on: https://code.wireshark.org/review/11195 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2015-10-20 15:31:52 +00:00
cap_file_.retapPackets();
RTP Player: Fix for #16452 Reason for #16452 is that RTP Player dialog is destroyed before tap finishes when Escape key pressed. The only solution I found is to check whether dialog does exists when tap finishes. If so, data are processed and shown. If dialog was destroyed in meanwhile, no data processing is done.
2021-01-02 00:58:21 +00:00
// Check if dialog exists still
if (destroyCheck.data()) {
if (!listener_removed_) {
remove_tap_listener(this);
listener_removed_ = true;
}
rescanPackets(true);
}
Split RTP player tapping, decoding, and plotting. In RtpAudioStream split tapping+decoding into separate member functions. Store RTP payloads in memory. In RtpPlayerDialog split tapping+plotting. This more closely resembles what we're doing in the GTK+ UI and paves the way for jitter support and other changes. Change-Id: I244c225cec8930545622e6582b7be35ebe45b237 Reviewed-on: https://code.wireshark.org/review/11195 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2015-10-20 15:31:52 +00:00
}
void RtpPlayerDialog::rescanPackets(bool rescale_axes)
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
{
RTP Player: Added notification message during long decoding When RTP player updates view, it might take long time. So message is shown in hint area.
2021-02-04 12:51:44 +00:00
// Show information for a user - it can last long time...
ui->hintLabel->setText("<i><small>" + tr("Decoding streams...") + "</i></small>");
wsApp->processEvents();
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
QAudioDeviceInfo cur_out_device = getCurrentDeviceInfo();
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
int row_count = ui->streamTreeWidget->topLevelItemCount();
rtp_player_dialog: Route audio for a stream to left/right speaker in RTP player Column 'Play' added to player. Double click on a stream in the column changes audio routing for the stream. When soundcard supports only one channel, there are Mute/Play option. When soundcard supports two or more channels, there are Mute/L/L+R/R options. Muted channel is drawn with dotted line. Change-Id: If120c902195da46f98a1663c589f20c6a1da0ba7 Reviewed-on: https://code.wireshark.org/review/35687 Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-07 21:16:42 +00:00
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
// Reset stream values
for (int row = 0; row < row_count; row++) {
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
QTreeWidgetItem *ti = ui->streamTreeWidget->topLevelItem(row);
RtpAudioStream *audio_stream = ti->data(stream_data_col_, Qt::UserRole).value<RtpAudioStream*>();
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
audio_stream->setStereoRequired(stereo_available_);
audio_stream->reset(first_stream_rel_start_time_);
audio_stream->setJitterBufferSize((int) ui->jitterSpinBox->value());
RtpAudioStream::TimingMode timing_mode = RtpAudioStream::JitterBuffer;
switch (ui->timingComboBox->currentIndex()) {
case RtpAudioStream::RtpTimestamp:
timing_mode = RtpAudioStream::RtpTimestamp;
break;
case RtpAudioStream::Uninterrupted:
timing_mode = RtpAudioStream::Uninterrupted;
break;
default:
break;
rtp_player_dialog: Route audio for a stream to left/right speaker in RTP player Column 'Play' added to player. Double click on a stream in the column changes audio routing for the stream. When soundcard supports only one channel, there are Mute/Play option. When soundcard supports two or more channels, there are Mute/L/L+R/R options. Muted channel is drawn with dotted line. Change-Id: If120c902195da46f98a1663c589f20c6a1da0ba7 Reviewed-on: https://code.wireshark.org/review/35687 Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-07 21:16:42 +00:00
}
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
audio_stream->setTimingMode(timing_mode);
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
audio_stream->decode(cur_out_device);
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
}
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
for (int col = 0; col < ui->streamTreeWidget->columnCount() - 1; col++) {
ui->streamTreeWidget->resizeColumnToContents(col);
}
createPlot(rescale_axes);
updateWidgets();
}
void RtpPlayerDialog::createPlot(bool rescale_axes)
{
RTP Player: Fixed incorrect handling of legend in graph Legend and it's elements were shown only when first waveform requires it. Patch fixes it and all legend elements are shown when required by any waveform.
2021-03-22 10:25:41 +00:00
bool legend_out_of_sequence = false;
bool legend_jitter_dropped = false;
bool legend_wrong_timestamps = false;
bool legend_inserted_silences = false;
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
bool relative_timestamps = !ui->todCheckBox->isChecked();
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
int row_count = ui->streamTreeWidget->topLevelItemCount();
RTP Player: UI improvements Changes: - all waveforms has common scale therefore louder/quiter signal is visible - when stream/streams are deleted from view, Y axis is rescaled and waveforms are rearranged to reuse empty space
2021-03-23 08:48:20 +00:00
gint16 total_max_sample_value = 1;
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
ui->audioPlot->clearGraphs();
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
Update to qcustomplot 2.0.1 Upgrade the internal version of QCustomPlot to 2.0.1 Change-Id: I1eb372d8e6a2f6c1bbdde4c74596785bf2d405c2 Reviewed-on: https://code.wireshark.org/review/17980 Reviewed-by: Roland Knall <rknall@gmail.com> Petri-Dish: Roland Knall <rknall@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2016-09-25 09:57:55 +00:00
if (relative_timestamps) {
ui->audioPlot->xAxis->setTicker(number_ticker_);
} else {
ui->audioPlot->xAxis->setTicker(datetime_ticker_);
}
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
RTP Player: UI improvements Changes: - all waveforms has common scale therefore louder/quiter signal is visible - when stream/streams are deleted from view, Y axis is rescaled and waveforms are rearranged to reuse empty space
2021-03-23 08:48:20 +00:00
// Calculate common Y scale for graphs
for (int row = 0; row < row_count; row++) {
QTreeWidgetItem *ti = ui->streamTreeWidget->topLevelItem(row);
RtpAudioStream *audio_stream = ti->data(stream_data_col_, Qt::UserRole).value<RtpAudioStream*>();
gint16 max_sample_value = audio_stream->getMaxSampleValue();
if (max_sample_value > total_max_sample_value) {
total_max_sample_value = max_sample_value;
}
}
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
// Clear existing graphs
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
for (int row = 0; row < row_count; row++) {
QTreeWidgetItem *ti = ui->streamTreeWidget->topLevelItem(row);
RtpAudioStream *audio_stream = ti->data(stream_data_col_, Qt::UserRole).value<RtpAudioStream*>();
int y_offset = row_count - row - 1;
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
AudioRouting audio_routing = audio_stream->getAudioRouting();
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
ti->setData(graph_audio_data_col_, Qt::UserRole, QVariant());
ti->setData(graph_sequence_data_col_, Qt::UserRole, QVariant());
ti->setData(graph_jitter_data_col_, Qt::UserRole, QVariant());
ti->setData(graph_timestamp_data_col_, Qt::UserRole, QVariant());
ti->setData(graph_silence_data_col_, Qt::UserRole, QVariant());
Split RTP player tapping, decoding, and plotting. In RtpAudioStream split tapping+decoding into separate member functions. Store RTP payloads in memory. In RtpPlayerDialog split tapping+plotting. This more closely resembles what we're doing in the GTK+ UI and paves the way for jitter support and other changes. Change-Id: I244c225cec8930545622e6582b7be35ebe45b237 Reviewed-on: https://code.wireshark.org/review/11195 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2015-10-20 15:31:52 +00:00
RTP Player: UI improvements Changes: - all waveforms has common scale therefore louder/quiter signal is visible - when stream/streams are deleted from view, Y axis is rescaled and waveforms are rearranged to reuse empty space
2021-03-23 08:48:20 +00:00
// Set common scale
audio_stream->setMaxSampleValue(total_max_sample_value);
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
// Waveform
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
RtpAudioGraph *audio_graph = new RtpAudioGraph(ui->audioPlot, audio_stream->color());
audio_graph->setMuted(audio_routing.isMuted());
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
audio_graph->setData(audio_stream->visualTimestamps(relative_timestamps), audio_stream->visualSamples(y_offset));
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
ti->setData(graph_audio_data_col_, Qt::UserRole, QVariant::fromValue<RtpAudioGraph *>(audio_graph));
//RTP_STREAM_DEBUG("Plotting %s, %d samples", ti->text(src_addr_col_).toUtf8().constData(), audio_graph->wave->data()->size());
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
RTP Player: Time span column shows time of day when Time of Day enabled Time span column shows relative start and end time of stream. When Time of Day enabled, column now shows absolute start and end time.
2021-01-01 18:22:07 +00:00
QString span_str;
if (ui->todCheckBox->isChecked()) {
QDateTime date_time1 = QDateTime::fromMSecsSinceEpoch((audio_stream->startRelTime() + first_stream_abs_start_time_ - audio_stream->startRelTime()) * 1000.0);
QDateTime date_time2 = QDateTime::fromMSecsSinceEpoch((audio_stream->stopRelTime() + first_stream_abs_start_time_ - audio_stream->startRelTime()) * 1000.0);
QString time_str1 = date_time1.toString("yyyy-MM-dd hh:mm:ss.zzz");
QString time_str2 = date_time2.toString("yyyy-MM-dd hh:mm:ss.zzz");
span_str = QString("%1 - %2 (%3)")
.arg(time_str1)
.arg(time_str2)
.arg(QString::number(audio_stream->stopRelTime() - audio_stream->startRelTime(), 'f', prefs.gui_decimal_places1));
} else {
span_str = QString("%1 - %2 (%3)")
All RTP dialogs: Added option to specify count of decimal places Preferences are extended with advanced settings: gui.decimal_places1 - 2 gui.decimal_places2 - 4 gui.decimal_places3 - 6 rtp_analysis_dialog, rtp_player_dialog and rtp_stream_dialog uses new settings for formating numbers. Same information in all dialogs uses same settings. It solves request #15481. Note: Other UI dialogs can be adapted later.
2020-12-30 12:28:31 +00:00
.arg(QString::number(audio_stream->startRelTime(), 'f', prefs.gui_decimal_places1))
.arg(QString::number(audio_stream->stopRelTime(), 'f', prefs.gui_decimal_places1))
.arg(QString::number(audio_stream->stopRelTime() - audio_stream->startRelTime(), 'f', prefs.gui_decimal_places1));
RTP Player: Time span column shows time of day when Time of Day enabled Time span column shows relative start and end time of stream. When Time of Day enabled, column now shows absolute start and end time.
2021-01-01 18:22:07 +00:00
}
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
ti->setText(time_span_col_, span_str);
ti->setText(sample_rate_col_, QString::number(audio_stream->sampleRate()));
RTP Player: Added play rate of decoded audio It shows which rate wireshark really decoded the stream and how it will be played.
2021-03-29 06:49:51 +00:00
ti->setText(play_rate_col_, QString::number(audio_stream->playRate()));
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
ti->setText(payload_col_, audio_stream->payloadNames().join(", "));
if (audio_stream->outOfSequence() > 0) {
// Sequence numbers
QCPGraph *seq_graph = ui->audioPlot->addGraph();
seq_graph->setLineStyle(QCPGraph::lsNone);
RTP player: Size of symbols increased Size of symbols (Wrong Timestamp, ...) in graph increased from 4pt to size of main UI font.
2020-12-29 14:03:59 +00:00
seq_graph->setScatterStyle(QCPScatterStyle(QCPScatterStyle::ssSquare, tango_aluminium_6, Qt::white, wsApp->font().pointSize())); // Arbitrary
Update to qcustomplot 2.0.1 Upgrade the internal version of QCustomPlot to 2.0.1 Change-Id: I1eb372d8e6a2f6c1bbdde4c74596785bf2d405c2 Reviewed-on: https://code.wireshark.org/review/17980 Reviewed-by: Roland Knall <rknall@gmail.com> Petri-Dish: Roland Knall <rknall@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2016-09-25 09:57:55 +00:00
seq_graph->setSelectable(QCP::stNone);
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
seq_graph->setData(audio_stream->outOfSequenceTimestamps(relative_timestamps), audio_stream->outOfSequenceSamples(y_offset));
RTP dissector: Remove unused functions Patch removes (just comments out) functions not used by the code as found by Martin Mathieson.
2021-02-05 09:19:48 +00:00
ti->setData(graph_sequence_data_col_, Qt::UserRole, QVariant::fromValue<QCPGraph *>(seq_graph));
RTP Player: Fixed incorrect handling of legend in graph Legend and it's elements were shown only when first waveform requires it. Patch fixes it and all legend elements are shown when required by any waveform.
2021-03-22 10:25:41 +00:00
if (legend_out_of_sequence) {
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
seq_graph->removeFromLegend();
RTP Player: Fixed incorrect handling of legend in graph Legend and it's elements were shown only when first waveform requires it. Patch fixes it and all legend elements are shown when required by any waveform.
2021-03-22 10:25:41 +00:00
} else {
seq_graph->setName(tr("Out of Sequence"));
legend_out_of_sequence = true;
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
}
}
Add jitter logic to RtpAudioStream. Copy the jitter logic from rtp_player.c to rtp_audio_stream.cpp. This still isn't correct but the RTP player should now be complete enough to start looking at the bug list at the top of rtp_player_dialog.cpp. Disable timing and jitter controls while we're playing while we're here. Fixes bug 11635. Bug: 11635 Change-Id: Ie583ade522702cbe1bbcea4475a535caa1d74fa2 Reviewed-on: https://code.wireshark.org/review/11295 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2015-10-26 15:34:58 +00:00
if (audio_stream->jitterDropped() > 0) {
// Jitter drops
QCPGraph *seq_graph = ui->audioPlot->addGraph();
seq_graph->setLineStyle(QCPGraph::lsNone);
RTP player: Size of symbols increased Size of symbols (Wrong Timestamp, ...) in graph increased from 4pt to size of main UI font.
2020-12-29 14:03:59 +00:00
seq_graph->setScatterStyle(QCPScatterStyle(QCPScatterStyle::ssCircle, tango_scarlet_red_5, Qt::white, wsApp->font().pointSize())); // Arbitrary
Update to qcustomplot 2.0.1 Upgrade the internal version of QCustomPlot to 2.0.1 Change-Id: I1eb372d8e6a2f6c1bbdde4c74596785bf2d405c2 Reviewed-on: https://code.wireshark.org/review/17980 Reviewed-by: Roland Knall <rknall@gmail.com> Petri-Dish: Roland Knall <rknall@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2016-09-25 09:57:55 +00:00
seq_graph->setSelectable(QCP::stNone);
Add jitter logic to RtpAudioStream. Copy the jitter logic from rtp_player.c to rtp_audio_stream.cpp. This still isn't correct but the RTP player should now be complete enough to start looking at the bug list at the top of rtp_player_dialog.cpp. Disable timing and jitter controls while we're playing while we're here. Fixes bug 11635. Bug: 11635 Change-Id: Ie583ade522702cbe1bbcea4475a535caa1d74fa2 Reviewed-on: https://code.wireshark.org/review/11295 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2015-10-26 15:34:58 +00:00
seq_graph->setData(audio_stream->jitterDroppedTimestamps(relative_timestamps), audio_stream->jitterDroppedSamples(y_offset));
RTP dissector: Remove unused functions Patch removes (just comments out) functions not used by the code as found by Martin Mathieson.
2021-02-05 09:19:48 +00:00
ti->setData(graph_jitter_data_col_, Qt::UserRole, QVariant::fromValue<QCPGraph *>(seq_graph));
RTP Player: Fixed incorrect handling of legend in graph Legend and it's elements were shown only when first waveform requires it. Patch fixes it and all legend elements are shown when required by any waveform.
2021-03-22 10:25:41 +00:00
if (legend_jitter_dropped) {
Add jitter logic to RtpAudioStream. Copy the jitter logic from rtp_player.c to rtp_audio_stream.cpp. This still isn't correct but the RTP player should now be complete enough to start looking at the bug list at the top of rtp_player_dialog.cpp. Disable timing and jitter controls while we're playing while we're here. Fixes bug 11635. Bug: 11635 Change-Id: Ie583ade522702cbe1bbcea4475a535caa1d74fa2 Reviewed-on: https://code.wireshark.org/review/11295 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2015-10-26 15:34:58 +00:00
seq_graph->removeFromLegend();
RTP Player: Fixed incorrect handling of legend in graph Legend and it's elements were shown only when first waveform requires it. Patch fixes it and all legend elements are shown when required by any waveform.
2021-03-22 10:25:41 +00:00
} else {
seq_graph->setName(tr("Jitter Drops"));
legend_jitter_dropped = true;
Add jitter logic to RtpAudioStream. Copy the jitter logic from rtp_player.c to rtp_audio_stream.cpp. This still isn't correct but the RTP player should now be complete enough to start looking at the bug list at the top of rtp_player_dialog.cpp. Disable timing and jitter controls while we're playing while we're here. Fixes bug 11635. Bug: 11635 Change-Id: Ie583ade522702cbe1bbcea4475a535caa1d74fa2 Reviewed-on: https://code.wireshark.org/review/11295 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2015-10-26 15:34:58 +00:00
}
}
if (audio_stream->wrongTimestamps() > 0) {
// Wrong timestamps
QCPGraph *seq_graph = ui->audioPlot->addGraph();
seq_graph->setLineStyle(QCPGraph::lsNone);
RTP player: Size of symbols increased Size of symbols (Wrong Timestamp, ...) in graph increased from 4pt to size of main UI font.
2020-12-29 14:03:59 +00:00
seq_graph->setScatterStyle(QCPScatterStyle(QCPScatterStyle::ssDiamond, tango_sky_blue_5, Qt::white, wsApp->font().pointSize())); // Arbitrary
Update to qcustomplot 2.0.1 Upgrade the internal version of QCustomPlot to 2.0.1 Change-Id: I1eb372d8e6a2f6c1bbdde4c74596785bf2d405c2 Reviewed-on: https://code.wireshark.org/review/17980 Reviewed-by: Roland Knall <rknall@gmail.com> Petri-Dish: Roland Knall <rknall@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2016-09-25 09:57:55 +00:00
seq_graph->setSelectable(QCP::stNone);
Add jitter logic to RtpAudioStream. Copy the jitter logic from rtp_player.c to rtp_audio_stream.cpp. This still isn't correct but the RTP player should now be complete enough to start looking at the bug list at the top of rtp_player_dialog.cpp. Disable timing and jitter controls while we're playing while we're here. Fixes bug 11635. Bug: 11635 Change-Id: Ie583ade522702cbe1bbcea4475a535caa1d74fa2 Reviewed-on: https://code.wireshark.org/review/11295 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2015-10-26 15:34:58 +00:00
seq_graph->setData(audio_stream->wrongTimestampTimestamps(relative_timestamps), audio_stream->wrongTimestampSamples(y_offset));
RTP dissector: Remove unused functions Patch removes (just comments out) functions not used by the code as found by Martin Mathieson.
2021-02-05 09:19:48 +00:00
ti->setData(graph_timestamp_data_col_, Qt::UserRole, QVariant::fromValue<QCPGraph *>(seq_graph));
RTP Player: Fixed incorrect handling of legend in graph Legend and it's elements were shown only when first waveform requires it. Patch fixes it and all legend elements are shown when required by any waveform.
2021-03-22 10:25:41 +00:00
if (legend_wrong_timestamps) {
Add jitter logic to RtpAudioStream. Copy the jitter logic from rtp_player.c to rtp_audio_stream.cpp. This still isn't correct but the RTP player should now be complete enough to start looking at the bug list at the top of rtp_player_dialog.cpp. Disable timing and jitter controls while we're playing while we're here. Fixes bug 11635. Bug: 11635 Change-Id: Ie583ade522702cbe1bbcea4475a535caa1d74fa2 Reviewed-on: https://code.wireshark.org/review/11295 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2015-10-26 15:34:58 +00:00
seq_graph->removeFromLegend();
RTP Player: Fixed incorrect handling of legend in graph Legend and it's elements were shown only when first waveform requires it. Patch fixes it and all legend elements are shown when required by any waveform.
2021-03-22 10:25:41 +00:00
} else {
seq_graph->setName(tr("Wrong Timestamps"));
legend_wrong_timestamps = true;
Add jitter logic to RtpAudioStream. Copy the jitter logic from rtp_player.c to rtp_audio_stream.cpp. This still isn't correct but the RTP player should now be complete enough to start looking at the bug list at the top of rtp_player_dialog.cpp. Disable timing and jitter controls while we're playing while we're here. Fixes bug 11635. Bug: 11635 Change-Id: Ie583ade522702cbe1bbcea4475a535caa1d74fa2 Reviewed-on: https://code.wireshark.org/review/11295 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2015-10-26 15:34:58 +00:00
}
}
if (audio_stream->insertedSilences() > 0) {
// Inserted silence
QCPGraph *seq_graph = ui->audioPlot->addGraph();
seq_graph->setLineStyle(QCPGraph::lsNone);
RTP player: Size of symbols increased Size of symbols (Wrong Timestamp, ...) in graph increased from 4pt to size of main UI font.
2020-12-29 14:03:59 +00:00
seq_graph->setScatterStyle(QCPScatterStyle(QCPScatterStyle::ssTriangle, tango_butter_5, Qt::white, wsApp->font().pointSize())); // Arbitrary
Update to qcustomplot 2.0.1 Upgrade the internal version of QCustomPlot to 2.0.1 Change-Id: I1eb372d8e6a2f6c1bbdde4c74596785bf2d405c2 Reviewed-on: https://code.wireshark.org/review/17980 Reviewed-by: Roland Knall <rknall@gmail.com> Petri-Dish: Roland Knall <rknall@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2016-09-25 09:57:55 +00:00
seq_graph->setSelectable(QCP::stNone);
Add jitter logic to RtpAudioStream. Copy the jitter logic from rtp_player.c to rtp_audio_stream.cpp. This still isn't correct but the RTP player should now be complete enough to start looking at the bug list at the top of rtp_player_dialog.cpp. Disable timing and jitter controls while we're playing while we're here. Fixes bug 11635. Bug: 11635 Change-Id: Ie583ade522702cbe1bbcea4475a535caa1d74fa2 Reviewed-on: https://code.wireshark.org/review/11295 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2015-10-26 15:34:58 +00:00
seq_graph->setData(audio_stream->insertedSilenceTimestamps(relative_timestamps), audio_stream->insertedSilenceSamples(y_offset));
RTP dissector: Remove unused functions Patch removes (just comments out) functions not used by the code as found by Martin Mathieson.
2021-02-05 09:19:48 +00:00
ti->setData(graph_silence_data_col_, Qt::UserRole, QVariant::fromValue<QCPGraph *>(seq_graph));
RTP Player: Fixed incorrect handling of legend in graph Legend and it's elements were shown only when first waveform requires it. Patch fixes it and all legend elements are shown when required by any waveform.
2021-03-22 10:25:41 +00:00
if (legend_inserted_silences) {
Add jitter logic to RtpAudioStream. Copy the jitter logic from rtp_player.c to rtp_audio_stream.cpp. This still isn't correct but the RTP player should now be complete enough to start looking at the bug list at the top of rtp_player_dialog.cpp. Disable timing and jitter controls while we're playing while we're here. Fixes bug 11635. Bug: 11635 Change-Id: Ie583ade522702cbe1bbcea4475a535caa1d74fa2 Reviewed-on: https://code.wireshark.org/review/11295 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2015-10-26 15:34:58 +00:00
seq_graph->removeFromLegend();
RTP Player: Fixed incorrect handling of legend in graph Legend and it's elements were shown only when first waveform requires it. Patch fixes it and all legend elements are shown when required by any waveform.
2021-03-22 10:25:41 +00:00
} else {
seq_graph->setName(tr("Inserted Silence"));
legend_inserted_silences = true;
Add jitter logic to RtpAudioStream. Copy the jitter logic from rtp_player.c to rtp_audio_stream.cpp. This still isn't correct but the RTP player should now be complete enough to start looking at the bug list at the top of rtp_player_dialog.cpp. Disable timing and jitter controls while we're playing while we're here. Fixes bug 11635. Bug: 11635 Change-Id: Ie583ade522702cbe1bbcea4475a535caa1d74fa2 Reviewed-on: https://code.wireshark.org/review/11295 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2015-10-26 15:34:58 +00:00
}
}
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
}
RTP Player: Fixed incorrect handling of legend in graph Legend and it's elements were shown only when first waveform requires it. Patch fixes it and all legend elements are shown when required by any waveform.
2021-03-22 10:25:41 +00:00
ui->audioPlot->legend->setVisible(legend_out_of_sequence || legend_jitter_dropped || legend_wrong_timestamps || legend_inserted_silences);
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
ui->audioPlot->replot();
if (rescale_axes) resetXAxis();
}
RTP Player: Dialog is nonmodal now and can be called multiple ways Changes: - refactored main_dialog handling of telephony dialogs - RTP Player dialog is nonmodal now and can be left open - it is possible to issue three actions on RTP Player dialog from other dialogs (other dialog have selected set of RTP streams before action) - replace - removes existing streams from RTP dialog and shows new set - add - adds new set to existing list in RTP dialog - remove - remove streams in set from list in RTP dialog - Sequence Dialog: - was modified to hold rtpstream_info_t for RTP streams - added Play button - VoIP features (RTP Play button, select/deselect RTP stream) are disabled after creation and must be enabled. It handles that RTP Play button is not shown e.g. in TCP sequence show
2021-03-27 20:01:43 +00:00
void RtpPlayerDialog::addSingleRtpStream(rtpstream_info_t *rtpstream)
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
{
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
AudioRouting audio_routing = AudioRouting(AUDIO_UNMUTED, channel_mono);
rtp_player_dialog: Route audio for a stream to left/right speaker in RTP player Column 'Play' added to player. Double click on a stream in the column changes audio routing for the stream. When soundcard supports only one channel, there are Mute/Play option. When soundcard supports two or more channels, there are Mute/L/L+R/R options. Muted channel is drawn with dotted line. Change-Id: If120c902195da46f98a1663c589f20c6a1da0ba7 Reviewed-on: https://code.wireshark.org/review/35687 Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-07 21:16:42 +00:00
RTP: Common functions for allocation/deallocation of rtpstream_info_t Change-Id: I9a0a11d238473a7c57d85547dca0713ed421a500 Reviewed-on: https://code.wireshark.org/review/28417 Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2018-06-21 13:16:03 +00:00
if (!rtpstream) return;
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
// Find the RTP streams associated with this conversation.
// gtk/rtp_player.c:mark_rtp_stream_to_play does this differently.
RtpAudioStream *audio_stream = NULL;
int tli_count = ui->streamTreeWidget->topLevelItemCount();
for (int row = 0; row < tli_count; row++) {
QTreeWidgetItem *ti = ui->streamTreeWidget->topLevelItem(row);
RtpAudioStream *row_stream = ti->data(stream_data_col_, Qt::UserRole).value<RtpAudioStream*>();
RTP: Common functions for allocation/deallocation of rtpstream_info_t Change-Id: I9a0a11d238473a7c57d85547dca0713ed421a500 Reviewed-on: https://code.wireshark.org/review/28417 Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2018-06-21 13:16:03 +00:00
if (row_stream->isMatch(rtpstream)) {
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
audio_stream = row_stream;
break;
}
}
if (!audio_stream) {
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
audio_stream = new RtpAudioStream(this, rtpstream, stereo_available_);
Qt: Don't expose ColorUtils::graph_colors_. Make graph_colors_ private and accessible via getters. Blind attempt at fixing bug 11833. Bug: 11833 Change-Id: I03b7e90c686374d2d0f046f7e5fe87e43939dc82 Reviewed-on: https://code.wireshark.org/review/12318 Reviewed-by: Gerald Combs <gerald@wireshark.org> Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-11-30 22:42:42 +00:00
audio_stream->setColor(ColorUtils::graphColor(tli_count));
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
QTreeWidgetItem *ti = new RtpPlayerTreeWidgetItem(ui->streamTreeWidget);
RTP: Common functions for allocation/deallocation of rtpstream_info_t Change-Id: I9a0a11d238473a7c57d85547dca0713ed421a500 Reviewed-on: https://code.wireshark.org/review/28417 Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2018-06-21 13:16:03 +00:00
ti->setText(src_addr_col_, address_to_qstring(&rtpstream->id.src_addr));
ti->setText(src_port_col_, QString::number(rtpstream->id.src_port));
ti->setText(dst_addr_col_, address_to_qstring(&rtpstream->id.dst_addr));
ti->setText(dst_port_col_, QString::number(rtpstream->id.dst_port));
ti->setText(ssrc_col_, int_to_qstring(rtpstream->id.ssrc, 8, 16));
RTP Player: Showing of setup frame for RTP stream fixed When RTP stream has no setup frame, but is decoded by Decode as or with rtp_udp active, setup frame was shown as SETUP <number>, but correct is RTP <number>.
2021-03-26 23:39:30 +00:00
// 0xFFFFFFFF mean no setup frame
// first_packet_num == setup_frame_number happens, when
// rtp_udp is active or Decode as was used
if ((rtpstream->setup_frame_number == 0xFFFFFFFF) ||
(rtpstream->rtp_stats.first_packet_num == rtpstream->setup_frame_number)
) {
RTP player: It is possible to select SETUP packet related to RTP stream When SETUP (or related signalling packet) exists, it is selected by Shift+G. If RTP stream is "alone", first RTP packet of the stream is selected.
2021-03-09 13:47:07 +00:00
int packet = rtpstream->rtp_stats.first_packet_num;
ti->setText(first_pkt_col_, QString("RTP %1").arg(packet));
ti->setData(first_pkt_col_, Qt::UserRole, QVariant(packet));
} else {
int packet = rtpstream->setup_frame_number;
ti->setText(first_pkt_col_, QString("SETUP %1").arg(rtpstream->setup_frame_number));
ti->setData(first_pkt_col_, Qt::UserRole, QVariant(packet));
}
RTP: Common functions for allocation/deallocation of rtpstream_info_t Change-Id: I9a0a11d238473a7c57d85547dca0713ed421a500 Reviewed-on: https://code.wireshark.org/review/28417 Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2018-06-21 13:16:03 +00:00
ti->setText(num_pkts_col_, QString::number(rtpstream->packet_count));
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
ti->setData(stream_data_col_, Qt::UserRole, QVariant::fromValue(audio_stream));
rtp_player_dialog: Route audio for a stream to left/right speaker in RTP player Column 'Play' added to player. Double click on a stream in the column changes audio routing for the stream. When soundcard supports only one channel, there are Mute/Play option. When soundcard supports two or more channels, there are Mute/L/L+R/R options. Muted channel is drawn with dotted line. Change-Id: If120c902195da46f98a1663c589f20c6a1da0ba7 Reviewed-on: https://code.wireshark.org/review/35687 Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-07 21:16:42 +00:00
if (stereo_available_) {
if (tli_count%2) {
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
audio_routing.setChannel(channel_stereo_right);
rtp_player_dialog: Route audio for a stream to left/right speaker in RTP player Column 'Play' added to player. Double click on a stream in the column changes audio routing for the stream. When soundcard supports only one channel, there are Mute/Play option. When soundcard supports two or more channels, there are Mute/L/L+R/R options. Muted channel is drawn with dotted line. Change-Id: If120c902195da46f98a1663c589f20c6a1da0ba7 Reviewed-on: https://code.wireshark.org/review/35687 Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-07 21:16:42 +00:00
} else {
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
audio_routing.setChannel(channel_stereo_left);
rtp_player_dialog: Route audio for a stream to left/right speaker in RTP player Column 'Play' added to player. Double click on a stream in the column changes audio routing for the stream. When soundcard supports only one channel, there are Mute/Play option. When soundcard supports two or more channels, there are Mute/L/L+R/R options. Muted channel is drawn with dotted line. Change-Id: If120c902195da46f98a1663c589f20c6a1da0ba7 Reviewed-on: https://code.wireshark.org/review/35687 Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-07 21:16:42 +00:00
}
} else {
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
audio_routing.setChannel(channel_mono);
rtp_player_dialog: Route audio for a stream to left/right speaker in RTP player Column 'Play' added to player. Double click on a stream in the column changes audio routing for the stream. When soundcard supports only one channel, there are Mute/Play option. When soundcard supports two or more channels, there are Mute/L/L+R/R options. Muted channel is drawn with dotted line. Change-Id: If120c902195da46f98a1663c589f20c6a1da0ba7 Reviewed-on: https://code.wireshark.org/review/35687 Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-07 21:16:42 +00:00
}
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
ti->setToolTip(channel_col_, QString(tr("Double click to change audio routing")));
formatAudioRouting(ti, audio_routing);
audio_stream->setAudioRouting(audio_routing);
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
for (int col = 0; col < ui->streamTreeWidget->columnCount(); col++) {
Qt: Fix 5.13 deprecated warnings With Qt 5.13 some methods are deprecated. Fixing those warnings Change-Id: Ia290f06f2b681de1d5b437624de77d8a5c2f5266 Reviewed-on: https://code.wireshark.org/review/33761 Petri-Dish: Roland Knall <rknall@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Roland Knall <rknall@gmail.com>
2019-06-28 10:37:02 +00:00
QBrush fgBrush = ti->foreground(col);
fgBrush.setColor(audio_stream->color());
ti->setForeground(col, fgBrush);
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
}
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
connect(audio_stream, SIGNAL(finishedPlaying(RtpAudioStream *)), this, SLOT(playFinished(RtpAudioStream *)));
Qt: Make the RTP player output device selectable. Add a combobox for selecting the output device and populate it with our available devices. Let the user know if our output format isn't supported. Ping-Bug: 13105 Change-Id: I299c7d0f191bb66d93896338036000e2c377781f Reviewed-on: https://code.wireshark.org/review/19046 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2016-12-02 23:52:02 +00:00
connect(audio_stream, SIGNAL(playbackError(QString)), this, SLOT(setPlaybackError(QString)));
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
}
rtp_player: Player is able to set start of audio play by double click Patch adds ability to set start of audio play by double clicking on waveform. Patch fixes unreported issue with placing waveform at incorrect place when switched relative/absolute time mode (check/uncheck Time of Day). Change-Id: Ib8ce24aea870e2443e033afbb6d6e9fbcf222431 Reviewed-on: https://code.wireshark.org/review/35621 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-01 23:38:35 +00:00
// Update start/stop time nevertheless stream is new or already seen
// because voip_calls_dialog.cpp splits same stream to multiple pieces
updateStartStopTime(rtpstream, tli_count);
RTP: Code cleanup 2 Removed unused items. Change-Id: Ic83d7bfae12424f11e3fc2a9a678c28a0ff6b72e Reviewed-on: https://code.wireshark.org/review/28345 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2018-06-20 12:06:37 +00:00
RTP_STREAM_DEBUG("adding stream %d to layout, %u packets, start %u",
Split RTP player tapping, decoding, and plotting. In RtpAudioStream split tapping+decoding into separate member functions. Store RTP payloads in memory. In RtpPlayerDialog split tapping+plotting. This more closely resembles what we're doing in the GTK+ UI and paves the way for jitter support and other changes. Change-Id: I244c225cec8930545622e6582b7be35ebe45b237 Reviewed-on: https://code.wireshark.org/review/11195 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2015-10-20 15:31:52 +00:00
ui->streamTreeWidget->topLevelItemCount(),
RTP: Common functions for allocation/deallocation of rtpstream_info_t Change-Id: I9a0a11d238473a7c57d85547dca0713ed421a500 Reviewed-on: https://code.wireshark.org/review/28417 Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2018-06-21 13:16:03 +00:00
rtpstream->packet_count,
rtpstream->start_fd ? rtpstream->start_fd->num : 0);
RTP Player: Dialog is nonmodal now and can be called multiple ways Changes: - refactored main_dialog handling of telephony dialogs - RTP Player dialog is nonmodal now and can be left open - it is possible to issue three actions on RTP Player dialog from other dialogs (other dialog have selected set of RTP streams before action) - replace - removes existing streams from RTP dialog and shows new set - add - adds new set to existing list in RTP dialog - remove - remove streams in set from list in RTP dialog - Sequence Dialog: - was modified to hold rtpstream_info_t for RTP streams - added Play button - VoIP features (RTP Play button, select/deselect RTP stream) are disabled after creation and must be enabled. It handles that RTP Play button is not shown e.g. in TCP sequence show
2021-03-27 20:01:43 +00:00
}
void RtpPlayerDialog::replaceRtpStreams(QVector<rtpstream_info_t *> stream_infos)
{
// Delete all existing rows
on_actionSelectAll_triggered();
on_actionRemoveStream_triggered();
// Add all new streams
for (int i=0; i < stream_infos.size(); i++) {
addSingleRtpStream(stream_infos[i]);
}
setMarkers();
QTimer::singleShot(0, this, SLOT(retapPackets()));
}
void RtpPlayerDialog::addRtpStreams(QVector<rtpstream_info_t *> stream_infos)
{
int tli_count = ui->streamTreeWidget->topLevelItemCount();
// Add new streams
for (int i=0; i < stream_infos.size(); i++) {
addSingleRtpStream(stream_infos[i]);
}
if (tli_count == 0) {
setMarkers();
}
QTimer::singleShot(0, this, SLOT(retapPackets()));
}
void RtpPlayerDialog::removeRtpStreams(QVector<rtpstream_info_t *> stream_infos)
{
int tli_count = ui->streamTreeWidget->topLevelItemCount();
if (last_ti_) {
highlightItem(last_ti_, false);
last_ti_ = NULL;
}
for (int i=0; i < stream_infos.size(); i++) {
for (int row = 0; row < tli_count; row++) {
QTreeWidgetItem *ti = ui->streamTreeWidget->topLevelItem(row);
RtpAudioStream *row_stream = ti->data(stream_data_col_, Qt::UserRole).value<RtpAudioStream*>();
if (row_stream->isMatch(stream_infos[i])) {
removeRow(ti);
tli_count--;
break;
}
}
}
updateGraphs();
updateWidgets();
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
}
rtp_player: Player is able to set start of audio play by double click Patch adds ability to set start of audio play by double clicking on waveform. Patch fixes unreported issue with placing waveform at incorrect place when switched relative/absolute time mode (check/uncheck Time of Day). Change-Id: Ib8ce24aea870e2443e033afbb6d6e9fbcf222431 Reviewed-on: https://code.wireshark.org/review/35621 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-01 23:38:35 +00:00
void RtpPlayerDialog::setMarkers()
{
setStartPlayMarker(0);
drawStartPlayMarker();
}
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
void RtpPlayerDialog::showEvent(QShowEvent *)
{
QList<int> split_sizes = ui->splitter->sizes();
int tot_size = split_sizes[0] + split_sizes[1];
int plot_size = tot_size * 3 / 4;
split_sizes.clear();
split_sizes << plot_size << tot_size - plot_size;
ui->splitter->setSizes(split_sizes);
}
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
bool RtpPlayerDialog::eventFilter(QObject *, QEvent *event)
{
if (event->type() == QEvent::KeyPress) {
QKeyEvent &keyEvent = static_cast<QKeyEvent&>(*event);
int pan_secs = keyEvent.modifiers() & Qt::ShiftModifier ? 1 : 10;
switch(keyEvent.key()) {
case Qt::Key_Minus:
case Qt::Key_Underscore: // Shifted minus on U.S. keyboards
case Qt::Key_O: // GTK+
case Qt::Key_R:
on_actionZoomOut_triggered();
return true;
case Qt::Key_Plus:
case Qt::Key_Equal: // Unshifted plus on U.S. keyboards
case Qt::Key_I: // GTK+
if (keyEvent.modifiers() == Qt::ControlModifier) {
RTP player: It is possible to select SETUP packet related to RTP stream When SETUP (or related signalling packet) exists, it is selected by Shift+G. If RTP stream is "alone", first RTP packet of the stream is selected.
2021-03-09 13:47:07 +00:00
// Ctrl+I
Voice dialogs: Added Select All/None/Invert to all of it
2021-03-13 20:28:06 +00:00
on_actionSelectInvert_triggered();
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
return true;
} else {
RTP player: It is possible to select SETUP packet related to RTP stream When SETUP (or related signalling packet) exists, it is selected by Shift+G. If RTP stream is "alone", first RTP packet of the stream is selected.
2021-03-09 13:47:07 +00:00
// I
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
on_actionZoomIn_triggered();
return true;
}
break;
case Qt::Key_Right:
case Qt::Key_L:
panXAxis(pan_secs);
return true;
case Qt::Key_Left:
case Qt::Key_H:
panXAxis(-1 * pan_secs);
return true;
case Qt::Key_0:
case Qt::Key_ParenRight: // Shifted 0 on U.S. keyboards
on_actionReset_triggered();
return true;
case Qt::Key_G:
RTP player: It is possible to select SETUP packet related to RTP stream When SETUP (or related signalling packet) exists, it is selected by Shift+G. If RTP stream is "alone", first RTP packet of the stream is selected.
2021-03-09 13:47:07 +00:00
if (keyEvent.modifiers() == Qt::ShiftModifier) {
// Goto SETUP frame, use correct call based on caller
QPoint pos1 = ui->audioPlot->mapFromGlobal(QCursor::pos());
QPoint pos2 = ui->streamTreeWidget->mapFromGlobal(QCursor::pos());
if (ui->audioPlot->rect().contains(pos1)) {
// audio plot, by mouse coords
on_actionGoToSetupPacketPlot_triggered();
} else if (ui->streamTreeWidget->rect().contains(pos2)) {
// packet tree, by cursor
on_actionGoToSetupPacketTree_triggered();
}
return true;
} else {
on_actionGoToPacket_triggered();
return true;
}
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
case Qt::Key_A:
if (keyEvent.modifiers() == Qt::ControlModifier) {
RTP player: It is possible to select SETUP packet related to RTP stream When SETUP (or related signalling packet) exists, it is selected by Shift+G. If RTP stream is "alone", first RTP packet of the stream is selected.
2021-03-09 13:47:07 +00:00
// Ctrl+A
Voice dialogs: Added Select All/None/Invert to all of it
2021-03-13 20:28:06 +00:00
on_actionSelectAll_triggered();
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
return true;
} else if (keyEvent.modifiers() == (Qt::ShiftModifier | Qt::ControlModifier)) {
RTP player: It is possible to select SETUP packet related to RTP stream When SETUP (or related signalling packet) exists, it is selected by Shift+G. If RTP stream is "alone", first RTP packet of the stream is selected.
2021-03-09 13:47:07 +00:00
// Ctrl+Shift+A
Voice dialogs: Added Select All/None/Invert to all of it
2021-03-13 20:28:06 +00:00
on_actionSelectNone_triggered();
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
return true;
}
break;
case Qt::Key_M:
RTP Player: Added additional shortcuts Added shortcuts: - Mute/Unmute/Invert Mute - Play/Pause/Stop Removed outdated help in tooltip text. It makes no sense to use it, all actions are available in context menu.
2021-03-11 08:22:15 +00:00
if (keyEvent.modifiers() == Qt::ShiftModifier) {
on_actionAudioRoutingUnmute_triggered();
return true;
} else if (keyEvent.modifiers() == Qt::ControlModifier) {
on_actionAudioRoutingMuteInvert_triggered();
return true;
} else {
on_actionAudioRoutingMute_triggered();
return true;
}
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
case Qt::Key_Delete:
on_actionRemoveStream_triggered();
return true;
case Qt::Key_X:
if (keyEvent.modifiers() == Qt::ControlModifier) {
RTP player: It is possible to select SETUP packet related to RTP stream When SETUP (or related signalling packet) exists, it is selected by Shift+G. If RTP stream is "alone", first RTP packet of the stream is selected.
2021-03-09 13:47:07 +00:00
// Ctrl+X
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
on_actionRemoveStream_triggered();
return true;
}
break;
case Qt::Key_Down:
case Qt::Key_Up:
case Qt::Key_PageUp:
case Qt::Key_PageDown:
case Qt::Key_Home:
case Qt::Key_End:
RTP player: It is possible to select SETUP packet related to RTP stream When SETUP (or related signalling packet) exists, it is selected by Shift+G. If RTP stream is "alone", first RTP packet of the stream is selected.
2021-03-09 13:47:07 +00:00
// Route keys to QTreeWidget
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
ui->streamTreeWidget->setFocus();
break;
RTP Player: Added additional shortcuts Added shortcuts: - Mute/Unmute/Invert Mute - Play/Pause/Stop Removed outdated help in tooltip text. It makes no sense to use it, all actions are available in context menu.
2021-03-11 08:22:15 +00:00
case Qt::Key_P:
on_actionPlay_triggered();
return true;
case Qt::Key_S:
on_actionStop_triggered();
return true;
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
}
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
}
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
return false;
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
}
RTP dissector: Remove unused functions Patch removes (just comments out) functions not used by the code as found by Martin Mathieson.
2021-02-05 09:19:48 +00:00
void RtpPlayerDialog::contextMenuEvent(QContextMenuEvent *event)
{
list_ctx_menu_->exec(event->globalPos());
}
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
void RtpPlayerDialog::updateWidgets()
{
bool enable_play = true;
RTP player: Added pause button Pause button added. Button is visible only when media are played.
2020-12-29 13:41:14 +00:00
bool enable_pause = false;
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
bool enable_stop = false;
Add jitter logic to RtpAudioStream. Copy the jitter logic from rtp_player.c to rtp_audio_stream.cpp. This still isn't correct but the RTP player should now be complete enough to start looking at the bug list at the top of rtp_player_dialog.cpp. Disable timing and jitter controls while we're playing while we're here. Fixes bug 11635. Bug: 11635 Change-Id: Ie583ade522702cbe1bbcea4475a535caa1d74fa2 Reviewed-on: https://code.wireshark.org/review/11295 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2015-10-26 15:34:58 +00:00
bool enable_timing = true;
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
RTP Player: Possibility to remove stream from RTP player When many streams are analyzed, some streams can be found as useless during replay. Patch adds option to remove it from list in RTP player. Once stream is removed, it can't be returned except close/reopen RTP player with new selection.
2021-02-04 14:03:45 +00:00
if (ui->streamTreeWidget->topLevelItemCount() < 1)
enable_play = false;
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
for (int row = 0; row < ui->streamTreeWidget->topLevelItemCount(); row++) {
QTreeWidgetItem *ti = ui->streamTreeWidget->topLevelItem(row);
RTP dissector: Remove unused functions Patch removes (just comments out) functions not used by the code as found by Martin Mathieson.
2021-02-05 09:19:48 +00:00
RtpAudioStream *audio_stream = ti->data(stream_data_col_, Qt::UserRole).value<RtpAudioStream*>();
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
if (audio_stream->outputState() != QAudio::IdleState) {
enable_play = false;
RTP player: Added pause button Pause button added. Button is visible only when media are played.
2020-12-29 13:41:14 +00:00
enable_pause = true;
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
enable_stop = true;
Add jitter logic to RtpAudioStream. Copy the jitter logic from rtp_player.c to rtp_audio_stream.cpp. This still isn't correct but the RTP player should now be complete enough to start looking at the bug list at the top of rtp_player_dialog.cpp. Disable timing and jitter controls while we're playing while we're here. Fixes bug 11635. Bug: 11635 Change-Id: Ie583ade522702cbe1bbcea4475a535caa1d74fa2 Reviewed-on: https://code.wireshark.org/review/11295 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2015-10-26 15:34:58 +00:00
enable_timing = false;
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
}
}
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
ui->actionAudioRoutingP->setVisible(!stereo_available_);
ui->actionAudioRoutingL->setVisible(stereo_available_);
ui->actionAudioRoutingLR->setVisible(stereo_available_);
ui->actionAudioRoutingR->setVisible(stereo_available_);
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
ui->playButton->setEnabled(enable_play);
RTP player: Added pause button Pause button added. Button is visible only when media are played.
2020-12-29 13:41:14 +00:00
if (enable_play) {
ui->playButton->setVisible(true);
ui->pauseButton->setVisible(false);
} else if (enable_pause) {
ui->playButton->setVisible(false);
ui->pauseButton->setVisible(true);
}
Qt: Make the RTP player output device selectable. Add a combobox for selecting the output device and populate it with our available devices. Let the user know if our output format isn't supported. Ping-Bug: 13105 Change-Id: I299c7d0f191bb66d93896338036000e2c377781f Reviewed-on: https://code.wireshark.org/review/19046 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2016-12-02 23:52:02 +00:00
ui->outputDeviceComboBox->setEnabled(enable_play);
RTP player: Added pause button Pause button added. Button is visible only when media are played.
2020-12-29 13:41:14 +00:00
ui->pauseButton->setEnabled(enable_pause);
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
ui->stopButton->setEnabled(enable_stop);
RTP Player: Added additional shortcuts Added shortcuts: - Mute/Unmute/Invert Mute - Play/Pause/Stop Removed outdated help in tooltip text. It makes no sense to use it, all actions are available in context menu.
2021-03-11 08:22:15 +00:00
ui->actionStop->setEnabled(enable_stop);
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
cur_play_pos_->setVisible(enable_stop);
Add jitter logic to RtpAudioStream. Copy the jitter logic from rtp_player.c to rtp_audio_stream.cpp. This still isn't correct but the RTP player should now be complete enough to start looking at the bug list at the top of rtp_player_dialog.cpp. Disable timing and jitter controls while we're playing while we're here. Fixes bug 11635. Bug: 11635 Change-Id: Ie583ade522702cbe1bbcea4475a535caa1d74fa2 Reviewed-on: https://code.wireshark.org/review/11295 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2015-10-26 15:34:58 +00:00
ui->jitterSpinBox->setEnabled(enable_timing);
ui->timingComboBox->setEnabled(enable_timing);
ui->todCheckBox->setEnabled(enable_timing);
Qt: Make the RTP player output device selectable. Add a combobox for selecting the output device and populate it with our available devices. Let the user know if our output format isn't supported. Ping-Bug: 13105 Change-Id: I299c7d0f191bb66d93896338036000e2c377781f Reviewed-on: https://code.wireshark.org/review/19046 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2016-12-02 23:52:02 +00:00
updateHintLabel();
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
ui->audioPlot->replot();
}
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
void RtpPlayerDialog::handleItemHighlight(QTreeWidgetItem *ti, bool scroll)
{
if (ti) {
if (ti != last_ti_) {
if (last_ti_) {
highlightItem(last_ti_, false);
}
highlightItem(ti, true);
if (scroll)
ui->streamTreeWidget->scrollToItem(ti, QAbstractItemView::EnsureVisible);
ui->audioPlot->replot();
last_ti_ = ti;
}
} else {
if (last_ti_) {
highlightItem(last_ti_, false);
ui->audioPlot->replot();
last_ti_ = NULL;
}
}
}
void RtpPlayerDialog::highlightItem(QTreeWidgetItem *ti, bool highlight)
{
QFont font;
RtpAudioGraph *audio_graph;
font.setBold(highlight);
for(int i=0; i<ui->streamTreeWidget->columnCount(); i++) {
ti->setFont(i, font);
}
audio_graph = ti->data(graph_audio_data_col_, Qt::UserRole).value<RtpAudioGraph*>();
if (audio_graph) {
audio_graph->setHighlight(highlight);
}
}
void RtpPlayerDialog::itemEntered(QTreeWidgetItem *item, int column _U_)
{
handleItemHighlight(item, false);
}
void RtpPlayerDialog::mouseMovePlot(QMouseEvent *event)
{
updateHintLabel();
QTreeWidgetItem *ti = findItemByCoords(event->pos());
handleItemHighlight(ti, true);
}
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
void RtpPlayerDialog::graphClicked(QMouseEvent *event)
{
updateWidgets();
if (event->button() == Qt::RightButton) {
RTP dissector: Remove unused functions Patch removes (just comments out) functions not used by the code as found by Martin Mathieson.
2021-02-05 09:19:48 +00:00
graph_ctx_menu_->exec(event->globalPos());
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
}
}
rtp_player: Player is able to set start of audio play by double click Patch adds ability to set start of audio play by double clicking on waveform. Patch fixes unreported issue with placing waveform at incorrect place when switched relative/absolute time mode (check/uncheck Time of Day). Change-Id: Ib8ce24aea870e2443e033afbb6d6e9fbcf222431 Reviewed-on: https://code.wireshark.org/review/35621 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-01 23:38:35 +00:00
void RtpPlayerDialog::graphDoubleClicked(QMouseEvent *event)
{
updateWidgets();
if (event->button() == Qt::LeftButton) {
// Move start play line
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
double ts = ui->audioPlot->xAxis->pixelToCoord(event->pos().x());
rtp_player: Player is able to set start of audio play by double click Patch adds ability to set start of audio play by double clicking on waveform. Patch fixes unreported issue with placing waveform at incorrect place when switched relative/absolute time mode (check/uncheck Time of Day). Change-Id: Ib8ce24aea870e2443e033afbb6d6e9fbcf222431 Reviewed-on: https://code.wireshark.org/review/35621 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-01 23:38:35 +00:00
setStartPlayMarker(ts);
drawStartPlayMarker();
ui->audioPlot->replot();
}
}
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
void RtpPlayerDialog::plotClicked(QCPAbstractPlottable *plottable _U_, int dataIndex _U_, QMouseEvent *event)
{
// Delivered plottable very often points to different element than a mouse
// so we find right one by mouse coordinates
QTreeWidgetItem *ti = findItemByCoords(event->pos());
if (ti) {
if (event->modifiers() == Qt::NoModifier) {
ti->setSelected(true);
} else if (event->modifiers() == Qt::ControlModifier) {
ti->setSelected(!ti->isSelected());
}
}
}
QTreeWidgetItem *RtpPlayerDialog::findItemByCoords(QPoint point)
{
QCPAbstractPlottable *plottable=ui->audioPlot->plottableAt(point);
if (plottable) {
return findItem(plottable);
}
return NULL;
}
QTreeWidgetItem *RtpPlayerDialog::findItem(QCPAbstractPlottable *plottable)
RTP Player: Selected waveform is emphased. Waveform can be selected by click. When there is many waveforms, it is not clear which one is selected in list. It is important for selecting packets from graph. Patch emphasis selected waveform. In addition, it allows to select waveform by clicking on it.
2021-02-04 13:23:43 +00:00
{
for (int row = 0; row < ui->streamTreeWidget->topLevelItemCount(); row++) {
QTreeWidgetItem *ti = ui->streamTreeWidget->topLevelItem(row);
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
RtpAudioGraph *audio_graph = ti->data(graph_audio_data_col_, Qt::UserRole).value<RtpAudioGraph*>();
if (audio_graph->isMyPlottable(plottable)) {
return ti;
RTP Player: Selected waveform is emphased. Waveform can be selected by click. When there is many waveforms, it is not clear which one is selected in list. It is important for selecting packets from graph. Patch emphasis selected waveform. In addition, it allows to select waveform by clicking on it.
2021-02-04 13:23:43 +00:00
}
}
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
return NULL;
RTP Player: Selected waveform is emphased. Waveform can be selected by click. When there is many waveforms, it is not clear which one is selected in list. It is important for selecting packets from graph. Patch emphasis selected waveform. In addition, it allows to select waveform by clicking on it.
2021-02-04 13:23:43 +00:00
}
Qt: Make the RTP player output device selectable. Add a combobox for selecting the output device and populate it with our available devices. Let the user know if our output format isn't supported. Ping-Bug: 13105 Change-Id: I299c7d0f191bb66d93896338036000e2c377781f Reviewed-on: https://code.wireshark.org/review/19046 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2016-12-02 23:52:02 +00:00
void RtpPlayerDialog::updateHintLabel()
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
{
int packet_num = getHoveredPacket();
QString hint = "<small><i>";
rtp_player: Player is able to set start of audio play by double click Patch adds ability to set start of audio play by double clicking on waveform. Patch fixes unreported issue with placing waveform at incorrect place when switched relative/absolute time mode (check/uncheck Time of Day). Change-Id: Ib8ce24aea870e2443e033afbb6d6e9fbcf222431 Reviewed-on: https://code.wireshark.org/review/35621 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-01 23:38:35 +00:00
double start_pos = getStartPlayMarker();
if (packet_num == 0) {
hint += tr("Start: %1. Double click to set start of playback.")
.arg(getFormatedTime(start_pos));
} else if (packet_num > 0) {
hint += tr("Start: %1, cursor: %2. Press \"G\" to go to packet %3. Double click to set start of playback.")
.arg(getFormatedTime(start_pos))
.arg(getFormatedHoveredTime())
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
.arg(packet_num);
Qt: Make the RTP player output device selectable. Add a combobox for selecting the output device and populate it with our available devices. Let the user know if our output format isn't supported. Ping-Bug: 13105 Change-Id: I299c7d0f191bb66d93896338036000e2c377781f Reviewed-on: https://code.wireshark.org/review/19046 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2016-12-02 23:52:02 +00:00
} else if (!playback_error_.isEmpty()) {
hint += playback_error_;
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
}
hint += "</i></small>";
ui->hintLabel->setText(hint);
}
void RtpPlayerDialog::resetXAxis()
{
QCustomPlot *ap = ui->audioPlot;
double pixel_pad = 10.0; // per side
ap->rescaleAxes(true);
double axis_pixels = ap->xAxis->axisRect()->width();
RTP Player: x axis centering correction When waveform start much later than at time 0s, centering of waveform was incorrect. The reason was that range was taken before rescalingAxes. Patch solves the issue.
2021-01-02 12:45:43 +00:00
ap->xAxis->scaleRange((axis_pixels + (pixel_pad * 2)) / axis_pixels, ap->xAxis->range().center());
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
axis_pixels = ap->yAxis->axisRect()->height();
ap->yAxis->scaleRange((axis_pixels + (pixel_pad * 2)) / axis_pixels, ap->yAxis->range().center());
ap->replot();
}
RTP Player: UI improvements Changes: - all waveforms has common scale therefore louder/quiter signal is visible - when stream/streams are deleted from view, Y axis is rescaled and waveforms are rearranged to reuse empty space
2021-03-23 08:48:20 +00:00
void RtpPlayerDialog::updateGraphs()
{
QCustomPlot *ap = ui->audioPlot;
// Create new plots, just existing ones
createPlot(false);
// Rescale Y axis
double pixel_pad = 10.0; // per side
double axis_pixels = ap->yAxis->axisRect()->height();
ap->yAxis->rescale(true);
ap->yAxis->scaleRange((axis_pixels + (pixel_pad * 2)) / axis_pixels, ap->yAxis->range().center());
ap->replot();
}
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
void RtpPlayerDialog::playFinished(RtpAudioStream *stream)
{
playing_streams_.removeOne(stream);
if (playing_streams_.isEmpty()) {
updateWidgets();
}
}
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
void RtpPlayerDialog::setPlayPosition(double secs)
{
double cur_secs = cur_play_pos_->point1->key();
rtp_player: Player is able to set start of audio play by double click Patch adds ability to set start of audio play by double clicking on waveform. Patch fixes unreported issue with placing waveform at incorrect place when switched relative/absolute time mode (check/uncheck Time of Day). Change-Id: Ib8ce24aea870e2443e033afbb6d6e9fbcf222431 Reviewed-on: https://code.wireshark.org/review/35621 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-01 23:38:35 +00:00
if (ui->todCheckBox->isChecked()) {
secs += first_stream_abs_start_time_;
} else {
secs += first_stream_rel_start_time_;
}
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
if (secs > cur_secs) {
cur_play_pos_->point1->setCoords(secs, 0.0);
cur_play_pos_->point2->setCoords(secs, 1.0);
ui->audioPlot->replot();
}
}
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
void RtpPlayerDialog::setPlaybackError(const QString playback_error)
{
playback_error_ = playback_error;
updateHintLabel();
}
Add a "failed" return for tap packet routines. This allows taps that can fail to report an error and fail; a failed tap's packet routine won't be called again, so they don't have to keep track of whether they've failed themselves. We make the return value from the packet routine an enum. Don't have a separate type for the per-packet routine for "follow" taps; they're expected to act like tap packet routines, so just use the type for tap packet routines. One tap packet routine returned -1; that's not a valid return value, and wasn't one before this change (the return value was a boolean), so presume the intent was "don't redraw". Another tap routine's early return, without doing any work, returned TRUE; this is presumably an error (no work done, no need to redraw), so presumably it should be "don't redraw". Clean up some white space while we're at it. Change-Id: Ia7d2b717b2cace4b13c2b886e699aa4d79cc82c8 Reviewed-on: https://code.wireshark.org/review/31283 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2019-01-01 03:36:12 +00:00
tap_packet_status RtpPlayerDialog::tapPacket(void *tapinfo_ptr, packet_info *pinfo, epan_dissect_t *, const void *rtpinfo_ptr)
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
{
RtpPlayerDialog *rtp_player_dialog = dynamic_cast<RtpPlayerDialog *>((RtpPlayerDialog*)tapinfo_ptr);
Add a "failed" return for tap packet routines. This allows taps that can fail to report an error and fail; a failed tap's packet routine won't be called again, so they don't have to keep track of whether they've failed themselves. We make the return value from the packet routine an enum. Don't have a separate type for the per-packet routine for "follow" taps; they're expected to act like tap packet routines, so just use the type for tap packet routines. One tap packet routine returned -1; that's not a valid return value, and wasn't one before this change (the return value was a boolean), so presume the intent was "don't redraw". Another tap routine's early return, without doing any work, returned TRUE; this is presumably an error (no work done, no need to redraw), so presumably it should be "don't redraw". Clean up some white space while we're at it. Change-Id: Ia7d2b717b2cace4b13c2b886e699aa4d79cc82c8 Reviewed-on: https://code.wireshark.org/review/31283 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2019-01-01 03:36:12 +00:00
if (!rtp_player_dialog) return TAP_PACKET_DONT_REDRAW;
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
const struct _rtp_info *rtpinfo = (const struct _rtp_info *)rtpinfo_ptr;
Add a "failed" return for tap packet routines. This allows taps that can fail to report an error and fail; a failed tap's packet routine won't be called again, so they don't have to keep track of whether they've failed themselves. We make the return value from the packet routine an enum. Don't have a separate type for the per-packet routine for "follow" taps; they're expected to act like tap packet routines, so just use the type for tap packet routines. One tap packet routine returned -1; that's not a valid return value, and wasn't one before this change (the return value was a boolean), so presume the intent was "don't redraw". Another tap routine's early return, without doing any work, returned TRUE; this is presumably an error (no work done, no need to redraw), so presumably it should be "don't redraw". Clean up some white space while we're at it. Change-Id: Ia7d2b717b2cace4b13c2b886e699aa4d79cc82c8 Reviewed-on: https://code.wireshark.org/review/31283 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2019-01-01 03:36:12 +00:00
if (!rtpinfo) return TAP_PACKET_DONT_REDRAW;
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
RTP Player: Shows streams independently on active display filter This patch solves issue #13325.
2021-01-01 16:12:11 +00:00
/* ignore RTP Version != 2 */
if (rtpinfo->info_version != 2)
Add a "failed" return for tap packet routines. This allows taps that can fail to report an error and fail; a failed tap's packet routine won't be called again, so they don't have to keep track of whether they've failed themselves. We make the return value from the packet routine an enum. Don't have a separate type for the per-packet routine for "follow" taps; they're expected to act like tap packet routines, so just use the type for tap packet routines. One tap packet routine returned -1; that's not a valid return value, and wasn't one before this change (the return value was a boolean), so presume the intent was "don't redraw". Another tap routine's early return, without doing any work, returned TRUE; this is presumably an error (no work done, no need to redraw), so presumably it should be "don't redraw". Clean up some white space while we're at it. Change-Id: Ia7d2b717b2cace4b13c2b886e699aa4d79cc82c8 Reviewed-on: https://code.wireshark.org/review/31283 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2019-01-01 03:36:12 +00:00
return TAP_PACKET_DONT_REDRAW;
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
rtp_player_dialog->addPacket(pinfo, rtpinfo);
Add a "failed" return for tap packet routines. This allows taps that can fail to report an error and fail; a failed tap's packet routine won't be called again, so they don't have to keep track of whether they've failed themselves. We make the return value from the packet routine an enum. Don't have a separate type for the per-packet routine for "follow" taps; they're expected to act like tap packet routines, so just use the type for tap packet routines. One tap packet routine returned -1; that's not a valid return value, and wasn't one before this change (the return value was a boolean), so presume the intent was "don't redraw". Another tap routine's early return, without doing any work, returned TRUE; this is presumably an error (no work done, no need to redraw), so presumably it should be "don't redraw". Clean up some white space while we're at it. Change-Id: Ia7d2b717b2cace4b13c2b886e699aa4d79cc82c8 Reviewed-on: https://code.wireshark.org/review/31283 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2019-01-01 03:36:12 +00:00
return TAP_PACKET_DONT_REDRAW;
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
}
void RtpPlayerDialog::addPacket(packet_info *pinfo, const _rtp_info *rtpinfo)
{
for (int row = 0; row < ui->streamTreeWidget->topLevelItemCount(); row++) {
QTreeWidgetItem *ti = ui->streamTreeWidget->topLevelItem(row);
RtpAudioStream *row_stream = ti->data(stream_data_col_, Qt::UserRole).value<RtpAudioStream*>();
if (row_stream->isMatch(pinfo, rtpinfo)) {
row_stream->addRtpPacket(pinfo, rtpinfo);
return;
}
}
// qDebug() << "=ap no match!" << address_to_qstring(&pinfo->src) << address_to_qstring(&pinfo->dst);
}
void RtpPlayerDialog::zoomXAxis(bool in)
{
QCustomPlot *ap = ui->audioPlot;
double h_factor = ap->axisRect()->rangeZoomFactor(Qt::Horizontal);
if (!in) {
h_factor = pow(h_factor, -1);
}
ap->xAxis->scaleRange(h_factor, ap->xAxis->range().center());
ap->replot();
}
// XXX I tried using seconds but pixels make more sense at varying zoom
// levels.
void RtpPlayerDialog::panXAxis(int x_pixels)
{
QCustomPlot *ap = ui->audioPlot;
double h_pan;
h_pan = ap->xAxis->range().size() * x_pixels / ap->xAxis->axisRect()->width();
if (x_pixels) {
ap->xAxis->moveRange(h_pan);
ap->replot();
}
}
void RtpPlayerDialog::on_playButton_clicked()
{
rtp_player: Player is able to set start of audio play by double click Patch adds ability to set start of audio play by double clicking on waveform. Patch fixes unreported issue with placing waveform at incorrect place when switched relative/absolute time mode (check/uncheck Time of Day). Change-Id: Ib8ce24aea870e2443e033afbb6d6e9fbcf222431 Reviewed-on: https://code.wireshark.org/review/35621 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-01 23:38:35 +00:00
double start_time;
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
ui->hintLabel->setText("<i><small>" + tr("Preparing to play...") + "</i></small>");
wsApp->processEvents();
RTP player: Added pause button Pause button added. Button is visible only when media are played.
2020-12-29 13:41:14 +00:00
ui->pauseButton->setChecked(false);
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
// Protect start time against move of marker during the play
start_marker_time_play_ = start_marker_time_;
cur_play_pos_->point1->setCoords(start_marker_time_play_, 0.0);
cur_play_pos_->point2->setCoords(start_marker_time_play_, 1.0);
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
cur_play_pos_->setVisible(true);
Qt: Make the RTP player output device selectable. Add a combobox for selecting the output device and populate it with our available devices. Let the user know if our output format isn't supported. Ping-Bug: 13105 Change-Id: I299c7d0f191bb66d93896338036000e2c377781f Reviewed-on: https://code.wireshark.org/review/19046 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2016-12-02 23:52:02 +00:00
playback_error_.clear();
rtp_player: Player is able to set start of audio play by double click Patch adds ability to set start of audio play by double clicking on waveform. Patch fixes unreported issue with placing waveform at incorrect place when switched relative/absolute time mode (check/uncheck Time of Day). Change-Id: Ib8ce24aea870e2443e033afbb6d6e9fbcf222431 Reviewed-on: https://code.wireshark.org/review/35621 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-01 23:38:35 +00:00
if (ui->todCheckBox->isChecked()) {
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
start_time = start_marker_time_play_;
rtp_player: Player is able to set start of audio play by double click Patch adds ability to set start of audio play by double clicking on waveform. Patch fixes unreported issue with placing waveform at incorrect place when switched relative/absolute time mode (check/uncheck Time of Day). Change-Id: Ib8ce24aea870e2443e033afbb6d6e9fbcf222431 Reviewed-on: https://code.wireshark.org/review/35621 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-01 23:38:35 +00:00
} else {
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
start_time = start_marker_time_play_ - first_stream_rel_start_time_;
rtp_player: Player is able to set start of audio play by double click Patch adds ability to set start of audio play by double clicking on waveform. Patch fixes unreported issue with placing waveform at incorrect place when switched relative/absolute time mode (check/uncheck Time of Day). Change-Id: Ib8ce24aea870e2443e033afbb6d6e9fbcf222431 Reviewed-on: https://code.wireshark.org/review/35621 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-01 23:38:35 +00:00
}
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
QAudioDeviceInfo cur_out_device = getCurrentDeviceInfo();
playing_streams_.clear();
rtp_player: Player is able to set start of audio play by double click Patch adds ability to set start of audio play by double clicking on waveform. Patch fixes unreported issue with placing waveform at incorrect place when switched relative/absolute time mode (check/uncheck Time of Day). Change-Id: Ib8ce24aea870e2443e033afbb6d6e9fbcf222431 Reviewed-on: https://code.wireshark.org/review/35621 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-01 23:38:35 +00:00
int row_count = ui->streamTreeWidget->topLevelItemCount();
for (int row = 0; row < row_count; row++) {
QTreeWidgetItem *ti = ui->streamTreeWidget->topLevelItem(row);
RtpAudioStream *audio_stream = ti->data(stream_data_col_, Qt::UserRole).value<RtpAudioStream*>();
// All streams starts at first_stream_rel_start_time_
audio_stream->setStartPlayTime(start_time);
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
if (audio_stream->prepareForPlay(cur_out_device)) {
playing_streams_ << audio_stream;
}
rtp_player: Player is able to set start of audio play by double click Patch adds ability to set start of audio play by double clicking on waveform. Patch fixes unreported issue with placing waveform at incorrect place when switched relative/absolute time mode (check/uncheck Time of Day). Change-Id: Ib8ce24aea870e2443e033afbb6d6e9fbcf222431 Reviewed-on: https://code.wireshark.org/review/35621 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-01 23:38:35 +00:00
}
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
// Prepare silent stream for progress marker
if (!marker_stream_) {
marker_stream_ = getSilenceAudioOutput();
} else {
marker_stream_->stop();
}
// Start progress marker and then audio streams
marker_stream_->start(new AudioSilenceGenerator());
for( int i = 0; i<playing_streams_.count(); ++i ) {
playing_streams_[i]->startPlaying();
}
updateWidgets();
}
QAudioDeviceInfo RtpPlayerDialog::getCurrentDeviceInfo()
{
QAudioDeviceInfo cur_out_device = QAudioDeviceInfo::defaultOutputDevice();
QString cur_out_name = currentOutputDeviceName();
foreach (QAudioDeviceInfo out_device, QAudioDeviceInfo::availableDevices(QAudio::AudioOutput)) {
if (cur_out_name == out_device.deviceName()) {
cur_out_device = out_device;
}
}
return cur_out_device;
}
QAudioOutput *RtpPlayerDialog::getSilenceAudioOutput()
{
QAudioOutput *o;
QAudioDeviceInfo cur_out_device = getCurrentDeviceInfo();
QAudioFormat format;
format.setSampleRate(8000);
format.setSampleSize(SAMPLE_BYTES * 8); // bits
format.setSampleType(QAudioFormat::SignedInt);
if (stereo_available_) {
format.setChannelCount(2);
} else {
format.setChannelCount(1);
}
format.setCodec("audio/pcm");
o = new QAudioOutput(cur_out_device, format, this);
o->setNotifyInterval(100); // ~15 fps
connect(o, SIGNAL(notify()), this, SLOT(outputNotify()));
return o;
}
void RtpPlayerDialog::outputNotify()
{
double secs = marker_stream_->processedUSecs() / 1000000.0;
if (ui->todCheckBox->isChecked()) {
secs += start_marker_time_play_;
} else {
secs += start_marker_time_play_;
secs -= first_stream_rel_start_time_;
}
setPlayPosition(secs);
}
void RtpPlayerDialog::on_pauseButton_clicked()
{
for( int i = 0; i<playing_streams_.count(); ++i ) {
playing_streams_[i]->pausePlaying();
}
if (ui->pauseButton->isChecked()) {
marker_stream_->suspend();
} else {
marker_stream_->resume();
}
updateWidgets();
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
}
void RtpPlayerDialog::on_stopButton_clicked()
{
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
// We need copy of list because items will be removed during stopPlaying()
QList<RtpAudioStream *> ps=QList<RtpAudioStream *>(playing_streams_);
for( int i = 0; i<ps.count(); ++i ) {
ps[i]->stopPlaying();
}
marker_stream_->stop();
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
cur_play_pos_->setVisible(false);
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
updateWidgets();
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
}
void RtpPlayerDialog::on_actionReset_triggered()
{
resetXAxis();
}
void RtpPlayerDialog::on_actionZoomIn_triggered()
{
zoomXAxis(true);
}
void RtpPlayerDialog::on_actionZoomOut_triggered()
{
zoomXAxis(false);
}
void RtpPlayerDialog::on_actionMoveLeft10_triggered()
{
panXAxis(-10);
}
void RtpPlayerDialog::on_actionMoveRight10_triggered()
{
panXAxis(10);
}
void RtpPlayerDialog::on_actionMoveLeft1_triggered()
{
panXAxis(-1);
}
void RtpPlayerDialog::on_actionMoveRight1_triggered()
{
panXAxis(1);
}
void RtpPlayerDialog::on_actionGoToPacket_triggered()
{
int packet_num = getHoveredPacket();
if (packet_num > 0) emit goToPacket(packet_num);
}
RTP player: It is possible to select SETUP packet related to RTP stream When SETUP (or related signalling packet) exists, it is selected by Shift+G. If RTP stream is "alone", first RTP packet of the stream is selected.
2021-03-09 13:47:07 +00:00
void RtpPlayerDialog::handleGoToSetupPacket(QTreeWidgetItem *ti)
{
if (ti) {
bool ok;
int packet_num = ti->data(first_pkt_col_, Qt::UserRole).toInt(&ok);
if (ok) {
emit goToPacket(packet_num);
}
}
}
void RtpPlayerDialog::on_actionGoToSetupPacketPlot_triggered()
{
QPoint pos = ui->audioPlot->mapFromGlobal(QCursor::pos());
handleGoToSetupPacket(findItemByCoords(pos));
}
void RtpPlayerDialog::on_actionGoToSetupPacketTree_triggered()
{
handleGoToSetupPacket(last_ti_);
}
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
// Make waveform graphs selectable and update the treewidget selection accordingly.
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
void RtpPlayerDialog::on_streamTreeWidget_itemSelectionChanged()
{
for (int row = 0; row < ui->streamTreeWidget->topLevelItemCount(); row++) {
QTreeWidgetItem *ti = ui->streamTreeWidget->topLevelItem(row);
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
RtpAudioGraph *audio_graph = ti->data(graph_audio_data_col_, Qt::UserRole).value<RtpAudioGraph*>();
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
if (audio_graph) {
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
audio_graph->setSelected(ti->isSelected());
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
}
}
ui->audioPlot->replot();
}
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
// Change channel audio routing if double clicked channel column
rtp_player_dialog: Route audio for a stream to left/right speaker in RTP player Column 'Play' added to player. Double click on a stream in the column changes audio routing for the stream. When soundcard supports only one channel, there are Mute/Play option. When soundcard supports two or more channels, there are Mute/L/L+R/R options. Muted channel is drawn with dotted line. Change-Id: If120c902195da46f98a1663c589f20c6a1da0ba7 Reviewed-on: https://code.wireshark.org/review/35687 Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-07 21:16:42 +00:00
void RtpPlayerDialog::on_streamTreeWidget_itemDoubleClicked(QTreeWidgetItem *item, const int column)
{
if (column == channel_col_) {
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
RtpAudioStream *audio_stream = item->data(stream_data_col_, Qt::UserRole).value<RtpAudioStream*>();
if (!audio_stream)
return;
AudioRouting audio_routing = audio_stream->getAudioRouting();
audio_routing = audio_routing.getNextChannel(stereo_available_);
changeAudioRoutingOnItem(item, audio_routing);
rtp_player_dialog: Route audio for a stream to left/right speaker in RTP player Column 'Play' added to player. Double click on a stream in the column changes audio routing for the stream. When soundcard supports only one channel, there are Mute/Play option. When soundcard supports two or more channels, there are Mute/L/L+R/R options. Muted channel is drawn with dotted line. Change-Id: If120c902195da46f98a1663c589f20c6a1da0ba7 Reviewed-on: https://code.wireshark.org/review/35687 Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-07 21:16:42 +00:00
}
}
RTP Player: Dialog is nonmodal now and can be called multiple ways Changes: - refactored main_dialog handling of telephony dialogs - RTP Player dialog is nonmodal now and can be left open - it is possible to issue three actions on RTP Player dialog from other dialogs (other dialog have selected set of RTP streams before action) - replace - removes existing streams from RTP dialog and shows new set - add - adds new set to existing list in RTP dialog - remove - remove streams in set from list in RTP dialog - Sequence Dialog: - was modified to hold rtpstream_info_t for RTP streams - added Play button - VoIP features (RTP Play button, select/deselect RTP stream) are disabled after creation and must be enabled. It handles that RTP Play button is not shown e.g. in TCP sequence show
2021-03-27 20:01:43 +00:00
void RtpPlayerDialog::removeRow(QTreeWidgetItem *ti)
RTP Player: Possibility to remove stream from RTP player When many streams are analyzed, some streams can be found as useless during replay. Patch adds option to remove it from list in RTP player. Once stream is removed, it can't be returned except close/reopen RTP player with new selection.
2021-02-04 14:03:45 +00:00
{
RTP Player: Dialog is nonmodal now and can be called multiple ways Changes: - refactored main_dialog handling of telephony dialogs - RTP Player dialog is nonmodal now and can be left open - it is possible to issue three actions on RTP Player dialog from other dialogs (other dialog have selected set of RTP streams before action) - replace - removes existing streams from RTP dialog and shows new set - add - adds new set to existing list in RTP dialog - remove - remove streams in set from list in RTP dialog - Sequence Dialog: - was modified to hold rtpstream_info_t for RTP streams - added Play button - VoIP features (RTP Play button, select/deselect RTP stream) are disabled after creation and must be enabled. It handles that RTP Play button is not shown e.g. in TCP sequence show
2021-03-27 20:01:43 +00:00
RtpAudioStream *audio_stream = ti->data(stream_data_col_, Qt::UserRole).value<RtpAudioStream*>();
if (audio_stream) {
ti->setData(stream_data_col_, Qt::UserRole, QVariant());
delete audio_stream;
RTP Player: Possibility to remove stream from RTP player When many streams are analyzed, some streams can be found as useless during replay. Patch adds option to remove it from list in RTP player. Once stream is removed, it can't be returned except close/reopen RTP player with new selection.
2021-02-04 14:03:45 +00:00
}
RTP Player: Dialog is nonmodal now and can be called multiple ways Changes: - refactored main_dialog handling of telephony dialogs - RTP Player dialog is nonmodal now and can be left open - it is possible to issue three actions on RTP Player dialog from other dialogs (other dialog have selected set of RTP streams before action) - replace - removes existing streams from RTP dialog and shows new set - add - adds new set to existing list in RTP dialog - remove - remove streams in set from list in RTP dialog - Sequence Dialog: - was modified to hold rtpstream_info_t for RTP streams - added Play button - VoIP features (RTP Play button, select/deselect RTP stream) are disabled after creation and must be enabled. It handles that RTP Play button is not shown e.g. in TCP sequence show
2021-03-27 20:01:43 +00:00
RtpAudioGraph *audio_graph = ti->data(graph_audio_data_col_, Qt::UserRole).value<RtpAudioGraph*>();
if (audio_graph) {
ti->setData(graph_audio_data_col_, Qt::UserRole, QVariant());
audio_graph->remove(ui->audioPlot);
}
RTP Player: Possibility to remove stream from RTP player When many streams are analyzed, some streams can be found as useless during replay. Patch adds option to remove it from list in RTP player. Once stream is removed, it can't be returned except close/reopen RTP player with new selection.
2021-02-04 14:03:45 +00:00
RTP Player: Dialog is nonmodal now and can be called multiple ways Changes: - refactored main_dialog handling of telephony dialogs - RTP Player dialog is nonmodal now and can be left open - it is possible to issue three actions on RTP Player dialog from other dialogs (other dialog have selected set of RTP streams before action) - replace - removes existing streams from RTP dialog and shows new set - add - adds new set to existing list in RTP dialog - remove - remove streams in set from list in RTP dialog - Sequence Dialog: - was modified to hold rtpstream_info_t for RTP streams - added Play button - VoIP features (RTP Play button, select/deselect RTP stream) are disabled after creation and must be enabled. It handles that RTP Play button is not shown e.g. in TCP sequence show
2021-03-27 20:01:43 +00:00
QCPGraph *graph;
graph = ti->data(graph_sequence_data_col_, Qt::UserRole).value<QCPGraph*>();
if (graph) {
ti->setData(graph_sequence_data_col_, Qt::UserRole, QVariant());
ui->audioPlot->removeGraph(graph);
}
RTP Player: Possibility to remove stream from RTP player When many streams are analyzed, some streams can be found as useless during replay. Patch adds option to remove it from list in RTP player. Once stream is removed, it can't be returned except close/reopen RTP player with new selection.
2021-02-04 14:03:45 +00:00
RTP Player: Dialog is nonmodal now and can be called multiple ways Changes: - refactored main_dialog handling of telephony dialogs - RTP Player dialog is nonmodal now and can be left open - it is possible to issue three actions on RTP Player dialog from other dialogs (other dialog have selected set of RTP streams before action) - replace - removes existing streams from RTP dialog and shows new set - add - adds new set to existing list in RTP dialog - remove - remove streams in set from list in RTP dialog - Sequence Dialog: - was modified to hold rtpstream_info_t for RTP streams - added Play button - VoIP features (RTP Play button, select/deselect RTP stream) are disabled after creation and must be enabled. It handles that RTP Play button is not shown e.g. in TCP sequence show
2021-03-27 20:01:43 +00:00
graph = ti->data(graph_jitter_data_col_, Qt::UserRole).value<QCPGraph*>();
if (graph) {
ti->setData(graph_jitter_data_col_, Qt::UserRole, QVariant());
ui->audioPlot->removeGraph(graph);
}
RTP Player: Possibility to remove stream from RTP player When many streams are analyzed, some streams can be found as useless during replay. Patch adds option to remove it from list in RTP player. Once stream is removed, it can't be returned except close/reopen RTP player with new selection.
2021-02-04 14:03:45 +00:00
RTP Player: Dialog is nonmodal now and can be called multiple ways Changes: - refactored main_dialog handling of telephony dialogs - RTP Player dialog is nonmodal now and can be left open - it is possible to issue three actions on RTP Player dialog from other dialogs (other dialog have selected set of RTP streams before action) - replace - removes existing streams from RTP dialog and shows new set - add - adds new set to existing list in RTP dialog - remove - remove streams in set from list in RTP dialog - Sequence Dialog: - was modified to hold rtpstream_info_t for RTP streams - added Play button - VoIP features (RTP Play button, select/deselect RTP stream) are disabled after creation and must be enabled. It handles that RTP Play button is not shown e.g. in TCP sequence show
2021-03-27 20:01:43 +00:00
graph = ti->data(graph_timestamp_data_col_, Qt::UserRole).value<QCPGraph*>();
if (graph) {
ti->setData(graph_timestamp_data_col_, Qt::UserRole, QVariant());
ui->audioPlot->removeGraph(graph);
}
RTP Player: Possibility to remove stream from RTP player When many streams are analyzed, some streams can be found as useless during replay. Patch adds option to remove it from list in RTP player. Once stream is removed, it can't be returned except close/reopen RTP player with new selection.
2021-02-04 14:03:45 +00:00
RTP Player: Dialog is nonmodal now and can be called multiple ways Changes: - refactored main_dialog handling of telephony dialogs - RTP Player dialog is nonmodal now and can be left open - it is possible to issue three actions on RTP Player dialog from other dialogs (other dialog have selected set of RTP streams before action) - replace - removes existing streams from RTP dialog and shows new set - add - adds new set to existing list in RTP dialog - remove - remove streams in set from list in RTP dialog - Sequence Dialog: - was modified to hold rtpstream_info_t for RTP streams - added Play button - VoIP features (RTP Play button, select/deselect RTP stream) are disabled after creation and must be enabled. It handles that RTP Play button is not shown e.g. in TCP sequence show
2021-03-27 20:01:43 +00:00
graph = ti->data(graph_silence_data_col_, Qt::UserRole).value<QCPGraph*>();
if (graph) {
ti->setData(graph_silence_data_col_, Qt::UserRole, QVariant());
ui->audioPlot->removeGraph(graph);
}
RTP Player: Possibility to remove stream from RTP player When many streams are analyzed, some streams can be found as useless during replay. Patch adds option to remove it from list in RTP player. Once stream is removed, it can't be returned except close/reopen RTP player with new selection.
2021-02-04 14:03:45 +00:00
RTP Player: Dialog is nonmodal now and can be called multiple ways Changes: - refactored main_dialog handling of telephony dialogs - RTP Player dialog is nonmodal now and can be left open - it is possible to issue three actions on RTP Player dialog from other dialogs (other dialog have selected set of RTP streams before action) - replace - removes existing streams from RTP dialog and shows new set - add - adds new set to existing list in RTP dialog - remove - remove streams in set from list in RTP dialog - Sequence Dialog: - was modified to hold rtpstream_info_t for RTP streams - added Play button - VoIP features (RTP Play button, select/deselect RTP stream) are disabled after creation and must be enabled. It handles that RTP Play button is not shown e.g. in TCP sequence show
2021-03-27 20:01:43 +00:00
delete ti;
}
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
RTP Player: Dialog is nonmodal now and can be called multiple ways Changes: - refactored main_dialog handling of telephony dialogs - RTP Player dialog is nonmodal now and can be left open - it is possible to issue three actions on RTP Player dialog from other dialogs (other dialog have selected set of RTP streams before action) - replace - removes existing streams from RTP dialog and shows new set - add - adds new set to existing list in RTP dialog - remove - remove streams in set from list in RTP dialog - Sequence Dialog: - was modified to hold rtpstream_info_t for RTP streams - added Play button - VoIP features (RTP Play button, select/deselect RTP stream) are disabled after creation and must be enabled. It handles that RTP Play button is not shown e.g. in TCP sequence show
2021-03-27 20:01:43 +00:00
void RtpPlayerDialog::on_actionRemoveStream_triggered()
{
QList<QTreeWidgetItem *> items = ui->streamTreeWidget->selectedItems();
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
RTP Player: Dialog is nonmodal now and can be called multiple ways Changes: - refactored main_dialog handling of telephony dialogs - RTP Player dialog is nonmodal now and can be left open - it is possible to issue three actions on RTP Player dialog from other dialogs (other dialog have selected set of RTP streams before action) - replace - removes existing streams from RTP dialog and shows new set - add - adds new set to existing list in RTP dialog - remove - remove streams in set from list in RTP dialog - Sequence Dialog: - was modified to hold rtpstream_info_t for RTP streams - added Play button - VoIP features (RTP Play button, select/deselect RTP stream) are disabled after creation and must be enabled. It handles that RTP Play button is not shown e.g. in TCP sequence show
2021-03-27 20:01:43 +00:00
if (last_ti_) {
highlightItem(last_ti_, false);
last_ti_ = NULL;
}
for(int i = 0; i<items.count(); i++ ) {
removeRow(items[i]);
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
}
RTP Player: UI improvements Changes: - all waveforms has common scale therefore louder/quiter signal is visible - when stream/streams are deleted from view, Y axis is rescaled and waveforms are rearranged to reuse empty space
2021-03-23 08:48:20 +00:00
// TODO: Recalculate legend
// - Graphs used for legend could be removed above and we must add new
// - If no legend is required, it should be removed
// Redraw existing waveforms and rescale Y axis
updateGraphs();
RTP Player: Possibility to remove stream from RTP player When many streams are analyzed, some streams can be found as useless during replay. Patch adds option to remove it from list in RTP player. Once stream is removed, it can't be returned except close/reopen RTP player with new selection.
2021-02-04 14:03:45 +00:00
updateWidgets();
}
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
// If called with channel_any, just muted flag should be changed
void RtpPlayerDialog::changeAudioRoutingOnItem(QTreeWidgetItem *ti, AudioRouting new_audio_routing)
{
if (!ti)
return;
RtpAudioStream *audio_stream = ti->data(stream_data_col_, Qt::UserRole).value<RtpAudioStream*>();
if (!audio_stream)
return;
AudioRouting audio_routing = audio_stream->getAudioRouting();
audio_routing.mergeAudioRouting(new_audio_routing);
formatAudioRouting(ti, audio_routing);
audio_stream->setAudioRouting(audio_routing);
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
RtpAudioGraph *audio_graph = ti->data(graph_audio_data_col_, Qt::UserRole).value<RtpAudioGraph*>();
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
if (audio_graph) {
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
audio_graph->setSelected(ti->isSelected());
audio_graph->setMuted(audio_routing.isMuted());
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
ui->audioPlot->replot();
}
}
// Find current item and apply change on it
void RtpPlayerDialog::changeAudioRouting(AudioRouting new_audio_routing)
RTP dissector: Remove unused functions Patch removes (just comments out) functions not used by the code as found by Martin Mathieson.
2021-02-05 09:19:48 +00:00
{
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
QList<QTreeWidgetItem *> items = ui->streamTreeWidget->selectedItems();
for(int i = 0; i<items.count(); i++ ) {
QTreeWidgetItem *ti = items[i];
changeAudioRoutingOnItem(ti, new_audio_routing);
}
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
}
// Invert mute/unmute on item
void RtpPlayerDialog::invertAudioMutingOnItem(QTreeWidgetItem *ti)
{
RTP dissector: Remove unused functions Patch removes (just comments out) functions not used by the code as found by Martin Mathieson.
2021-02-05 09:19:48 +00:00
if (!ti)
return;
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
RtpAudioStream *audio_stream = ti->data(stream_data_col_, Qt::UserRole).value<RtpAudioStream*>();
if (!audio_stream)
return;
AudioRouting audio_routing = audio_stream->getAudioRouting();
// Invert muting
if (audio_routing.isMuted()) {
changeAudioRoutingOnItem(ti, AudioRouting(AUDIO_UNMUTED, channel_any));
RTP dissector: Remove unused functions Patch removes (just comments out) functions not used by the code as found by Martin Mathieson.
2021-02-05 09:19:48 +00:00
} else {
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
changeAudioRoutingOnItem(ti, AudioRouting(AUDIO_MUTED, channel_any));
RTP dissector: Remove unused functions Patch removes (just comments out) functions not used by the code as found by Martin Mathieson.
2021-02-05 09:19:48 +00:00
}
}
void RtpPlayerDialog::on_actionAudioRoutingP_triggered()
{
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
changeAudioRouting(AudioRouting(AUDIO_UNMUTED, channel_mono));
RTP dissector: Remove unused functions Patch removes (just comments out) functions not used by the code as found by Martin Mathieson.
2021-02-05 09:19:48 +00:00
}
void RtpPlayerDialog::on_actionAudioRoutingL_triggered()
{
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
changeAudioRouting(AudioRouting(AUDIO_UNMUTED, channel_stereo_left));
RTP dissector: Remove unused functions Patch removes (just comments out) functions not used by the code as found by Martin Mathieson.
2021-02-05 09:19:48 +00:00
}
void RtpPlayerDialog::on_actionAudioRoutingLR_triggered()
{
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
changeAudioRouting(AudioRouting(AUDIO_UNMUTED, channel_stereo_both));
RTP dissector: Remove unused functions Patch removes (just comments out) functions not used by the code as found by Martin Mathieson.
2021-02-05 09:19:48 +00:00
}
void RtpPlayerDialog::on_actionAudioRoutingR_triggered()
{
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
changeAudioRouting(AudioRouting(AUDIO_UNMUTED, channel_stereo_right));
RTP dissector: Remove unused functions Patch removes (just comments out) functions not used by the code as found by Martin Mathieson.
2021-02-05 09:19:48 +00:00
}
RTP Player: Added additional shortcuts Added shortcuts: - Mute/Unmute/Invert Mute - Play/Pause/Stop Removed outdated help in tooltip text. It makes no sense to use it, all actions are available in context menu.
2021-03-11 08:22:15 +00:00
void RtpPlayerDialog::on_actionAudioRoutingMute_triggered()
{
changeAudioRouting(AudioRouting(AUDIO_MUTED, channel_any));
}
void RtpPlayerDialog::on_actionAudioRoutingUnmute_triggered()
{
changeAudioRouting(AudioRouting(AUDIO_UNMUTED, channel_any));
}
void RtpPlayerDialog::on_actionAudioRoutingMuteInvert_triggered()
RTP dissector: Remove unused functions Patch removes (just comments out) functions not used by the code as found by Martin Mathieson.
2021-02-05 09:19:48 +00:00
{
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
QList<QTreeWidgetItem *> items = ui->streamTreeWidget->selectedItems();
for(int i = 0; i<items.count(); i++ ) {
QTreeWidgetItem *ti = items[i];
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
invertAudioMutingOnItem(ti);
RTP dissector: Remove unused functions Patch removes (just comments out) functions not used by the code as found by Martin Mathieson.
2021-02-05 09:19:48 +00:00
}
}
rtp_player_dialog: Route audio for a stream to left/right speaker in RTP player Column 'Play' added to player. Double click on a stream in the column changes audio routing for the stream. When soundcard supports only one channel, there are Mute/Play option. When soundcard supports two or more channels, there are Mute/L/L+R/R options. Muted channel is drawn with dotted line. Change-Id: If120c902195da46f98a1663c589f20c6a1da0ba7 Reviewed-on: https://code.wireshark.org/review/35687 Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-07 21:16:42 +00:00
const QString RtpPlayerDialog::getFormatedTime(double f_time)
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
{
rtp_player: Player is able to set start of audio play by double click Patch adds ability to set start of audio play by double clicking on waveform. Patch fixes unreported issue with placing waveform at incorrect place when switched relative/absolute time mode (check/uncheck Time of Day). Change-Id: Ib8ce24aea870e2443e033afbb6d6e9fbcf222431 Reviewed-on: https://code.wireshark.org/review/35621 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-01 23:38:35 +00:00
QString time_str;
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
rtp_player: Player is able to set start of audio play by double click Patch adds ability to set start of audio play by double clicking on waveform. Patch fixes unreported issue with placing waveform at incorrect place when switched relative/absolute time mode (check/uncheck Time of Day). Change-Id: Ib8ce24aea870e2443e033afbb6d6e9fbcf222431 Reviewed-on: https://code.wireshark.org/review/35621 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-01 23:38:35 +00:00
if (ui->todCheckBox->isChecked()) {
rtp_player_dialog: Route audio for a stream to left/right speaker in RTP player Column 'Play' added to player. Double click on a stream in the column changes audio routing for the stream. When soundcard supports only one channel, there are Mute/Play option. When soundcard supports two or more channels, there are Mute/L/L+R/R options. Muted channel is drawn with dotted line. Change-Id: If120c902195da46f98a1663c589f20c6a1da0ba7 Reviewed-on: https://code.wireshark.org/review/35687 Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-07 21:16:42 +00:00
QDateTime date_time = QDateTime::fromMSecsSinceEpoch(f_time * 1000.0);
rtp_player: Player is able to set start of audio play by double click Patch adds ability to set start of audio play by double clicking on waveform. Patch fixes unreported issue with placing waveform at incorrect place when switched relative/absolute time mode (check/uncheck Time of Day). Change-Id: Ib8ce24aea870e2443e033afbb6d6e9fbcf222431 Reviewed-on: https://code.wireshark.org/review/35621 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-01 23:38:35 +00:00
time_str = date_time.toString("yyyy-MM-dd hh:mm:ss.zzz");
} else {
All RTP dialogs: Added option to specify count of decimal places Preferences are extended with advanced settings: gui.decimal_places1 - 2 gui.decimal_places2 - 4 gui.decimal_places3 - 6 rtp_analysis_dialog, rtp_player_dialog and rtp_stream_dialog uses new settings for formating numbers. Same information in all dialogs uses same settings. It solves request #15481. Note: Other UI dialogs can be adapted later.
2020-12-30 12:28:31 +00:00
time_str = QString::number(f_time, 'f', 6);
rtp_player: Player is able to set start of audio play by double click Patch adds ability to set start of audio play by double clicking on waveform. Patch fixes unreported issue with placing waveform at incorrect place when switched relative/absolute time mode (check/uncheck Time of Day). Change-Id: Ib8ce24aea870e2443e033afbb6d6e9fbcf222431 Reviewed-on: https://code.wireshark.org/review/35621 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-01 23:38:35 +00:00
time_str += " s";
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
}
rtp_player: Player is able to set start of audio play by double click Patch adds ability to set start of audio play by double clicking on waveform. Patch fixes unreported issue with placing waveform at incorrect place when switched relative/absolute time mode (check/uncheck Time of Day). Change-Id: Ib8ce24aea870e2443e033afbb6d6e9fbcf222431 Reviewed-on: https://code.wireshark.org/review/35621 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-01 23:38:35 +00:00
return time_str;
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
}
rtp_player: Player is able to set start of audio play by double click Patch adds ability to set start of audio play by double clicking on waveform. Patch fixes unreported issue with placing waveform at incorrect place when switched relative/absolute time mode (check/uncheck Time of Day). Change-Id: Ib8ce24aea870e2443e033afbb6d6e9fbcf222431 Reviewed-on: https://code.wireshark.org/review/35621 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-01 23:38:35 +00:00
const QString RtpPlayerDialog::getFormatedHoveredTime()
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
{
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
QPoint pos = ui->audioPlot->mapFromGlobal(QCursor::pos());
QTreeWidgetItem *ti = findItemByCoords(pos);
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
if (!ti) return tr("Unknown");
RTP Player: Decoding of audio improved Changes: In nearly all cases decoding match content of capture. The exception is #2270, where timestamps do not match recorded time which causes discrepancy in decoding. Decoding of audio correctly follows different soundcard rates. RTP Player shows first sample rate in each stream in place of rate of playing. Fixed incorrect time axis calculation Fixes #16837 Fixes #4960 Fixes #2270
2021-03-27 08:41:58 +00:00
double ts = ui->audioPlot->xAxis->pixelToCoord(pos.x());
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
rtp_player: Player is able to set start of audio play by double click Patch adds ability to set start of audio play by double clicking on waveform. Patch fixes unreported issue with placing waveform at incorrect place when switched relative/absolute time mode (check/uncheck Time of Day). Change-Id: Ib8ce24aea870e2443e033afbb6d6e9fbcf222431 Reviewed-on: https://code.wireshark.org/review/35621 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-01 23:38:35 +00:00
return getFormatedTime(ts);
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
}
int RtpPlayerDialog::getHoveredPacket()
{
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
QPoint pos = ui->audioPlot->mapFromGlobal(QCursor::pos());
QTreeWidgetItem *ti = findItemByCoords(pos);
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
if (!ti) return 0;
RTP Player: Audio routing information refactored Audio routing information storage refactored to separate mute information and channel routing. It is required for next changes.
2021-02-04 13:12:44 +00:00
RtpAudioStream *audio_stream = ti->data(stream_data_col_, Qt::UserRole).value<RtpAudioStream*>();
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
RTP Player: Decoding of audio improved Changes: In nearly all cases decoding match content of capture. The exception is #2270, where timestamps do not match recorded time which causes discrepancy in decoding. Decoding of audio correctly follows different soundcard rates. RTP Player shows first sample rate in each stream in place of rate of playing. Fixed incorrect time axis calculation Fixes #16837 Fixes #4960 Fixes #2270
2021-03-27 08:41:58 +00:00
double ts = ui->audioPlot->xAxis->pixelToCoord(pos.x());
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
return audio_stream->nearestPacket(ts, !ui->todCheckBox->isChecked());
}
Qt: Make the RTP player output device selectable. Add a combobox for selecting the output device and populate it with our available devices. Let the user know if our output format isn't supported. Ping-Bug: 13105 Change-Id: I299c7d0f191bb66d93896338036000e2c377781f Reviewed-on: https://code.wireshark.org/review/19046 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2016-12-02 23:52:02 +00:00
// Used by RtpAudioStreams to initialize QAudioOutput. We could alternatively
// pass the corresponding QAudioDeviceInfo directly.
Qt: Fixup the currentOutputDeviceName Q_PROPERTY. The CONSTANT attribute indicates that the same value will be returned every time. That isn't the case here so remove it. Change-Id: Ie7451e6aabcb4fa1a6960762d96ad190f32b3d7a Reviewed-on: https://code.wireshark.org/review/19130 Reviewed-by: Gerald Combs <gerald@wireshark.org> Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2016-12-07 17:58:28 +00:00
QString RtpPlayerDialog::currentOutputDeviceName()
Qt: Make the RTP player output device selectable. Add a combobox for selecting the output device and populate it with our available devices. Let the user know if our output format isn't supported. Ping-Bug: 13105 Change-Id: I299c7d0f191bb66d93896338036000e2c377781f Reviewed-on: https://code.wireshark.org/review/19046 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2016-12-02 23:52:02 +00:00
{
return ui->outputDeviceComboBox->currentText();
}
Qt RTP: Add the default device and adjust sample rates. In the RTP player dialog, list the default audio device first, ensure it's selected by default and ensure that the list items are unique. According to http://code.qt.io/cgit/qt/qtmultimedia.git/tree/src/plugins/windowsaudio/qwindowsaudiodeviceinfo.cpp?h=5.9 the default device on Windows uses the special WAVE_MAPPER id, which appears to support various sample rates even when the underlying hardware doesn't. Ensuring the names are unique fixes an issue I'm seeing on a test machine here. When decoding, check to see if our sample rate is supported by our output device and adjust accordingly. Bug: 13906 Change-Id: Iddc0beb2459bfac42276ff29d227c2619b0a8d90 Reviewed-on: https://code.wireshark.org/review/22756 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2017-07-21 18:03:04 +00:00
void RtpPlayerDialog::on_outputDeviceComboBox_currentIndexChanged(const QString &)
{
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
stereo_available_ = isStereoAvailable();
for (int row = 0; row < ui->streamTreeWidget->topLevelItemCount(); row++) {
QTreeWidgetItem *ti = ui->streamTreeWidget->topLevelItem(row);
RtpAudioStream *audio_stream = ti->data(stream_data_col_, Qt::UserRole).value<RtpAudioStream*>();
if (!audio_stream)
continue;
changeAudioRoutingOnItem(ti, audio_stream->getAudioRouting().convert(stereo_available_));
}
Qt RTP: Add the default device and adjust sample rates. In the RTP player dialog, list the default audio device first, ensure it's selected by default and ensure that the list items are unique. According to http://code.qt.io/cgit/qt/qtmultimedia.git/tree/src/plugins/windowsaudio/qwindowsaudiodeviceinfo.cpp?h=5.9 the default device on Windows uses the special WAVE_MAPPER id, which appears to support various sample rates even when the underlying hardware doesn't. Ensuring the names are unique fixes an issue I'm seeing on a test machine here. When decoding, check to see if our sample rate is supported by our output device and adjust accordingly. Bug: 13906 Change-Id: Iddc0beb2459bfac42276ff29d227c2619b0a8d90 Reviewed-on: https://code.wireshark.org/review/22756 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2017-07-21 18:03:04 +00:00
rescanPackets();
}
Add jitter logic to RtpAudioStream. Copy the jitter logic from rtp_player.c to rtp_audio_stream.cpp. This still isn't correct but the RTP player should now be complete enough to start looking at the bug list at the top of rtp_player_dialog.cpp. Disable timing and jitter controls while we're playing while we're here. Fixes bug 11635. Bug: 11635 Change-Id: Ie583ade522702cbe1bbcea4475a535caa1d74fa2 Reviewed-on: https://code.wireshark.org/review/11295 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2015-10-26 15:34:58 +00:00
void RtpPlayerDialog::on_jitterSpinBox_valueChanged(double)
{
rescanPackets();
}
void RtpPlayerDialog::on_timingComboBox_currentIndexChanged(int)
{
rescanPackets();
}
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
void RtpPlayerDialog::on_todCheckBox_toggled(bool)
{
QCPAxis *x_axis = ui->audioPlot->xAxis;
rtp_player: Player is able to set start of audio play by double click Patch adds ability to set start of audio play by double clicking on waveform. Patch fixes unreported issue with placing waveform at incorrect place when switched relative/absolute time mode (check/uncheck Time of Day). Change-Id: Ib8ce24aea870e2443e033afbb6d6e9fbcf222431 Reviewed-on: https://code.wireshark.org/review/35621 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-01 23:38:35 +00:00
double move;
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
// Create plot with new tod settings
createPlot();
// Move view to same place as was shown before the change
rtp_player: Player is able to set start of audio play by double click Patch adds ability to set start of audio play by double clicking on waveform. Patch fixes unreported issue with placing waveform at incorrect place when switched relative/absolute time mode (check/uncheck Time of Day). Change-Id: Ib8ce24aea870e2443e033afbb6d6e9fbcf222431 Reviewed-on: https://code.wireshark.org/review/35621 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-01 23:38:35 +00:00
if (ui->todCheckBox->isChecked()) {
// rel -> abs
// based on abs time of first sample
setStartPlayMarker(first_stream_abs_start_time_ + start_marker_time_ - first_stream_rel_start_time_);
move = first_stream_abs_start_time_ - first_stream_rel_start_time_;
} else {
// abs -> rel
// based on 0s
setStartPlayMarker(first_stream_rel_start_time_ + start_marker_time_);
move = - first_stream_abs_start_time_ + first_stream_rel_start_time_;
}
x_axis->moveRange(move);
drawStartPlayMarker();
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
ui->audioPlot->replot();
}
void RtpPlayerDialog::on_buttonBox_helpRequested()
{
wsApp->helpTopicAction(HELP_TELEPHONY_RTP_PLAYER_DIALOG);
}
rtp_player: Player is able to set start of audio play by double click Patch adds ability to set start of audio play by double clicking on waveform. Patch fixes unreported issue with placing waveform at incorrect place when switched relative/absolute time mode (check/uncheck Time of Day). Change-Id: Ib8ce24aea870e2443e033afbb6d6e9fbcf222431 Reviewed-on: https://code.wireshark.org/review/35621 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-01 23:38:35 +00:00
double RtpPlayerDialog::getStartPlayMarker()
{
double start_pos;
if (ui->todCheckBox->isChecked()) {
start_pos = start_marker_time_ + first_stream_abs_start_time_;
} else {
start_pos = start_marker_time_;
}
return start_pos;
}
void RtpPlayerDialog::drawStartPlayMarker()
{
double pos = getStartPlayMarker();
start_marker_pos_->point1->setCoords(pos, 0.0);
start_marker_pos_->point2->setCoords(pos, 1.0);
updateHintLabel();
}
rtp_player_dialog: Route audio for a stream to left/right speaker in RTP player Column 'Play' added to player. Double click on a stream in the column changes audio routing for the stream. When soundcard supports only one channel, there are Mute/Play option. When soundcard supports two or more channels, there are Mute/L/L+R/R options. Muted channel is drawn with dotted line. Change-Id: If120c902195da46f98a1663c589f20c6a1da0ba7 Reviewed-on: https://code.wireshark.org/review/35687 Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-07 21:16:42 +00:00
void RtpPlayerDialog::setStartPlayMarker(double new_time)
rtp_player: Player is able to set start of audio play by double click Patch adds ability to set start of audio play by double clicking on waveform. Patch fixes unreported issue with placing waveform at incorrect place when switched relative/absolute time mode (check/uncheck Time of Day). Change-Id: Ib8ce24aea870e2443e033afbb6d6e9fbcf222431 Reviewed-on: https://code.wireshark.org/review/35621 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-01 23:38:35 +00:00
{
if (ui->todCheckBox->isChecked()) {
rtp_player_dialog: Route audio for a stream to left/right speaker in RTP player Column 'Play' added to player. Double click on a stream in the column changes audio routing for the stream. When soundcard supports only one channel, there are Mute/Play option. When soundcard supports two or more channels, there are Mute/L/L+R/R options. Muted channel is drawn with dotted line. Change-Id: If120c902195da46f98a1663c589f20c6a1da0ba7 Reviewed-on: https://code.wireshark.org/review/35687 Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-07 21:16:42 +00:00
new_time = qBound(first_stream_abs_start_time_, new_time, first_stream_abs_start_time_ + streams_length_);
rtp_player: Player is able to set start of audio play by double click Patch adds ability to set start of audio play by double clicking on waveform. Patch fixes unreported issue with placing waveform at incorrect place when switched relative/absolute time mode (check/uncheck Time of Day). Change-Id: Ib8ce24aea870e2443e033afbb6d6e9fbcf222431 Reviewed-on: https://code.wireshark.org/review/35621 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-01 23:38:35 +00:00
// start_play_time is relative, we must calculate it
rtp_player_dialog: Route audio for a stream to left/right speaker in RTP player Column 'Play' added to player. Double click on a stream in the column changes audio routing for the stream. When soundcard supports only one channel, there are Mute/Play option. When soundcard supports two or more channels, there are Mute/L/L+R/R options. Muted channel is drawn with dotted line. Change-Id: If120c902195da46f98a1663c589f20c6a1da0ba7 Reviewed-on: https://code.wireshark.org/review/35687 Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-07 21:16:42 +00:00
start_marker_time_ = new_time - first_stream_abs_start_time_;
rtp_player: Player is able to set start of audio play by double click Patch adds ability to set start of audio play by double clicking on waveform. Patch fixes unreported issue with placing waveform at incorrect place when switched relative/absolute time mode (check/uncheck Time of Day). Change-Id: Ib8ce24aea870e2443e033afbb6d6e9fbcf222431 Reviewed-on: https://code.wireshark.org/review/35621 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-01 23:38:35 +00:00
} else {
rtp_player_dialog: Route audio for a stream to left/right speaker in RTP player Column 'Play' added to player. Double click on a stream in the column changes audio routing for the stream. When soundcard supports only one channel, there are Mute/Play option. When soundcard supports two or more channels, there are Mute/L/L+R/R options. Muted channel is drawn with dotted line. Change-Id: If120c902195da46f98a1663c589f20c6a1da0ba7 Reviewed-on: https://code.wireshark.org/review/35687 Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-07 21:16:42 +00:00
new_time = qBound(first_stream_rel_start_time_, new_time, first_stream_rel_start_time_ + streams_length_);
start_marker_time_ = new_time;
rtp_player: Player is able to set start of audio play by double click Patch adds ability to set start of audio play by double clicking on waveform. Patch fixes unreported issue with placing waveform at incorrect place when switched relative/absolute time mode (check/uncheck Time of Day). Change-Id: Ib8ce24aea870e2443e033afbb6d6e9fbcf222431 Reviewed-on: https://code.wireshark.org/review/35621 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-01 23:38:35 +00:00
}
}
void RtpPlayerDialog::updateStartStopTime(rtpstream_info_t *rtpstream, int tli_count)
{
// Calculate start time of first stream and end time of last stream
double stream_rel_start_time = nstime_to_sec(&rtpstream->start_rel_time);
double stream_abs_start_time = nstime_to_sec(&rtpstream->start_abs_time);
double stream_rel_stop_time = nstime_to_sec(&rtpstream->stop_rel_time);
if (tli_count == 0) {
// Take start/stop time for first stream
first_stream_rel_start_time_ = stream_rel_start_time;
first_stream_abs_start_time_ = stream_abs_start_time;
first_stream_rel_stop_time_ = stream_rel_stop_time;
} else {
// Calculate min/max for start/stop time for other streams
first_stream_rel_start_time_ = qMin(first_stream_rel_start_time_, stream_rel_start_time);
first_stream_abs_start_time_ = qMin(first_stream_abs_start_time_, stream_abs_start_time);
first_stream_rel_stop_time_ = qMax(first_stream_rel_stop_time_, stream_rel_stop_time);
}
streams_length_ = first_stream_rel_stop_time_ - first_stream_rel_start_time_;
}
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
void RtpPlayerDialog::formatAudioRouting(QTreeWidgetItem *ti, AudioRouting audio_routing)
rtp_player_dialog: Route audio for a stream to left/right speaker in RTP player Column 'Play' added to player. Double click on a stream in the column changes audio routing for the stream. When soundcard supports only one channel, there are Mute/Play option. When soundcard supports two or more channels, there are Mute/L/L+R/R options. Muted channel is drawn with dotted line. Change-Id: If120c902195da46f98a1663c589f20c6a1da0ba7 Reviewed-on: https://code.wireshark.org/review/35687 Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-07 21:16:42 +00:00
{
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
ti->setText(channel_col_, tr(audio_routing.formatAudioRoutingToString()));
rtp_player_dialog: Route audio for a stream to left/right speaker in RTP player Column 'Play' added to player. Double click on a stream in the column changes audio routing for the stream. When soundcard supports only one channel, there are Mute/Play option. When soundcard supports two or more channels, there are Mute/L/L+R/R options. Muted channel is drawn with dotted line. Change-Id: If120c902195da46f98a1663c589f20c6a1da0ba7 Reviewed-on: https://code.wireshark.org/review/35687 Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-07 21:16:42 +00:00
}
bool RtpPlayerDialog::isStereoAvailable()
{
RTP Player dialog improved Functional changes: Audio routing information is now stored in audio stream and not in table. It is handled by separate utils/rtp_audio_routing.cpp class which is able to convert it between mono/stereo etc. There is new utils/rtp_audio_routing_filter.cpp class which is able to route mono audio stream to any audio channel. Sample file separated from audio stream file - sample file is generated only during recap. So when we need new waveform, we just use existing sample file and no recap required - it is much faster. Audio stream exports just mono audio. Mono audio is then expanded to stereo and correct channel by AudioRoutingFilter during play as required. So when audio routing is changed, no recap and no audio export is required. When audio stream is muted, no audio is produced nor played. Most of signals between RtpPlayerDialog and RtpAudioStream were removed. Start/Pause/Stop is processed in RtpPlayDialog (just for non muted streams). Play possition is not received from every playing stream but from independent silence stream. Added Mute/Unmute function. Optimalization: When audio routing is changed, just graphs are updated. No retap nor audio decoding is required. When TOD is changed, just graphs are updated. No retap nor audio decoding is required.
2021-02-08 21:35:12 +00:00
QAudioDeviceInfo cur_out_device = getCurrentDeviceInfo();
rtp_player_dialog: Route audio for a stream to left/right speaker in RTP player Column 'Play' added to player. Double click on a stream in the column changes audio routing for the stream. When soundcard supports only one channel, there are Mute/Play option. When soundcard supports two or more channels, there are Mute/L/L+R/R options. Muted channel is drawn with dotted line. Change-Id: If120c902195da46f98a1663c589f20c6a1da0ba7 Reviewed-on: https://code.wireshark.org/review/35687 Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-01-07 21:16:42 +00:00
foreach(int count, cur_out_device.supportedChannelCounts()) {
if (count>1) {
return true;
}
}
return false;
}
Rtp player: It is possible to select multiple rows in stream's list and graph Changes: - It is possible to select multiple streams in list and in graph - Select All/None/Invert implemented in list of streams and in graph - Indication of "Selected" stream redesigned in graph - Mouse hovering shows related row/wave - All operations adapted to multiselection
2021-03-09 00:58:15 +00:00
void RtpPlayerDialog::invertSelection()
{
for (int row = 0; row < ui->streamTreeWidget->topLevelItemCount(); row++) {
QTreeWidgetItem *ti = ui->streamTreeWidget->topLevelItem(row);
ti->setSelected(!ti->isSelected());
}
}
void RtpPlayerDialog::on_actionSelectAll_triggered()
{
ui->streamTreeWidget->selectAll();
}
void RtpPlayerDialog::on_actionSelectInvert_triggered()
{
invertSelection();
}
void RtpPlayerDialog::on_actionSelectNone_triggered()
{
ui->streamTreeWidget->clearSelection();
}
RTP Player: Added additional shortcuts Added shortcuts: - Mute/Unmute/Invert Mute - Play/Pause/Stop Removed outdated help in tooltip text. It makes no sense to use it, all actions are available in context menu.
2021-03-11 08:22:15 +00:00
void RtpPlayerDialog::on_actionPlay_triggered()
{
if (ui->playButton->isEnabled()) {
ui->playButton->animateClick();
} else if (ui->pauseButton->isEnabled()) {
ui->pauseButton->animateClick();
}
}
void RtpPlayerDialog::on_actionStop_triggered()
{
if (ui->stopButton->isEnabled()) {
ui->stopButton->animateClick();
}
}
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
#if 0
// This also serves as a title in RtpAudioFrame.
static const QString stream_key_tmpl_ = "%1:%2 " UTF8_RIGHTWARDS_ARROW " %3:%4 0x%5";
RTP: Common functions for allocation/deallocation of rtpstream_info_t Change-Id: I9a0a11d238473a7c57d85547dca0713ed421a500 Reviewed-on: https://code.wireshark.org/review/28417 Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2018-06-21 13:16:03 +00:00
const QString RtpPlayerDialog::streamKey(const rtpstream_info_t *rtpstream)
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
{
const QString stream_key = QString(stream_key_tmpl_)
RTP: Common functions for allocation/deallocation of rtpstream_info_t Change-Id: I9a0a11d238473a7c57d85547dca0713ed421a500 Reviewed-on: https://code.wireshark.org/review/28417 Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2018-06-21 13:16:03 +00:00
.arg(address_to_display_qstring(&rtpstream->src_addr))
.arg(rtpstream->src_port)
.arg(address_to_display_qstring(&rtpstream->dst_addr))
.arg(rtpstream->dst_port)
.arg(rtpstream->ssrc, 0, 16);
Qt: Initial RTP playback. Note the "initial". This is woefully incomplete. See the "to do" lists below and in the code. This differs a bit from the GTK+ version in that you specify one or more streams to be decoded. Instead of showing waveforms in individual widgets, add them all to a single QCustomPlot. This conserves screen real estate and lets us more easily take advantage of the QCP API. It also looks better IMHO. Change a bunch of checks for QtMultimediaWidgets to QtMultimedia. We probably won't use the widgets until we make 5.0 our minimum Qt version and plain old QtMultimedia lets us support Qt 4 more easily (in theory at least). Add resampling code from libspeex. I initially used this to resample each packet to match the preferred rate of our output device, but this resulted in poorer audio quality than expected. Leave it in and use to create visual samples for QCP and to match rates any time the rate changes. The latter is currently untested. Add some debugging macros. Note that both the RTP player and RTP analysis dialogs decode audio data using different code. Note that voip_calls_packet and voip_calls_init_tap appear to be dead code. To do: - Add silence frames where needed. - Implement the jitter buffer. - Implement the playback timing controls. - Tapping / scanning streams might be too slow. Change-Id: I20dd3b66d3df53c9b1f3501262dc01458849f6b4 Bug: 9007 Reviewed-on: https://code.wireshark.org/review/10458 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2014-12-13 00:51:40 +00:00
return stream_key;
}
const QString RtpPlayerDialog::streamKey(const packet_info *pinfo, const struct _rtp_info *rtpinfo)
{
const QString stream_key = QString(stream_key_tmpl_)
.arg(address_to_display_qstring(&pinfo->src))
.arg(pinfo->srcport)
.arg(address_to_display_qstring(&pinfo->dst))
.arg(pinfo->destport)
.arg(rtpinfo->info_sync_src, 0, 16);
return stream_key;
}
#endif
#endif // QT_MULTIMEDIA_LIB