The Dissector.list, Listener.list and Field.list functions do not accept
a parameter. Passing one however results in a crash because the first
element is not the expected table, but the first function parameter.
Change-Id: I8483d0aa20716045472f30f5b41a449bed549b05
Reviewed-on: https://code.wireshark.org/review/23209
Petri-Dish: Peter Wu <peter@lekensteyn.nl>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Peter Wu <peter@lekensteyn.nl>
Also, fix error messages reported for lua_load() to speak of
precompilation, not execution (the script isn't *executed* until
lua_pcall() is called).
Change-Id: I7ac9ee6e5df7612f9af141e51958fbfad38a4083
Reviewed-on: https://code.wireshark.org/review/23023
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Support per-packet comments in ERF_TYPE_META through a new Anchor ID
extension header with per-Host unique 48-bit Anchor ID which links an
ERF_TYPE_META record with a packet record. There may be more than one
Anchor ID associated with a packet, where they are grouped by Host ID
extension header in the extension header list. Like other ERF_TYPE_META
existing comments should not be overwritten and instead a new record
generated. See erf_write_anchor_meta_update_phdr() for detailed comments
on the extension header stack required.
As Wireshark only supports one comment currently, use the one one with
the latest metadata generation time (gen_time). Do this for capture
comment too.
Write various wtap metadata in periodic per-second ERF_TYPE_META records
if non-WTAP_ENCAP_ERF or we have an updated capture comment.
Refactor erf_dump to create fake ERF header first then follow common
pseudoheadr and payload write code rather than two separate code paths.
Support an ERF_HOST_ID environment variable to define Wireshark's Host
ID when writing. Defaults to 0 for now.
ERF dissector updates to support Anchor ID extension header with basic
frame linking.
Update ERF_TYPE_META naming and descriptions to official name
(Provenance)
Core changes:
Add has_comment_changed to wtap_pkthdr, TRUE when a packet
opt_comment has unsaved changes by the user.
Add needs_reload to wtap_dumper which forces a full reload of the file
on save, otherwise wireshark gets confused by additional packets being
written.
Change-Id: I0bb04411548c7bcd2d6ed82af689fbeed104546c
Ping-Bug: 12303
Reviewed-on: https://code.wireshark.org/review/21873
Petri-Dish: Anders Broman <a.broman58@gmail.com>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Stephen Donnelly <stephen.donnelly@endace.com>
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Absolute and Relative time fields could not be converted to seconds
without converting to string and parsing to number.
Fixed conversion in generated code that was subject to precision loss
Usage:
f=Field.new("frame.delta_time")
delta=f().value:tonumber()
Change-Id: I6ef91c6238a6c2ed9adf6cae03f8913f0a09332e
Reviewed-on: https://code.wireshark.org/review/22316
Petri-Dish: Stig Bjørlykke <stig@bjorlykke.org>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Make two minor adjustments to allow building on Windows when the source
directory is specified in UNC notation (\\server\volume\directory)
instead of mapping such a directory to a drive letter.
Cmake's add_custom_command() calls "cd <work_dir>" if a working
directory is define as part of the rule. However,
cd \\server\volume\directory
is not allowed.
Modify the two occassions where the working directory is derived from
CMAKE_SOURCE_DIR.
For copying some install files, we can get away with using the absolute
path for each source file to be copied.
The perl script that creates the tap listing for lua does not depend on
a working directory at all. We can simply remove the WORKING_DIRECTORY
parameter.
Change-Id: Iac8e0addc44650692c1263fdca11f68315f50c63
Reviewed-on: https://code.wireshark.org/review/22236
Petri-Dish: Martin Kaiser <wireshark@kaiser.cx>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Peter Wu <peter@lekensteyn.nl>
Use WTAP_MAX_PACKET_SIZE_STANDARD, set to 256KB, for everything except
for D-Bus captures. Use WTAP_MAX_PACKET_SIZE_DBUS, set to 128MB, for
them, because that's the largest possible D-Bus message size. See
https://bugs.freedesktop.org/show_bug.cgi?id=100220
for an example of the problems caused by limiting the snapshot length to
256KB for D-Bus.
Have a snapshot length of 0 in a capture_file structure mean "there is
no snapshot length for the file"; we don't need the has_snap field in
that case, a value of 0 mean "no, we don't have a snapshot length".
In dumpcap, start out with a pipe buffer size of 2KB, and grow it as
necessary. When checking for a too-big packet from a pipe, check
against the appropriate maximum - 128MB for DLT_DBUS, 256KB for
everything else.
Change-Id: Ib2ce7a0cf37b971fbc0318024fd011e18add8b20
Reviewed-on: https://code.wireshark.org/review/21952
Petri-Dish: Guy Harris <guy@alum.mit.edu>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Guy Harris <guy@alum.mit.edu>
This function returns TRUE/FALSE depending whether the specified ProtoField/Dissector
needs to be dissected. By using this function in conjunction with the TreeItem.visible
attribute, Lua dissectors can be significantly sped up by making less C interop calls
which are relatively slow in terms of dissection especially when using sub-protocols
where the dissection of an entire protocol can be skipped.
Added tests for TreeItem:referenced to protofield.lua
Change-Id: I44feacb91a2a5b0e3c28c0ccd8d6b04cccd67261
Reviewed-on: https://code.wireshark.org/review/21387
Petri-Dish: Michael Mann <mmann78@netscape.net>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Peter Wu <peter@lekensteyn.nl>
The only place the time stamp precision is used is in the libpcap code,
where it determines whether to write out microsecond-precision or
nanosecond-precision time stamps; we can determine that by looking at
the type/subtype field, which is also part of that structure, so do
that.
We weren't setting it consistently - we were only setting it in libpcap
and a few other capture file writers, and not in other capture file
writers - and none of the writers other than libpcap used it.
Change-Id: If53779cf4823ca936b8bf3e8a7dbcfea5850e652
Reviewed-on: https://code.wireshark.org/review/21171
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Currently, this is only used to determine whether a protocol tree needs
to be built on the first pass or not - if there are postdissectors that
need fields, it does - but eventually we should be able to use it to
prime the dissection to deliver those fields in cases where we don't
need the *entire* protocol tree (rather than using a hack such as
cooking up a fake tap with a fake filter to do that).
Update MATE and TRANSUM to use it.
Clean up code to check whether we need a protocol tree, and add comments
before that code indicating, in each case, what the criteria are.
The array of postdissectors includes a length, so we don't need to
separately keep track of the number of postdissectors.
Clean up indentation while we're at it.
Change-Id: I71d4025848206d144bc54cc82941089a50e80ab7
Reviewed-on: https://code.wireshark.org/review/21029
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Add a "report a warning message" routine to the "report_err" code in
libwsutil, and rename files and routines appropriately, as they don't
only handle errors any more.
Have a routine read_enabled_and_disabled_protos() that reads all the
files that enable or disable protocols or heuristic dissectors, enables
and disables them based on the contents of those files, and reports
errors itself (as warnings) using the new "report a warning message"
routine. Fix that error reporting to report separately on the disabled
protocols, enabled protocols, and heuristic dissectors files.
Have a routine to set up the enabled and disabled protocols and
heuristic dissectors from the command-line arguments, so it's done the
same way in all programs.
If we try to enable or disable an unknown heuristic dissector via a
command-line argument, report an error.
Update a bunch of comments.
Update the name of disabled_protos_cleanup(), as it cleans up
information for disabled *and* enabled protocols and for heuristic
dissectors.
Support the command-line flags to enable and disable protocols and
heuristic dissectors in tfshark.
Change-Id: I9b8bd29947cccdf6dc34a0540b5509ef941391df
Reviewed-on: https://code.wireshark.org/review/20966
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Add ABSOLUTE_TIME_* defines to the base table in init.lua for use
in ProtoField.absolute_time.
Change-Id: I5c99eafdac97655d71fd4f3374294cd587afaf0a
Reviewed-on: https://code.wireshark.org/review/20543
Petri-Dish: Stig Bjørlykke <stig@bjorlykke.org>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Peter Wu <peter@lekensteyn.nl>
Also add documentation strings to the base values.
Change-Id: I00ac8f154fdd0382106ed27d740e16956520be97
Reviewed-on: https://code.wireshark.org/review/20554
Petri-Dish: Stig Bjørlykke <stig@bjorlykke.org>
Reviewed-by: Peter Wu <peter@lekensteyn.nl>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Stig Bjørlykke <stig@bjorlykke.org>
When using base.UNIT_STRING in a ProtoField the table must be given.
Change-Id: Ie4beb93b5597a97a99939ef2c60a1ee7ece328f2
Reviewed-on: https://code.wireshark.org/review/20542
Petri-Dish: Stig Bjørlykke <stig@bjorlykke.org>
Reviewed-by: Peter Wu <peter@lekensteyn.nl>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Stig Bjørlykke <stig@bjorlykke.org>
Check base value for signed integer before unsigned to avoid a case
where the valid bases for a unsigned integer is presented in a error
message when a signed type is used.
Change-Id: Idfb87597779652e32adceacad220d748afda5e85
Reviewed-on: https://code.wireshark.org/review/20541
Petri-Dish: Stig Bjørlykke <stig@bjorlykke.org>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Peter Wu <peter@lekensteyn.nl>
Remove double start-of-comments and some leading spaces.
Change-Id: I1c5ba49990076cc2d32da2d0a3bb0f54ec31e823
Reviewed-on: https://code.wireshark.org/review/20540
Petri-Dish: Stig Bjørlykke <stig@bjorlykke.org>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Peter Wu <peter@lekensteyn.nl>
Reject ProtoField signed integer types with base other than base.DEC
and base.UNIT_STRING. Update documentation for this.
Bug: 13484
Change-Id: I55a636c89704e96a598f1f185faabd8ad13d96f1
Reviewed-on: https://code.wireshark.org/review/20533
Petri-Dish: Stig Bjørlykke <stig@bjorlykke.org>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Stig Bjørlykke <stig@bjorlykke.org>
WSLUA documents FileHandler:read as returning a boolean or integer.
Ignore strings, do not treat strings as number even if it is possible.
Remove undocumented feature of FileHandler:seek_read that used returned
strings as the frame data. Use FrameInfo:read_data instead. This avoids
interpreting numbers as data which caused surprising behavior when
combining with FileHandler:read.
Documentation has been improved to document the return value, this is
consistent with the documentation in the WSLUA tests.
Change-Id: I31f74ec33ff0e43c90d83c1904af43801886cfa1
Reviewed-on: https://code.wireshark.org/review/19366
Petri-Dish: Anders Broman <a.broman58@gmail.com>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Peter Wu <peter@lekensteyn.nl>
Reviewed-by: Anders Broman <a.broman58@gmail.com>
The default separator is really an empty string, not a single space.
This has been wrongly documented since the beginning.
Change-Id: I5598daec1486ce17cfeeaf5697f9759172db9cf5
Fixes: v1.11.3-rc1-1760-g860747e1e7 ("Adds some Lua helper functions: some commonly used functions, and to help troubleshooting Lua scripts")
Reviewed-on: https://code.wireshark.org/review/20296
Reviewed-by: Michael Mann <mmann78@netscape.net>
This emphasizes that there is no such thing as *the* routine to
construct a subset tvbuff; you need to choose one of
tvb_new_subset_remaining() (if you want a new tvbuff that contains
everything past a certain point in an existing tvbuff),
tvb_new_subset_length() (if you want a subset that contains everything
past a certain point, for some number of bytes, in an existing tvbuff),
and tvb_new_subset_length_caplen() (for all other cases).
Many of the calls to tvb_new_subset_length_caplen() should really be
calling one of the other routines; that's the next step. (This also
makes it easier to find the calls that need fixing.)
Change-Id: Ieb3d676d8cda535451c119487d7cd3b559221f2b
Reviewed-on: https://code.wireshark.org/review/19597
Reviewed-by: Guy Harris <guy@alum.mit.edu>
This is mostly to address memory leaks in range preferences (the biggest
user of range functionality) on shutdown.
Now range preferences must use epan scoped memory when referencing
internal preference structures to keep consistency.
Change-Id: Idc644f59b5b42fa1d46891542b53ff13ea754157
Reviewed-on: https://code.wireshark.org/review/19387
Petri-Dish: Michael Mann <mmann78@netscape.net>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Peter Wu <peter@lekensteyn.nl>
Reviewed-by: Michael Mann <mmann78@netscape.net>
Change-Id: I4d1cf878245b03665207a500fb7593be1435c3d3
Reviewed-on: https://code.wireshark.org/review/19371
Reviewed-by: Peter Wu <peter@lekensteyn.nl>
Petri-Dish: Peter Wu <peter@lekensteyn.nl>
Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Michael Mann <mmann78@netscape.net>
Add handling of STR_ASCII and STR_UNICODE as base types for string
and stringz. Add handling of SEP_DOT, SEP_DASH, SEP_COLON and
SEP_SPACE for bytes and uint_bytes. Add SEP_NONE for completeness.
Change-Id: Ida46c215fee7ec7132ec91ab5dd6cb3de4628920
Reviewed-on: https://code.wireshark.org/review/19337
Petri-Dish: Stig Bjørlykke <stig@bjorlykke.org>
Reviewed-by: Peter Wu <peter@lekensteyn.nl>
Reviewed-by: Stig Bjørlykke <stig@bjorlykke.org>
Fixes Lua on macOS, tested with an out-of-tree build:
WS_BIN_PATH=$PWD/run ../wireshark/test/test.sh -s wslua
Previously programs that were ran from the build directory would load
data files (radius/, diameter/, init.lua) from the source directory.
Then in the case of Lua, files were loaded from the program directory
($BUILDDIR/run/init.lua on Linux) or source directory
(sSOURCEDIR/epan/wslua/console.lua).
On macOS, this does not work for Lua since files are installed into
$BUILDDIR/run/Wireshark.app/Contents/Resources/share/wireshark/init.lua
instead. Since CMake always copies data files (radius, console.lua,
etc.) into the build directory, make get_datafile_dir() return this
"run" directory instead.
Change-Id: If97d2f5686271caf9ad4d4e4fc58e902dc592a98
Reviewed-on: https://code.wireshark.org/review/19330
Reviewed-by: Stig Bjørlykke <stig@bjorlykke.org>
Petri-Dish: Stig Bjørlykke <stig@bjorlykke.org>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Graham Bloice <graham.bloice@trihedral.com>
Reviewed-by: Peter Wu <peter@lekensteyn.nl>
Add support for using unit names in ProtoField integer types by
using base.UNIT_STRING.
Add unit name table argument in ProtoField.float() and
ProtoField.double() (and made backward compatibility).
The use of base.UNIT_STRING is not really the best API for adding
unit names in Lua, but is the simples solution without adding new
arguments to ProtoField.
Change-Id: Ib5d064480cffd970a41db1764440642f6c593bb2
Reviewed-on: https://code.wireshark.org/review/19313
Reviewed-by: Stig Bjørlykke <stig@bjorlykke.org>
Also added a note above field_display_e enum to indicate that this
values are parsed in make-init-lua.pl to build init.lua.
Change-Id: Ibd125684f9a68e1b8116fae0ccbc72147825d75d
Reviewed-on: https://code.wireshark.org/review/19336
Reviewed-by: Stig Bjørlykke <stig@bjorlykke.org>
Because luaL_argerror() does a longjmp all memory free must be done
before calling this.
Also rewrote true_false_string_from_table() to be a bit simpler
and to give argument error when too many strings in the table.
Change-Id: Ied0fa468f1274155c746fe2e086dacf1a8582b08
Reviewed-on: https://code.wireshark.org/review/19325
Petri-Dish: Stig Bjørlykke <stig@bjorlykke.org>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Anders Broman <a.broman58@gmail.com>
FT_CHAR is straightforward to support.
Split the list of "invalid" types into a list of "unsupported" types and
a short list of "invalid" types, containing FT_PCRE (which isn't a valid
type for a field) and "everything else". Add FT_IEEE_11073_SFLOAT and
FT_IEEE_11073_FLOAT to the "unsupported" list.
Flag the whole unsupported list as just "not handled yet".
Change-Id: I62d2d7eead53377e4e601594a035b4395fdbeead
Reviewed-on: https://code.wireshark.org/review/18979
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Change-Id: If1daef88dce166ec92f00a8c6e71bcb220d25e0c
Signed-off-by: Franklin "Snaipe" Mathieu <snaipe@diacritic.io>
Reviewed-on: https://code.wireshark.org/review/18917
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>
In the C API, one can register a heuristic for the same protocol on different
lists by specifying another unique short_name. This is impossible in the
lua API, as the protocol name is used as the short name itself.
This change fixes that by creating an unique shortname composed of the
protocol name and the target list name.
Change-Id: I2c30ce6e4f7a3b38879180c64cf8564f779163b4
Signed-off-by: Franklin "Snaipe" Mathieu <snaipe@diacritic.io>
Reviewed-on: https://code.wireshark.org/review/18711
Petri-Dish: Michael Mann <mmann78@netscape.net>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Peter Wu <peter@lekensteyn.nl>
* Added support for 3-byte integers in :int() and :le_int()
* Added support for 5, 6, and 7-byte integers in :int64() and :le_int64()
Change-Id: If9ab4ea806191bc63effe45a081b9c65693c2367
Signed-off-by: Franklin "Snaipe" Mathieu <snaipe@diacritic.io>
Reviewed-on: https://code.wireshark.org/review/18672
Reviewed-by: Peter Wu <peter@lekensteyn.nl>
Petri-Dish: Peter Wu <peter@lekensteyn.nl>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com>
* Implement Tvb:ipv6()
* Handle FT_IPv6 protofields during insertion in the lua tree
* Implement Address.ipv6(hostname)
Change-Id: I585c392681b3aef02ed8ee956f74051d77fb28d4
Signed-off-by: Franklin "Snaipe" Mathieu <snaipe@diacritic.io>
Reviewed-on: https://code.wireshark.org/review/18442
Reviewed-by: Peter Wu <peter@lekensteyn.nl>
Petri-Dish: Peter Wu <peter@lekensteyn.nl>
Tested-by: Peter Wu <peter@lekensteyn.nl>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Michael Mann <mmann78@netscape.net>
Previously the metatables for classes were the same for the class and
its instances. This results in issues like calling __gc on the class
table on exit.
Make it possible to declare separate class methods (functions) and
instance methods. Observe that all attributes apply to the instances
only, so make these just available on the instance.
The attribute/methods lookup method (via __index/__newindex) have been
rewritten to use upvalues, removing the technical need for the
properties __getters/__setters/__methods. The "lua globals" test still
checks for these, but it could be removed in the future.
To fix bug 12968, the __gc method is removed from the class method.
Future patches should remove the WSLUA_REGISTER_CLASS,
WSLUA_REGISTER_META and WSLUA_REGISTER_ATTRIBUTES macros completely and
create split class functions/methods (such that __call for an instance
cannot accidentally be invoked on the class).
Removed duplicate "fragmented" property from Pinfo (which triggered an
error) and replaced exit() by g_error() for debugger friendliness.
Remove lua_shiftstring since checkstring always returns non-NULL.
Bug: 12968
Change-Id: I57f8a93d08bb84c79b0e94cf2c82d8402fc16646
Reviewed-on: https://code.wireshark.org/review/18026
Petri-Dish: Peter Wu <peter@lekensteyn.nl>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Balint Reczey <balint@balintreczey.hu>
Internal change, this allows use of negative indices for
lua_rawgetfield/lua_rawsetfield convenience functions and is closer to
lua_getfield/lua_setfield semantics in terms of indices.
Add lua_absindex for compatibility with Lua 5.1 (do not bother exporting
it yet since it is only used in this file, also do not rename it since
it is the same functionality with no Wireshark-specific changes.)
Change-Id: I322954ef461164ec514ed9f034ae2691775e67da
Reviewed-on: https://code.wireshark.org/review/18025
Petri-Dish: Peter Wu <peter@lekensteyn.nl>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Anders Broman <a.broman58@gmail.com>
Improve example with better formatting, clarification comments and more
common variable names.
Extend make-wsluarm.pl to support arguments containing underscores.
Fixes the description of dissect_tcp_pdus.
Change TvbRange.tvb(tvb) into tvbrange:tvb() and ByteArray.tvb(name)
into bytearray:tvb(name), these are really instance methods.
Change-Id: I1e20ef46195dc6c06f9ac790d3432db283d21a5e
Reviewed-on: https://code.wireshark.org/review/18226
Reviewed-by: Anders Broman <a.broman58@gmail.com>
ti->item is NULL for the root item, prevent a NULL-deref by Lua script.
Bug: 13017
Change-Id: I5e7f71d014a01bab615288df76509b3ef97b1bf6
Reviewed-on: https://code.wireshark.org/review/18247
Petri-Dish: Peter Wu <peter@lekensteyn.nl>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Anders Broman <a.broman58@gmail.com>
Don't set `lua_tvb` (or any of the other global variables) to NULL after a
Lua dissector is called: it's possible that the caller is also a Lua dissector
which may want/need that (global) variable to still be set (to the value it
had before the sub-dissector was called).
This fixes the problem reported in:
https://ask.wireshark.org/questions/56110/lua-error-tvbs-can-only-be-created-and-used-in-dissectors
Making these variables not be a globals (as suggested at the top of
init_wslua.c) might be a better solution--for another day.
Change-Id: I14fb8ec35b62abeda3f3471a323b88c80537a06e
Reviewed-on: https://code.wireshark.org/review/18095
Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Peter Wu <peter@lekensteyn.nl>
Tested-by: Peter Wu <peter@lekensteyn.nl>
Lua 5.2.4 built with -DLUA_USE_APICHECK detected a stack issue:
tshark: lapi.c:175: lua_settop: Assertion `(-(idx+1) <= (L->top - (func + 1))) && "invalid new top"' failed.
Function File_read always assumes that File_read_line pushes a value on
the stack (which clearly did not happen). On read failure, it would then
pop the stack (tripping the assertion) to push nil.
The other user (File_lines) is also affected by this change, but the Lua
5.2.4 documentation says that it should also return nil on EOF, fitting
this implementation.
Change-Id: I9cc8a5319523b2b56f4ae4735bbdbc1196387386
Reviewed-on: https://code.wireshark.org/review/18016
Petri-Dish: Peter Wu <peter@lekensteyn.nl>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Peter Wu <peter@lekensteyn.nl>
Instead of checking for the boolean "FALSE", just set an empty string.
This avoids the need to check for WERROR_COMMON_FLAGS before using it.
The transformation is the same for all files, remove
"if (WERROR_COMMON_FLAGS)" and "endif()", reindent and add quotes (since
we have a string here and not a list).
Modelines have been added where missing.
Change-Id: I0ab05ae507c51fa77336d49a99a226399cc81b92
Reviewed-on: https://code.wireshark.org/review/17997
Petri-Dish: Peter Wu <peter@lekensteyn.nl>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: João Valverde <j@v6e.pt>
Reviewed-by: Dario Lombardo <lomato@gmail.com>
Tested-by: Dario Lombardo <lomato@gmail.com>
Have all dissector tables have a "supports Decode As" flag, which
defaults to FALSE, and which is set to TRUE if a register_decode_as()
refers to it.
When adding a dissector to a dissector table with a given key, only add
it for Decode As if the dissector table supports it.
For non-FT_STRING dissector tables, always check for multiple entries
for the same protocol with different dissectors, and report an error if
we found them.
This means there's no need for the creator of a dissector table to
specify whether duplicates of that sort should be allowed - we always do
the check when registering something for "Decode As" (in a non-FT_STRING
dissector table), and just don't bother registering anything for "Decode
As" if the dissector table doesn't support "Decode As", so there's no
check done for those dissector tables.
Change-Id: I4a1fdea3bddc2af27a65cfbca23edc99b26c0eed
Reviewed-on: https://code.wireshark.org/review/17402
Petri-Dish: Guy Harris <guy@alum.mit.edu>
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Setting our compiler warning flags in CMAKE_C_FLAGS does not allow
using different flags per target.
Allow for that possibility by setting the internal WS_WARNINGS_{C,CXX}_FLAGS
and using the COMPILE_OPTIONS property to set them.
This change is just setting mechanism and there should be no difference
in generated warnings.
The check_X_compiler_flag cmake test is changed to test each flag individually.
We need a list, not a space separated string, and the aggregate test is not
significant.
Change-Id: I59fc5cd7e130c7a5e001c598e3df3e13f83a6a25
Reviewed-on: https://code.wireshark.org/review/17150
Petri-Dish: João Valverde <j@v6e.pt>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: João Valverde <j@v6e.pt>
Change-Id: Ibc43b1976d5827e8c40252a5200852fbcd00b70c
Reviewed-on: https://code.wireshark.org/review/16763
Petri-Dish: João Valverde <j@v6e.pt>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Michael Mann <mmann78@netscape.net>
Reviewed-by: João Valverde <j@v6e.pt>
The g_warning calls seem legitimate, so "hide" them from checkAPIs.pl.
Change-Id: I6d25b08e22aeeb0244e07836385f2b67d6261546
Reviewed-on: https://code.wireshark.org/review/16703
Reviewed-by: Michael Mann <mmann78@netscape.net>
Many of the complaints from checkAPI.pl for use of printf are when its embedded
in an #ifdef and checkAPI isn't smart enough to figure that out.
The other (non-ifdef) use is dumping internal structures (which is a type of
debug functionality)
Add a "ws_debug_printf" macro for printf to pacify the warnings.
Change-Id: I63610e1adbbaf2feffb4ec9d4f817247d833f7fd
Reviewed-on: https://code.wireshark.org/review/16623
Reviewed-by: Michael Mann <mmann78@netscape.net>
Petri-Dish: Michael Mann <mmann78@netscape.net>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Anders Broman <a.broman58@gmail.com>
A string option, if present, always has a value; it might be a null
*string*, but you won't get a null pointer (if the option isn't present,
it simply isn't present).
Fix some comments while we're at it.
Change-Id: I9c1420f56998a7d04de5c5cc2e92631b181f303a
Reviewed-on: https://code.wireshark.org/review/16564
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Peter Wu