Lua can create a file reader/writer, to open new capture file
formats or write to new ones. To save local state, it can save
things in Lua itself; but since there can be multiple open files
at the same time (for example during a reload), the Lua script
won't know for which file and state its read/write functions are
being invoked for. To remedy this, and also provide a convenient
way to store such state, this commit adds the ability for a Lua
script to store a Lua table in the wtap/wtap_dumper's priv
data member, just like C-code-based reader/writers do.
Change-Id: Ifc9e0d5f0379accee56f2a04b6080238670fec52
Reviewed-on: https://code.wireshark.org/review/766
Reviewed-by: Hadriel Kaplan <hadrielk@yahoo.com>
Reviewed-by: Anders Broman <a.broman58@gmail.com>
There's a relatively new feature in 1.11.3 to select a specific file format
reader, instead of relying on magics or heuristics. If you select a file
reader and open a file, open it, and then click the reload-file button or go
to View->Reload or press the ctrl-R keymap, the file is reloaded but using the
magic/heuristics again instead of the file format reader you previously chose.
Likewise, the Lua relaod() function has the same issue (which is how I found
this problem).
I have tested this change by hand, using a Lua script, but I didn't add it
to the testsuite because I need another change for my test script to work
correctly. (an enhancement rather than a bug fix, which I'll submit separately)
Change-Id: I48c2d9ea443e37fd9d41be43d6b6cd5a866d5b01
Reviewed-on: https://code.wireshark.org/review/764
Reviewed-by: Anders Broman <a.broman58@gmail.com>
wslua_file.c:92:13: error: request for implicit conversion from 'WFILE_T' to 'FILE_T' not permitted in C++ [-Werror=c++-compat]
Change-Id: Iff9cc716333802a3902429a8c68e5f4cdac2ee9e
Reviewed-on: https://code.wireshark.org/review/732
Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com>
Tested-by: Alexis La Goutte <alexis.lagoutte@gmail.com>
This enables a Lua script to implement a brand new capture file format reader/writer, so that for example one could write a script to read from vendor-specific "logs" of packets, and show them as normal packets in wireshark.
Change-Id: Id394edfffa94529f39789844c382b7ab6cc2d814
Reviewed-on: https://code.wireshark.org/review/431
Reviewed-by: Hadriel Kaplan <hadrielk@yahoo.com>
Reviewed-by: Michael Mann <mmann78@netscape.net>
Reviewed-by: Anders Broman <a.broman58@gmail.com>
This fixes a "Argument with 'nonnull' attribute passed null" warning
generated by the clang static analyzer. It's a false positive, but
easy to remedy.
Change-Id: Id737d1ac29765ed26a416c5cd13bedafee478fb6
Reviewed-on: https://code.wireshark.org/review/661
Reviewed-by: Anders Broman <a.broman58@gmail.com>
This adds the ability for Lua scripts to register heuristic dissectors
for any protocol that has registered a heuristic dissector list, such
as UDP, TCP, and ~50 others. The Lua function can also establish a
conversation tied to its Proto dissector, to avoid having to check the
heuristics for the same flow. The example dissector in the testsuite
has also been enhanced to include a heuristic dissector, to verify
the functionality and provide an example implementation.
Change-Id: Ie232602779f43d3418fe8db09c61d5fc0b59597a
Reviewed-on: https://code.wireshark.org/review/576
Reviewed-by: Anders Broman <a.broman58@gmail.com>
Due to the change I made previously for how methods are accessed, if you try
to access one that doesn't exist (for example mistype it or whatever), you get
an internal Lua error about a loop in table get, as opposed to the right error
message about the field not existing.
That's because I had set the class' metatable __index metamethod to point to
the class table, which of course has the metatable with the __index
metamethod, causing a lookup loop. Blech.
Change-Id: I20d3717feadd45f652c2640e1671846184e7082d
Reviewed-on: https://code.wireshark.org/review/593
Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com>
Reviewed-by: Anders Broman <a.broman58@gmail.com>
Should make the licensecheck buildbot happy.
Also add "Public domain MIT/X11 (BSD like)" to the list of permitted licenses,
since it is a combination of two permitted licenses.
Change-Id: Ibc4ead09af89e9225c4e0589a2b7d06dcee6a44e
Reviewed-on: https://code.wireshark.org/review/581
Reviewed-by: Hadriel Kaplan <hadrielk@yahoo.com>
Reviewed-by: Evan Huus <eapache@gmail.com>
Change-Id: I3b87e156ab35e14e3c6e3800ee2058b1a6be57d6
Reviewed-on: https://code.wireshark.org/review/577
Reviewed-by: Martin Kaiser <wireshark@kaiser.cx>
While Lua's built-in pattern support is ok for simple things, many people end
up wanting a real regex engine. Since Wireshark already includes the GLib
Regex library (a wrapper for PCRE), it makes sense to expose that library to
Lua scripts. This has been done using Lrexlib, one of the most popular regex
bindings for Lua. Lrexlib didn't support binding GLib's Regex in particular -
it does for PCRE but GLib is a different API - so I've done that. A fairly
thorough testsuite came along with that, which has been incorporated into the
wireshark wslua testuites as well in this commit.
Change-Id: I05811d1edf7af8d7c9f4f081de6850f31c0717c7
Reviewed-on: https://code.wireshark.org/review/332
Reviewed-by: Anders Broman <a.broman58@gmail.com>
Accessing a pref before it's registered causes a segfault, because prefs_p->next
is not being checked for NULL in wslua_proto.c:Prefs__index().
Change-Id: I270978ddb9238a9e8d2c533a96fc01ee0df385c7
Reviewed-on: https://code.wireshark.org/review/563
Reviewed-by: Evan Huus <eapache@gmail.com>
Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com>
(Using sed : sed -i '/^ \* \$Id\$/,+1 d')
Fix manually some typo (in export_object_dicom.c and crc16-plain.c)
Change-Id: I4c1ae68d1c4afeace8cb195b53c715cf9e1227a8
Reviewed-on: https://code.wireshark.org/review/497
Reviewed-by: Anders Broman <a.broman58@gmail.com>
Use FT_NONE instead of 0 to initialize an enum.
Drop use of lround/llround since they are not available on Windows.
Change-Id: I3961c1921304bafc090c763f0d6de8532f0b3510
Reviewed-on: https://code.wireshark.org/review/425
Reviewed-by: Evan Huus <eapache@gmail.com>
This fixes/addresses all the coverity warnings shown by
the buildbots. (I hope)
Change-Id: Ic2722df97c577d274e3cf3f0cbdca1902edde047
Reviewed-on: https://code.wireshark.org/review/423
Reviewed-by: Gerald Combs <gerald@wireshark.org>
The current API for Lua provides a global function
"all_field_infos()" which returns all the populated field_info nodes
in the current proto_tree.
By default all_field_infos() "works", in the literal sense: it returns
exactly the fields the previous dissectors of the packet have
populated at that instant of time. But of course dissectors don't
populate all the applicable fields most of the time, because of the
TRY_TO_FAKE_THIS_ITEM optimization where they don't fill in things
that aren't needed at the time by a display, color, or tap's dfilter.
So this commit offers a way to force the dissectors to populate
all the applicable field_infos in the tree, by setting the proto_tree
to be visible. Obviously that is going to impact performance, since
it basically bypasses the TRY_TO_FAKE_THIS_ITEM optimization; so the
patch only does this if the Lua script author told it to explicitly,
by adding an argument to Listener.new() and register_postdissector().
Change-Id: I11d3559fbe8c14fbadf1b51415a3701dc1200b7b
Reviewed-on: https://code.wireshark.org/review/286
Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com>
There are some common things people need to do, such as convert to/from hex or get
the raw binary string in a ByteArray/Tvb/TvbRange. These have been added, as well
as some tests for them in the testsuites. Also, functions have been added to allow
a script to get all the available tap types and filter fields, since they are
not exactly what one can see in the Wireshark gui.
Change-Id: I92e5e4eae713bb90d79b0c024eaa4e55b99cc96b
Reviewed-on: https://code.wireshark.org/review/249
Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com>
As discussed in bug 3513 and 9709, one can register more than one new ProtoFields for
the same field name. Of course C-code can do that too, and does a LOT apparently, but
if they're not similar ftypes then things can get scrweed up in display filters.
So this change prevents duplicate field registration of dissimilar ftypes. The
similarity is based on the discussion on the mailing list, and the listing in
README.developer has been updated to refelect that as well.
Also, this change adds a testscript for Proto/ProtoFields.
Change-Id: I43bd323f785245941a21289647332a19adec2a9d
Reviewed-on: https://code.wireshark.org/review/285
Reviewed-by: Evan Huus <eapache@gmail.com>
Several bugs have been introduced due to changing of perl scripts or #define names, such
that things exported into Lua have dissapeared or changed unintentionally. This commit
adds a test suite which compares the Lua global table with the ones from previous
releases (1.8 and 1.10), to verify nothing has gone missing. New items can be added, but
old ones cannot go away. The added script to verify these things, called 'verify_globals.lua',
also has the ability to display what's new - i.e., what was not in the olrder releases.
Lastly, this commit also fixes a bug: MENU_STAT_ENDPOINT became MENU_STAT_ENDPOINT_LIST
due to a change in the make-init-lua.pl perl script in this 1.11 release.
Change-Id: Iba143d1a436e706970635a5f8cc2b317955392bf
Reviewed-on: https://code.wireshark.org/review/284
Reviewed-by: Evan Huus <eapache@gmail.com>
Tested-by: Evan Huus <eapache@gmail.com>
Sort the list to match the .c list while we're at it, to make it easier
to check for errors.
(Why isn't this done with a Makefile.common file?)
Change-Id: I239964d53be9e48bddbd6180aabe118b5cac1fd0
Reviewed-on: https://code.wireshark.org/review/287
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Several bugs have been introduced due to changing of perl scripts or #define names, such
that things exported into Lua have dissapeared or changed unintentionally. This commit
adds a test suite which compares the Lua global table with the ones from previous
releases (1.8 and 1.10), to verify nothing has gone missing. New items can be added, but
old ones cannot go away. The added script to verify these things, called 'verify_globals.lua',
also has the ability to display what's new - i.e., what was not in the olrder releases.
Lastly, this commit also fixes a bug: MENU_STAT_ENDPOINT became MENU_STAT_ENDPOINT_LIST
due to a change in the make-init-lua.pl perl script in this 1.11 release.
Change-Id: Ic46172904256dc535b0fe4543237c07dddb3b9b5
Reviewed-on: https://code.wireshark.org/review/242
Reviewed-by: Evan Huus <eapache@gmail.com>
Over time the various wslua classes/functions have gotten moldy, with different
ways of doing similar things. Some of it can't be changed without breaking
backwards compatibility for Lua scripts, so I didn't do that. But I did what
I could. The biggest change is a refactoring of how accessors/attributes
are handled in the code, so that most of them work the same way using the
same code.
Specific changes made:
* Added null/expired checking macro to class declarations for many classes
* Removed extraneous pointer/expired checking, since checkFoo() does that already
* Fixed "errors" reported by clang static analyzer; they were false positives, but it was easier to get it to stop complaining by changing the code
* Moved internal wslua functions from wslua_utils.c into a new 'wslua_internals.c' file
* Changed Listener/NSTime/Pinfo/Proto to use a common setter/getter accessor/attribute code model, instead of each of them doing their own
* Fixed some API doc mistakes, mostly around attributes that were documented as read-only but were actually read-write
Change-Id: Idddafc5fbd3545ebff29e063acc767e1c743a1a9
Reviewed-on: https://code.wireshark.org/review/271
Reviewed-by: Evan Huus <eapache@gmail.com>
Tested-by: Evan Huus <eapache@gmail.com>
Setting the Pinfo.dst_port to a value actually changes the src_port's value,
due to a bug in wslua_pinfo.c, where both src_port and dst_port attributes use
the PARAM_PORT_SRC for their setter type enum.
Change-Id: I1b84ba8b343ec857d04a2d0809e16f17ba2a43e8
Reviewed-on: https://code.wireshark.org/review/269
Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com>
Tested-by: Alexis La Goutte <alexis.lagoutte@gmail.com>
It isn't picked up by licensecheck at bottom of file.
Change-Id: Ifa8302f90dc44eacd3722efca2471101902457a0
Reviewed-on: https://code.wireshark.org/review/217
Reviewed-by: Evan Huus <eapache@gmail.com>
wslua_struct.c(431) : error C2220: warning treated as error - no 'object' file generated
wslua_struct.c(431) : warning C4244: '=' : conversion from 'lua_Number' to 'size_t', possible loss of data
Change-Id: Id8b7a77eb123232116223077175d243195c8d4fa
Reviewed-on: https://code.wireshark.org/review/213
Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com>
Tested-by: Alexis La Goutte <alexis.lagoutte@gmail.com>
This is based on Roberto Ierusalimschy's struct library, along with additional
options based on Flemming Madsen's patch to the lua-users mailing list, and
some changes I made to support 64-bit integer packing/unpacking. Details
are in the top comments for wslua_struct.c. This also includes a test script.
Change-Id: Ifcd0116ba013d5c760927721c8d6e9f28965534b
Reviewed-on: https://code.wireshark.org/review/98
Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com>
Tested-by: Alexis La Goutte <alexis.lagoutte@gmail.com>
This change adds the ability to pass on to lua scripts loaded from the
command-line (tshark or wireshark) additional arguments supplied by the
command-line. This will help us in our testsuites, but also might be
useful for user-created scripts. The additional arguments are passed in
using the '-X' eXtension switch.
Change-Id: Ib94cdf1ffd194ca84692fee7816665e4ff95efbd
Reviewed-on: https://code.wireshark.org/review/156
Reviewed-by: Evan Huus <eapache@gmail.com>
Tested-by: Evan Huus <eapache@gmail.com>
OK, in all fairness this does a bit more than just fix that bug. It also
adds a 'Int64()' and 'UInt64()' __call metamethods. I generally dislike
using __call metamethods, because they're often unintuitive (for example
the wireshark Field and FielInfo use this in a bizarre fashion). But this
happens to be a perfect use for it, and very natural, imho. Another change
is to make the metatables of classes visible to scripts. There was never
really a good reason not to make them visible, and they have to be visible
to do things like use __call on a plain table... not to mention I need them
to be visible to run test scripts verifying evrything is kosher.
I also updated the test suite to test for the div/mod by zero.
Change-Id: Ia4c594c8f59d4e799090716bd032ba2815df032f
Reviewed-on: https://code.wireshark.org/review/149
Reviewed-by: Evan Huus <eapache@gmail.com>
The fix for bug 9712, which involved changing the make-init-lua.pl perl script
to handle the new C-code define names for filetypes in release 1.11,
unfortunately also stopped matching against the timestamp define names.
Arguably the timestamp ones should never have been matched to begin with, at
least not by the same regex function, because they're not "filetypes". But
they've been matched and exported into the Lua table forever, so we'll break
backward-compat if we don't keep doing it. Ugh.
The good news is I caught this bug using a new test script I wrote which
verifies previous release's Lua stuff didn't disappear. I'll submit that
separately when I get a chance to clean it up, hopefully next week.
Change-Id: Ibf1d1da0dc4ed62cc1bacf345742dccf8ec40e30
Reviewed-on: https://code.wireshark.org/review/143
Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com>
A recent fix (made by me) for bug 9707, in Change-Id:
If4ee1906aa60dd37366cf2ef9bc4168e0ea024b6, made the perl regex grab too much
of the menu name. It changed MENU_STAT_CONVERSATION, MENU_STAT_RESPONSE, and
MENU_ANALYZE_CONVERSATION's key names into their longer C-code names. Ugh.
The fix for this is a bit brittle, but I think it's impractical to avoid it
being brittle, due to needing to support legacy Lua scripts. I put comments
in stat_menu.h to warn of the danger.
Change-Id: I41408e9d4f5b5bd73e2871fccabff81c7cbd242d
Reviewed-on: https://code.wireshark.org/review/140
Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com>
Similar to bug 9725 and ProtoField.new(), the way the VALUESTRING argument is being checked
in the code for ProtoField.bool() ends up making it non-optional. This patch fixes that,
along with some minor API documentation fixes (text).
Change-Id: Iadb9a8ace9c5514fc623d882301fe16b637fe4ce
Reviewed-on: https://code.wireshark.org/review/125
Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com>
Reviewed-by: Evan Huus <eapache@gmail.com>
- packet-bencode.c had the wrong FSF address
- wslua_int64.c had the license as a footer instead of a header
Change-Id: I71204b36a1034af72874d6fe87929c31c9ff03df
Reviewed-on: https://code.wireshark.org/review/123
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Reviewed-by: Hadriel Kaplan <hadrielk@yahoo.com>
Reviewed-by: Evan Huus <eapache@gmail.com>
Tested-by: Evan Huus <eapache@gmail.com>
Using ProtoField.new() is dicey. Many of the optional arguments don't properly check the lua stack - they call lua_isnil() for their index number, instead of lua_gettop() to see the stack size. lua_isnil() may return false in such cases.
Change-Id: I83ca1e5fc34e71ec35899adbedabcee69571b9fe
Reviewed-on: https://code.wireshark.org/review/118
Reviewed-by: Stig Bjørlykke <stig@bjorlykke.org>
Tested-by: Stig Bjørlykke <stig@bjorlykke.org>
In Lua 5.2 the bitop library is missing - it's not getting loaded into
the lua global table as "bit", or anything else for that matter. Lua
5.2 has its own bit-operations library ("bit32") which is there, but
that one's not as good as bitop and would break back/forward
compatibility for lua scripts anyway.
Change-Id: I94b7d45bbeb2f637d1c76b0b5c9d8472eebfcaea
Reviewed-on: https://code.wireshark.org/review/100
Reviewed-by: Evan Huus <eapache@gmail.com>
Tested-by: Evan Huus <eapache@gmail.com>
Reviewed-by: Stig Bjørlykke <stig@bjorlykke.org>
usefulness, working around bug #9162 until Lua 5.3 is released.
The existing Int64 and UInt64 classes provide virtually no
usefullness, other than for creating a string of their value. While
one could then write Lua code to convert the string to Lua numbers and
such, ultimately Lua has no native 64-bit integer support, making such
a task difficult to handle in Lua. This change adds a host of
functions and operators to the existing Int64 (gint64) and UInt64
(guint64) classes, to enable true 64-bit integer support on par with
native Lua numbers.
A test script is also provided, which tests the functions/operators.
Change-Id: I4c5f8f5219b9a88198902283bd32ddf24c346bbe
Reviewed-on: https://code.wireshark.org/review/83
Tested-by: Evan Huus <eapache@gmail.com>
Reviewed-by: Evan Huus <eapache@gmail.com>
Load system init.lua from build-directory/epan/wslua
Set Lua datafile_path to source-directory/epan/wslua
Made dofile() search in source-directory/epan/wslua
Change-Id: I009234eb8193c1ed3260455b245c256c9747930f
This function can be used to check for files before calling dofile(),
which will fail for non-existing files.
Change-Id: Iae7b7ef6d8eb6e0e18f98fee7c740d2a5705eef3
declares the functions must be included, in order to make sure the
declarations match the function signature. Make it so - which means
creating the header file in the first place.
That means luaopen_bit() doesn't need to, and shouldn't be, declared in
wslua.h.
Have make-reg.pl generate a #include of lua_bitop.h. Also fix the "this
is autogenerated" warning.
svn path=/trunk/; revision=54725
Move COL_* enum to <epan/column-utils.h>
XXX Later we can rename epan/column-info.h to column-int.h (or smth like this)
svn path=/trunk/; revision=54352
as an argument, just as the add_packet_field method for a tree does.
Use tvb_get_string_enc() and tvb_get_stringz_enc() rather than
tvb_get_string(), tvb_get_stringz(), tvb_get_unicode_string(), and
tvb_get_unicode_stringz(). Treat "ustring" as meaning "UTF-16-encoded
Unicode string" rather than "UCS-2-encoded subset-of-Unicode string".
svn path=/trunk/; revision=54310
- Allow both string and integer for type and base (improve usability).
- Check valid base and mask for different types (avoid crashes).
- Remove ftypes.PROTOCOL from valid types.
This should fix bug #9549.
svn path=/trunk/; revision=54292
obvious that the returned string is ephemeral, and opens up the original names
in the API for versions that take a wmem pool (and thus can work in any scope).
svn path=/trunk/; revision=54249
knowledge of particular types of plugins. Instead, let particular types
of plugins register with the common plugin code, giving a name and a
routine to recognize that type of plugin.
In particular applications, only process the relevant plugin types.
Add a Makefile.common to the codecs directory.
svn path=/trunk/; revision=53710
subtypes, e.g. Network Monitor version 1 and Network Monitor version 2
are separate "file types", even though they both come from Network
Monitor.
Rename various functions, #defines, and variables appropriately.
svn path=/trunk/; revision=53166
In the process, fix various man page descriptions of the -t flag,
and add support for UTC absolute times in the iousers and iostat TShark
taps.
svn path=/trunk/; revision=53114
When a new Field is created, does as following:
* Check whether that field is registered, by using `proto_registrar_get_byname`. This is current behavior.
* (patched) If not registered, check whether that field is defined in LUA and will be registered. This is performed in `wslua_is_field_available` accessing LUA context.
* If not, an error "a field with this name must exist" occurs.
svn path=/trunk/; revision=52771
explicit, and frees up the "generic" names (like tvb_memdup) for new signatures
that take the appropriate wmem pool.
Majority of the conversion done with sed.
svn path=/trunk/; revision=52164
the same ByteArray twice (and thus calling the GC twice).
This should fix bug 4461.
Changed ByteArray concat to create a new array to be returned.
svn path=/trunk/; revision=51872
registered port numbers in PROTO.prefs_changed.
Now we only need this code in Lua plugins having port numbers prefs:
PROTO.prefs.ports = Pref.range ("Ports", 0, "Port numbers", 65535)
PROTO.prefs_changed()
local udp_table = DissectorTable.get ("udp.port")
udp_table:set (PROTO.prefs.ports, PROTO)
end
svn path=/trunk/; revision=51860
For those with dissectors outside the source tree, please see tools/convert_expert_add_info_format.pl for help with the conversion. Please do not use expert_add_info_format_internal, as it's support time will be very short lived.
svn path=/trunk/; revision=51844
when the preferences for the dissector are changed.
This is not a 100% correct implementation at the time because the
prefs_changed function in all Lua plugins will be called whenever
a preference in a single Lua plugin is changed.
svn path=/trunk/; revision=51818
Original (read from file) comments can be accessed by pkthdr->opt_comment
Keep user comments in seperated BST, add new method for epan session to get it.
svn path=/trunk/; revision=51090
Lua cannot store a 64 bit integer with full precision, which is used
for keys in tables, so this is not a 100% solution. But it will probably
be good enough for value strings, and it is better to have some support
than no support.
svn path=/trunk/; revision=50988
Remove ->prev_cap, for testing purpose also replace ->prev_dis with number of previously displayed frame number.
This patch reduce size of frame_data by 8B (amd64)
This is what (I think) was suggested by Guy in comment 13 (https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=5821#c13)
svn path=/trunk/; revision=50765
that directory since 2001 and reading from that directory was only left in for
backwards compatibility with versions prior to r4702. I think it's now safe
to remove that backwards compatibility.
This eliminates the last argument of get_persconffile_path().
https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=8437
svn path=/trunk/; revision=48797
is supported before trying to open for writing - the attempt to open for
writing will do the check for you. Instead, check for specific errors
if the attempt to open for writing fails, and use somewhat more specific
error messages for certain error codes. (We should perhaps check for
even more error codes in those cases.)
That gets rid of all external calls to wtap_dump_can_write_encap(), so
remove it from wtap.h and make it static.
svn path=/trunk/; revision=48691
Iterate backwards through the linked list of identically-named fields in the lua
bindings since the list is, in fact, created backwards by
proto_register_field_init(). There is some question about whether that is
actually intended, but the rest of the code seems to assume it's normal so we
will too. It was possibly a performance consideration, though that's not
well-documented if so.
Either way, this is the simplest and safest method of fixing the issue with the
lua bindings. See the bug for more analysis.
svn path=/trunk/; revision=48495
made implicit casts explicit and changed a number of fields and function return values from base_display_e to unsigned because the enum was not a comprehensive list of all possible values
https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=8416
svn path=/trunk/; revision=48289
epan/show_exception.c, as it's used outside
epan/dissectors/packet-frame.c. Update their callers to include
<epan/show_exception.h> to get their declaration.
Add a CATCH_NONFATAL_ERRORS macro that catches all exceptions that, if
there's more stuff in the packet to dissect after the dissector call
that threw the exception, doesn't mean you shouldn't go ahead and
dissect that stuff. Use it in all those cases, including ones where
BoundsError was inappropriately being caught (you want those passed up
to the top level, so that the packet is reported as having been cut
short in the capture process).
Add a CATCH_BOUNDS_ERRORS macro that catches all exceptions that
correspond to running past the end of the data for a tvbuff; use it
rather than explicitly catching those exceptions individually, and
rather than just catching all exceptions (the only place that
DissectorError should be caught, for example, is at the top level, so
dissector bugs show up in the protocol tree).
Don't catch and then immediately rethrow exceptions without doing
anything else; just let the exceptions go up to the final catcher.
Use show_exception() to report non-fatal errors, rather than doing it
yourself.
If a dissector is called from Lua, catch all non-fatal errors and use
show_exception() to report them rather than catching only
ReportedBoundsError and adding a proto_malformed item.
Don't catch exceptions when constructing a trailer tvbuff in
packet-ieee8023.c - just construct it after the payload has been
dissected, and let whatever exceptions that throws be handled at the top
level.
Avoid some TRY/CATCH/ENDTRY cases by using checks such as
tvb_bytes_exist() before even looking in the tvbuff.
svn path=/trunk/; revision=47924
Fix the Lua tostring() method on FieldInfo objects such that it always returns
a string, although the string might be '(unknown)', '(none)', or '(n/a)'. This
is more conformant to Lua's API style.
Also create a new 'FieldInfo.display' accessor table member, which Lua
scripts can use instead of tostring() to get what the GUI displays.
From me:
Misc indentation fixes, remove redundant 'return' statement.
svn path=/trunk/; revision=47783
function name isn't the same as the method name; this is used if we're
providing compatibility aliases for method names.
Use WSLUA_CLASS_FNREG() and WSLUA_CLASS_FNREG_ALIAS() for all method
registrations.
Fix the spelling of "prepend", but leave a compatibility alias in place.
svn path=/trunk/; revision=47667
Cast away some implicit 64-bit-to-32-bit conversion errors due to use of
sizeof.
Cast away some implicit 64-bit-to-32-bit conversion errors due to use of
strtol() and strtoul().
Change some data types to avoid those implicit conversion warnings.
When assigning a constant to a float, make sure the constant isn't a
double, by appending "f" to the constant.
Constify a bunch of variables, parameters, and return values to
eliminate warnings due to strings being given const qualifiers. Cast
away those warnings in some cases where an API we don't control forces
us to do so.
Enable a bunch of additional warnings by default. Note why at least
some of the other warnings aren't enabled.
randpkt.c and text2pcap.c are used to build programs, so they don't need
to be in EXTRA_DIST.
If the user specifies --enable-warnings-as-errors, add -Werror *even if
the user specified --enable-extra-gcc-flags; assume they know what
they're doing and are willing to have the compile fail due to the extra
GCC warnings being treated as errors.
svn path=/trunk/; revision=46748
This commit reduces size (from 144B to 128B on AMD64) of frame_data structure.
Part of bug 5821: Reduce per-packet memory requirements.
svn path=/trunk/; revision=45071
Fix leaks
- don't g_strdup a string just to use it in a g_strdup_printf
- clean up properly in error cases in lua bindings
- misc. other missing g_free() calls
- one missing fclose() in the new 80211_utils
https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=7454
svn path=/trunk/; revision=43617
proto_tree_add_item() calls.
Add new "add_packet_field" method to the TreeItem class, taking a
protocol field (*not* a protocol), TvbRange, and encoding value as
arguments.
Add the ENC_ values to init.lua. Make them all hex #defines so
make-init-lua.pl can easily extract them.
Export tvb_unicode_strsize() for use by Lua (and elsewhere as desired).
Note that it handles UTF-16 and UTF-8, and fix the comment to note that
its count of hexadectets *does* include the null terminator (that's what
the code does).
svn path=/trunk/; revision=42621
from makefiles (and thus from the buildbot).
The intention is to be able to tell when a human is running the tool so we
can provide more code-review guidance.
As a starter, enable the "too many proto_tree_add_text() calls" check when
a human is running the tool.
svn path=/trunk/; revision=41943
The Lua API does not have FT_*, it has ftypes.*, so use that in the
documentation. Also, list out each ft so the user knows what the available
options are.
svn path=/trunk/; revision=41035
The Lua API does not have BASE_*, it has base.*, so use that. Also, list out
each base so the user knows what the available options are.
svn path=/trunk/; revision=41024
size_t is sometimes an integer and sometimes a long. To avoid compiler
warnings when formatting it, cast it to a long.
svn path=/trunk/; revision=40682
except stack invalid, and will lead to a crash.
In this case it was when calling a dissector from a table in a Lua script.
svn path=/trunk/; revision=39748
Fix memory errors in Lua dissectors.
- Free Tvb when created from ByteArray.
- Free TvbRange correctly.
- Free string from get_persconffile_path and get_datafile_path.
- Some code cleanup.
svn path=/trunk/; revision=39744
- Adding a INT64/UINT64 field with a value string (currently unsupported).
- Trying to display a signed integer as hexadecimal.
svn path=/trunk/; revision=39562
This works between C and Lua.
In C the pinfo.private_table pointer must be initialized using
g_hash_table_new (g_str_hash, g_str_equal);
In Lua the values are available using pinfo.private.<key>, and the
table is created automatically on first usage. It's possible to use
this datatypes: nil, boolean, number and string, but every value
is converted to string so numbers must be converted using tonumber()
on usage. Boolean is either nil or an empty string.
svn path=/trunk/; revision=39461
Skip files starting with . also on windows, as the home directory
may be mounted from a server with a proper filesystem.
svn path=/trunk/; revision=38737
This object can be used to retreive other absolute and relative time fields,
create and modify nstime_t values and put generated time values in the tree.
Also added ProtoField.absolute_time and ProtoField.relative_time.
svn path=/trunk/; revision=38616
In order to compile the whole project with -DG_DISABLE_DEPRECATED
the mate plugin needs to replace its usage of GMemChunk.
All other places should be clean.
svn path=/trunk/; revision=38392
Added ability to display UTC time or UTC time with date. I liked having the
difference between UTC and local time, not just setting local=UTC.
https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=2629
svn path=/trunk/; revision=37898
Add NULL checks and fix typos in Lua error print-outs
From me:
Use the error message "GUI not available" when using GUI functions in tshark.
svn path=/trunk/; revision=37788
Fix TvbRange:string() so that it returns a string that contains all bytes
of the range, including zeroes (instead of being truncated at first zero).
svn path=/trunk/; revision=37763
with a variant of the patch submitted by Alexander Stein on the -dev list:
Allow FT_FRAMENUMs to be added (as base BASE_NONE). My change from Alexander's
patch is to disallow BASE_NONE for non-framenum integers.
svn path=/trunk/; revision=37239
This patch adds the capability to create BACnet statistics trees.
Find the respective menu items under 'Statistics->BACnet'.
Packets can be sorted by different criteria:
- Src/Dst IP adresses
- Instance ID
- Object Type
- Service
From me:
- Don't use C++/C99-style comments.
- Name variables for tick_stat_node() don't need to be static.
- Change updateBacnetInfoValue() to require 'data' to be ep_ allocated. Change
the couple of calls that did not send in ep_ allocated data to do so.
- Change one or two functions to be static.
- Do not use (memory-unsafe) g_sprintf().
- Use ep_strconcat() instead of leaking memory with g_strconcat().
- Put back one if(tree) that doesn't appear to do any harm.
- Remove variable declarations and #includes from the header file.
svn path=/trunk/; revision=36468
col_set_str simply stores the pointer passed to it, and luaL_checkstring
returns a temporary pointer. Changing col_set_str to col_add_str in
Columns_newindex seems to fix the issue.
svn path=/trunk/; revision=35795
- Define macros for certain CFLAGS in config.nmake iso of having defs in each makefile;
a. -DHAVE_CONFIG_H and -D_U_="" are now part of a macro named STANDARD_CFLAGS;
b. -WX has been replaced by WARNINGS_ARE_ERRORS (defined as -WX in config.nmake)
(This allows disabling "Warnings as Errors" by just changing config.nmake)
c. CVARSDLL definitions (not usage) have been removed from the various makefiles.
XXX: It appears the usage of CVARSDLL can also be removed (not yet done) since:
-DWIN32 and -DNULL=0 do not appear to be needed (any more);
-D_MT and _D_DLL are not needed since /MP causes these definitions.
d. Define a macro WARNINGS_CFLAGS with additional specific compiler (level4) warnings to be enabled.
E.G., 4295: array is too small to include a terminating null character
- config.nmake: reformat some long lines for readability;
- plugins\Makefile.nmake: clean-deps does nothing: remove it (and usage in top-level makefile);
- dissectors/Makefile.nmake: test to enable packet-rrc.obj target needs to include MSVC2010 ...
svn path=/trunk/; revision=35747
keys to have _uint in their names, to match the routines that handle
dissector tables with string keys. (Using _port can confuse people into
thinking they're intended solely for use with TCP/UDP/etc. ports when,
in fact, they work better for things such as Ethernet types, where the
binding of particular values to particular protocols are a lot
stronger.)
svn path=/trunk/; revision=35224
The Lua text string passed to column:set(text) by a dissector will eventually
be reclaimed. This causes the column to display incorrect characters.
svn path=/trunk/; revision=34618
Pull function names to column 0 - except when necessary for autoregistration.
Put function comments before the function.
For readability, put the macros for functions before the function too.
Put class functions before the class.
svn path=/trunk/; revision=34599
Doing a Dumper:close() and then forcing the dumper to be garbage collected will
cause a segfault.
The code attempts to handle the GC that could follow but fails to do it
properly. Assigning Dumper dp = NULL will only have an affect within the scope
of this method... Assumingly this should be a *dp = NULL.
svn path=/trunk/; revision=34597
https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=4788
- Compile the python code directly into epan - don't link it in as
a static lib.
- Call make-init-lua.pl with the top level directory instead of the
current directory. Change make-init-lua.pl accordingly.
svn path=/trunk/; revision=33009
argument indicating whether to include the time zone in the string. If
we're constructing a display filter, don't include the time zone,
otherwise do. Fixes bug 4756.
svn path=/trunk/; revision=32913
WSLUA improvement.
The existing ProtoField.bool looks like the following:
ProtoField.bool(abbr, [name], [desc])
But checking any protocol dissector with bool protofield, it can be seen that
in C code we have more options than thru the WSLUA interface.
E.g.: {&hf_gtp_qos_sig_ind,{"Signalling Indication", "gtp.sig_ind", FT_BOOLEAN,
8, TFS(>p_sig_ind), GTP_EXT_QOS_SIG_IND_MASK, NULL, HFILL}},
So I've extended ProtoField.bool to have more options like this:
ProtoField.bool(abbr, [name], [bitfield], [true_false_string], [mask], [desc])
https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=4700
svn path=/trunk/; revision=32771
- Add checking for linker flags
- Install plugins with the name including the Wireshark version.
This will make it easier to find matching plugin versions if
files get just copied over.
svn path=/trunk/; revision=32231
date as YYYY/DDD, where DDD is a 1-origin day of year. Move the formats
to a "time_fmt.h" file, included by the headers that use it. Have
abs_time_to_str() and abs_time_secs_to_str() take the date format value,
rather than a Boolean "show this as UTC" flag, as an argument. Document
the ABSOLUTE_TIME_ formats a bit better. Use that format in the CCSDS
and VCDU dissectors, rather than having those dissectors do the
formatting themselves.
svn path=/trunk/; revision=32034
* wslua doesnt null-terminate a char array properly after one strncpy call
* packet-smb-sidsnooping (disabled now but someone might enable it again)
has a potential overflow crash (i think?)
svn path=/trunk/; revision=31392
indicating whether the time should be shown as local time or UTC. For
now, always pass FALSE, meaning "show as local time".
Clean up some stuff in the SNMP dissector, use abs_time_secs_to_str()
for times with one-second resolution, and update a comment in various
macros in the WSP dissector, while we're at it.
svn path=/trunk/; revision=31227
We fill out the COL_DSTIDX column by using 'pinfo->dst_idx'. This member is only set by the MDS Header dissector based on 'mdshdr.dstidx'. So remove COL_DSTIDX and migrate to 'mdshdr.dstidx' custom column.
svn path=/trunk/; revision=29795
We fill out the COL_SRCIDX column by using 'pinfo->src_idx'. This member is only set by the MDS Header dissector based on 'mdshdr.srcidx'. So remove COL_SRCIDX and migrate to 'mdshdr.srcidx' custom column.
svn path=/trunk/; revision=29794
We fill out the COL_RXID column by using 'pinfo->rxid'. This member is only set by the Fibre Channel dissector based on 'fc.rx_id'. So remove COL_RXID and migrate to 'fc.rx_id' custom column.
svn path=/trunk/; revision=29793
We fill out the COL_OXID column by using 'pinfo->oxid'. This member is only set by the Fibre Channel dissector based on 'fc.ox_id'. So remove COL_OXID and migrate to 'fc.ox_id' custom column.
svn path=/trunk/; revision=29792
1) The tvb + name (aka. data_source) is only used when the protocol tree is visible
The current implementation of add_new_data_source() doesn't take this into account and simply allocates a data_source regardless. This is what packet_add_new_data_source() tries to rectify.
A couple of dissectors have already been switched over to the new packet_add_new_data_source(). Many are still missing. Help appreciated!
svn path=/trunk/; revision=29427
Use correct modifier for converting gint64 and guint64 to string.
Added __gc for Int64 and UInt64.
Fixed some indents.
svn path=/trunk/; revision=28655
a protocol tree;
the column values.
This includes stats-tree listeners.
Have the routines to build the packet list, and to retap packets, honor
those requirements. This means that cf_retap_packets() no longer needs
an argument to specify whether to construct the column values or not, so
get rid of that argument.
This also means that there's no need for a tap to have a fake filter
to ensure that the protocol tree will be built, so don't set up a fake
"frame" filter.
While we're at it, clean up some cases where "no filter" was represented
as a null string rather than a null pointer.
Have a routine to return an indication of the number of tap listeners
with filters; use that rather than the global num_tap_filters.
Clean up some indentation and some gboolean vs. gint items.
svn path=/trunk/; revision=28645
The value_string_from_table()'s use of the lua stack is not balanced,
if a table is iterated, it pops one more time than it should.
svn path=/trunk/; revision=28633
We can have a case where one Lua dissector calls another Lua dissector,
so we can't mark all objects when only one dissector is completed.
This fixes comment #37 and #39 in bug 2453.
svn path=/trunk/; revision=28325
Update Lua documentation:
- Fixed Pref.enum, Pref.range and Pref.statictext
- Fixed the equivalent comment for pinfo.cols
- Do not have a comment star (*) in the documentation
- Be consistent and start all sentences with a capital letter
svn path=/trunk/; revision=28304
Hadriel Kaplan