Remove some unused historical files.
Aggressively disable warnings to keep the lemon source
pristine and avoid the maintenance burden for lemon itself.
Lemon has its own lax policy for warnings that doesn't match our
own and they won't accept external patches to remove the
warnings, so just ignore them. Lemon is just executed to generate
code for the Wireshark build and the minor code issues it has
have no influence at runtime.
For lemon generated code we selectively disable some linting
warnings.
Remove patches for lemon and lempar, they are no longer required
with these changes to silence warnings.
When cross-compiling wireshark the lemon tool should be built
using the host machine compiler to be run on the host. Before
cmake this was done via autotools CC_FOR_BUILD but cmake only
supports one compiler toolchain per build and requires some
workarounds like running cmake twice using separately defined
toolchains.
This gets ugly and complicated fast when considering multiple
toolchains, especially for a simple tool like lemon, so just
allow builds to override the C compiler and wipe the cflags.
This way systems like Gentoo/ChromeOS/Yocto with a properly
setup cross-compile environment can just point to the native
BUILD_CC or similar while minimizing complexity.
The COMPILE_OPTIONS source file property was introduced with CMake 3.11,
disable warnings via a target property instead (available since 2.8.12).
Disable some VS Code Analysis warnings. These flags seem to be accepted
since VS2015, so they can be added unconditionally.
Change-Id: Idfbf154caf8c1168f1f871a640a25b816a2cfab8
Fixes: v2.9.0rc0-2274-g7ce9081fdc ("lemon: sync with upstream (2018-09-08)")
Reviewed-on: https://code.wireshark.org/review/30298
Petri-Dish: Peter Wu <peter@lekensteyn.nl>
Tested-by: Petri Dish Buildbot
Reviewed-by: Peter Wu <peter@lekensteyn.nl>
Changes:
- Drop the old basename modification that was present in the Wireshark
version of lemon.c. Use a new option available since 2018-04-20
("Add the -dDIRECTORY command-line option to LEMON.")
- Redo the static analyzer warning fixes, identifying the root causes
and adding assertions instead of hiding code with __clang_analyzer__.
- Ignore compiler warnings instead of adding config.h, _U_, extra const
keywords, unsigned/signed changes, etc.
- Remove lemon.html, it is out-of-date and external links are available.
In order to make future updates easier, document the exact steps that
were followed to create the lemon.c and lempar.c files. Future changes
SHOULD follow the same process.
My process to reach this updated lemon version:
1. Identify previous sync. Found v2.5.2rc0-147-g653af0f6d0 ("lemon: Sync
with latest trunk.") which seems based on sqlite commit 2b3d584ffe.
2. Check successive Wireshark patches. Identified many non-functional
changes to silence compiler warnings and static analyzer issues.
Found one feature (basename) that can be replaced with upstream -d.
3. Write minimal patches and document changes.
Upstream typos and coding style issues (other than trailing whitespace)
were deliberately not fixed to remain as close as possible to upstream.
Change-Id: I606f46dede86e34520f962a9e7163912392aad57
Reviewed-on: https://code.wireshark.org/review/30290
Petri-Dish: Peter Wu <peter@lekensteyn.nl>
Tested-by: Petri Dish Buildbot
Reviewed-by: Anders Broman <a.broman58@gmail.com>
Building without -DENABLE_ASAN=1 but with -fsanitize=address (as is done
by oss-fuzz) still triggers a memleak report. Closer inspection revealed
that two other members were leaked.
Change-Id: I3d022ac50c554891a73b9380b7879c1a60c3c798
Reviewed-on: https://code.wireshark.org/review/30289
Petri-Dish: Peter Wu <peter@lekensteyn.nl>
Tested-by: Petri Dish Buildbot
Reviewed-by: Anders Broman <a.broman58@gmail.com>
The static function handleswitch() is not compiled with __clang_analyzer__
defined, but it's used in OptInit() without any checks. Change to have
an empty implementation for __clang_analyzer__.
Change-Id: I73da2042ec64fdfa6910c02c059d36db491dc538
Reviewed-on: https://code.wireshark.org/review/28059
Petri-Dish: Stig Bjørlykke <stig@bjorlykke.org>
Tested-by: Petri Dish Buildbot
Reviewed-by: Dario Lombardo <lomato@gmail.com>
Reviewed-by: Anders Broman <a.broman58@gmail.com>
It has been replaced by cmake.
Change-Id: I83a5eddb8645dbbf6bca9f026066d2e995d8e87a
Reviewed-on: https://code.wireshark.org/review/26969
Petri-Dish: Dario Lombardo <lomato@gmail.com>
Tested-by: Petri Dish Buildbot
Reviewed-by: Gerald Combs <gerald@wireshark.org>
Reviewed-by: Anders Broman <a.broman58@gmail.com>
Add assert and remove some code from scan builds.
Change-Id: I32747d1a61f183e4c918d9f50ec8337eaef47f0d
Reviewed-on: https://code.wireshark.org/review/26483
Petri-Dish: Dario Lombardo <lomato@gmail.com>
Tested-by: Petri Dish Buildbot
Reviewed-by: Anders Broman <a.broman58@gmail.com>
Except for the one directory that (currently) has "not yet clean" files,
epan/dissectors, we don't need a separate variable to keep track of the
"clean" source files.
In the cases where not all files were in CLEAN_FILES, put them into the
variable used to enable -Werror or its equivalent.
Change-Id: Ic4119861c1d9e381adfe31e9977e1ac71d623f5b
Reviewed-on: https://code.wireshark.org/review/25830
Petri-Dish: Guy Harris <guy@alum.mit.edu>
Tested-by: Petri Dish Buildbot
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Autotools has the very useful feature by design of allowing the user
to override the default build flags (you break it you keep it).
Apparently CMake applies COMPILE_OPTIONS target property after
CMAKE_{C,CXX}_FLAGS so that doesn't work here. Prepend our flags to those
variables instead to make it work then.
Specific target flag overrides can still be added with COMPILER_OPTIONS
(e.g: generated files with -Wno-warning) but this is less effective and
then we're back at the point where this overrides user flags. It's less
of a concern though.
Change-Id: I44761a79be4289238e02d4e781fef0099628817b
Reviewed-on: https://code.wireshark.org/review/23675
Petri-Dish: João Valverde <j@v6e.pt>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: João Valverde <j@v6e.pt>
Reviewed-by: Peter Wu <peter@lekensteyn.nl>
For a sane plugin build environment. Include config.h as the first
header in the .c file instead.
Fix by moving required compiler attribute macros to a new
"ws_attributes.h" API header.
Change-Id: I34f58a927f68c1a0e59686c14d214825149749e1
Reviewed-on: https://code.wireshark.org/review/23400
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: I1fbc11dc433363b659ddba9a77e52ffd4e3968d7
Reviewed-on: https://code.wireshark.org/review/21850
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>
grammar.c:471:6: warning: no previous prototype for ‘DfilterInit’ [-Wmissing-prototypes]
grammar.c:646:6: warning: no previous prototype for ‘DfilterFinalize’ [-Wmissing-prototypes]
dtd_grammar.c:502:6: warning: no previous prototype for ‘DtdParseInit’ [-Wmissing-prototypes]
dtd_grammar.c:637:6: warning: no previous prototype for ‘DtdParseFinalize’ [-Wmissing-prototypes]
Change-Id: I9c43fb4d5ad50992e8e55163333793b20319aa74
Reviewed-on: https://code.wireshark.org/review/21516
Reviewed-by: Michael Mann <mmann78@netscape.net>
Change-Id: I5830581c843af41f48bc9645c2be7cf35fa00391
Reviewed-on: https://code.wireshark.org/review/21463
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>
This reverts commit 96c72df2df.
Generated a lot of warning (-Wmissing-prototypes)
and i try to kept a sync branch on https://github.com/alagoutte/sqlite/tree/wireshark
Change-Id: I42252a6e736f96783a9fa9805b84e525fe11726e
Reviewed-on: https://code.wireshark.org/review/21182
Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com>
This reduces the differences between our lemon.c and SQLite's.
Change-Id: Ie672fa564a43ffe2fae271968accfed2a864bb63
Reviewed-on: https://code.wireshark.org/review/21107
Petri-Dish: Guy Harris <guy@alum.mit.edu>
Reviewed-by: Guy Harris <guy@alum.mit.edu>
That makes the code more like Lemon as distributed.
Note that Lemon, as distributed, at least claims to have a bunch of
stuff "From the file "XXX.{c,h}", which might be the result of lemon.c
being an "amalgamation":
https://www.sqlite.org/amalgamation.html
although, unlike SQLite, the individual bits of the amalgamation don't
appear to be present in the SQLite Fossil repository, so maybe they've
been permanently glued together.
Change-Id: I361d0d16be0744b127110d7d237fdd84e30b6432
Reviewed-on: https://code.wireshark.org/review/21105
Petri-Dish: Guy Harris <guy@alum.mit.edu>
Reviewed-by: Guy Harris <guy@alum.mit.edu>
They accepted some of our changes, but with no space between "struct
XXX" and "*. Whatever.
Change-Id: I6b29462dc8b04ebf0822f8512d82f5f8df575447
Reviewed-on: https://code.wireshark.org/review/21101
Reviewed-by: Guy Harris <guy@alum.mit.edu>
lemon.c:877:10: warning: Access to field 'lhs' results in a dereference of a null pointer (loaded from field 'startRule')
lemon.c:1141:22: warning: Access to field 'lhs' results in a dereference of a null pointer (loaded from field 'startRule')
it is false positve lem.nrule==0 implies lem.rule==NULL
Change-Id: Id63086990762fbf5195bce34a28f25aeb7a246f7
Reviewed-on: https://code.wireshark.org/review/19445
Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com>
Reviewed-by: Peter Wu <peter@lekensteyn.nl>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Michael Mann <mmann78@netscape.net>
lemon: Tue Aug 16 16:46:40 2016
lempar: Tue Dec 6 17:59:05 2016 +0000
a copy of all Wireshark changes are available https://github.com/alagoutte/sqlite/tree/wireshark
Change-Id: I144d0f983e4ac960b5a7a2fd8cd379f6282579f8
Reviewed-on: https://code.wireshark.org/review/15987
Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com>
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>
If ASAN is enabled, use "-fno-sanitize=all" when building lemon. This
keeps us from having to set ASAN_OPTIONS=detect_leaks=0 in the
environment in order to build Wireshark.
Change-Id: I36f6d1a4f913ecabaf188f4c2b59216c8430d81a
Reviewed-on: https://code.wireshark.org/review/18098
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>
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>
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>
for people who don't immediately understand what fseek(fp, 0, 2) does
Change-Id: I297ba6fd718ef0f09c4d0b29bf433262c3c38435
Reviewed-on: https://code.wireshark.org/review/15976
Petri-Dish: Michael Mann <mmann78@netscape.net>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Michael Mann <mmann78@netscape.net>
Manually inline memory_error() in MemoryCheck(), so that static
analyzers know that, if MemoryCheck() sees a null pointer, it exists,
and the null pointer isn't subsequently used.
Use MemoryCheck() instead of the one place where we manually checked for
a null pointer and called memory_error().
Change-Id: Id6b0328cfd17cb14ec9d1e461420896a31573c71
Reviewed-on: https://code.wireshark.org/review/15851
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Always check for malloc() and realloc() failing. Use MemoryCheck() for
most of the checks - it's a bit cleaner and makes the error message the
same for those checks.
Change-Id: I533153c697b37b85adfa0259c1352efece0b0486
Reviewed-on: https://code.wireshark.org/review/15849
Reviewed-by: Guy Harris <guy@alum.mit.edu>
João Valverde