This reuses existing functionality and thereby making FindAsn2Wrs.cmake obsolete, resulting in a reduction of the cmake module maintaince load
Change-Id: Ic6ae53e65236246c0cbe86f8c2f066a2d034e0b9
Reviewed-on: https://code.wireshark.org/review/5805
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>
Add make-dissector-reg.py to the Debian packages.
Making it possible to generate out-of-source wireshark plugins.
Change-Id: I0bbe5b46205d39e229d31812341540b26a7336d6
Reviewed-on: https://code.wireshark.org/review/5802
Reviewed-by: Anders Broman <a.broman58@gmail.com>
A custom target is defined to be always invalid, so asn2wrs.py is always invoked.
A custom command is able to convert input file(s) into output file(s), the actual command is only invoked in the provided dependencies are changed.
As an advantage the dependencies are correct, so cmake is able to determine when the output file is required, so it can be build in time.
Change-Id: I84be5d408cfc87cf88d67b6073bea439febbe712
Reviewed-on: https://code.wireshark.org/review/5255
Reviewed-by: Anders Broman <a.broman58@gmail.com>
For out of source builds, the output directory should not point to wireshark source locations.
By default it does point to the wireshark source directory (to stay compatible with the build in dissectors).
When the A2W_OUTPUT_DIR variable is set to "_EMPTY_" the output lcoation argument (-O) is not invoked.
Use A2W_FLAGS to set an output location for out of source builds (or use A2W_OUTPUT_DIR, which has the same result)
Change-Id: I48b7486fcda982f0dce57cde3beacb99f946abd9
Reviewed-on: https://code.wireshark.org/review/5254
Reviewed-by: Anders Broman <a.broman58@gmail.com>
A first step to make it possible to reuse the UseAsn2Wrs for out of source builds.
Change-Id: Iefb34f943e06db3d6afed55ab69d3f26aa9d633d
Reviewed-on: https://code.wireshark.org/review/5253
Reviewed-by: Anders Broman <a.broman58@gmail.com>
This variable can be used to obtain the wireshark library(ies) (from the correct location).
Change-Id: I90cd5b098eb746ac573a2c611113287b06c3b82b
Reviewed-on: https://code.wireshark.org/review/5251
Petri-Dish: Michael Mann <mmann78@netscape.net>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Michael Mann <mmann78@netscape.net>
These files make it possible to use the cmake command find_package(Wirehark) to obtain information about the wireshark installation.
Change-Id: I5af7c4e7b53b99cd473e04905a92bac267cd9b83
Reviewed-on: https://code.wireshark.org/review/5235
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>
Add convenience targets for generating the release notes and the NEWS
file. Make sure we don't run multiple instances of a2x + AsciiDoc at the
same time.
Add the docbook directory to the build by default unless we're running
Windows. Explain why we don't yet build docs on Windows. Make each
docbook makefile target optional.
Split the ENABLE_GUIDES option into ENABLE_HTML_GUIDES and
ENABLE_PDF_GUIDES. Add a default "all_guides" target if either is on.
Remove the Debian patch that hacked around the PDF requirement.
Copy ws.css to the docbook build directory. Don't build PDF release
notes. I'm not sure we ever used them and I don't want to install Java
and FOP just to make a release.
Change-Id: Ia2f710000c17f9e0b4b514fd373d9a5902889553
Reviewed-on: https://code.wireshark.org/review/5712
Petri-Dish: Gerald Combs <gerald@wireshark.org>
Reviewed-by: Gerald Combs <gerald@wireshark.org>
Don't cache the LUA_LIBRARIES variable. This matches the behavior of the
other library modules and fixes a compilation problem on my machine
where /usr/local/lib/liblua.dylib wasn't showing up in the various
build.make and link.txt files.
Change-Id: Ib75ef303f2e67b266a246621718d0ea2ab885dca
Reviewed-on: https://code.wireshark.org/review/5603
Reviewed-by: Gerald Combs <gerald@wireshark.org>
Avoide rebuilding the guides each and every time.
Don't print the single file guides to stdout.
Change-Id: Ie94fc4b24676e9abc7258c4ea1c7fd3049a96fb6
Reviewed-on: https://code.wireshark.org/review/5367
Reviewed-by: Jörg Mayer <jmayer@loplof.de>
Have FindWSWinLibs check for WIRESHARK_LIB_DIR before WIRESHARK_BASE_DIR
to match the behavior of config.nmake and the Windows buildbots, which
set the former but not the latter.
It looks like the latest GLib DLL contains GThread. Adjust
FindGTHREAD2.cmake accordingly.
Change-Id: I0232b1819eca9ba2c39ae6834daddf65af979d7d
Reviewed-on: https://code.wireshark.org/review/4887
Reviewed-by: Gerald Combs <gerald@wireshark.org>
Current FindGTK2.cmake as included with Wireshark cannot do a 32-bit
build even after setting CMAKE_LIBRARY_PATH=/usr/lib32. After an update
to upstream FindGTK2 (v3.0.1-1824-g0b12815), things build again.
Compared to upstream FindGTK2, FindWSWinLibs and related hints have been
added (note that "${GTK2_HINTS}/lib" is not a typo, it is needed as some
headers are includes in locations like lib/gtk-2.0/include/). The
include paths have become relative again as there are no files like
FindPackageHandleStandardArgs in the Wireshark source tree. Diff between
upstream FindGTK2 and the version in this patch:
diff --git a/cmake/modules/FindGTK2.cmake b/cmake/modules/FindGTK2.cmake
index 72bb8eb..9b4f989 100644
--- a/cmake/modules/FindGTK2.cmake
+++ b/cmake/modules/FindGTK2.cmake
@@ -167,2 +167,5 @@
+include( FindWSWinLibs )
+FindWSWinLibs( "gtk2" "GTK2_HINTS" )
+
#=============================================================
@@ -176,4 +179,4 @@
-include(${CMAKE_CURRENT_LIST_DIR}/SelectLibraryConfigurations.cmake)
-include(${CMAKE_CURRENT_LIST_DIR}/CMakeParseArguments.cmake)
+include(SelectLibraryConfigurations)
+include(CMakeParseArguments)
@@ -289,2 +292,5 @@ function(_GTK2_FIND_INCLUDE_DIR _var _hdr)
${_suffixes}
+ HINTS
+ "${GTK2_HINTS}/include"
+ "${GTK2_HINTS}/lib"
)
@@ -395,2 +401,4 @@ function(_GTK2_FIND_LIBRARY _var _lib _expand_vc _append_version)
[HKEY_LOCAL_MACHINE\\SOFTWARE\\gtkmm\\2.4;Path]/lib
+ HINTS
+ "${GTK2_HINTS}/lib"
)
@@ -409,2 +417,4 @@ function(_GTK2_FIND_LIBRARY _var _lib _expand_vc _append_version)
[HKEY_LOCAL_MACHINE\\SOFTWARE\\gtkmm\\2.4;Path]/lib
+ HINTS
+ "${GTK2_HINTS}/lib"
)
@@ -829,3 +839,3 @@ set(_GTK2_did_we_find_everything true) # This gets set to GTK2_FOUND
-include(${CMAKE_CURRENT_LIST_DIR}/FindPackageHandleStandardArgs.cmake)
+include(FindPackageHandleStandardArgs)
Change-Id: I02103409a79ea30d1bf7cc0dfb43a9e41f8a4db3
Reviewed-on: https://code.wireshark.org/review/4293
Reviewed-by: Graham Bloice <graham.bloice@trihedral.com>
Reviewed-by: Anders Broman <a.broman58@gmail.com>
Changes:
* Fix glib2 search path.
* Add pkg-config support to CAP, GEOIP, GNUTLS, LUA, PortAudio (API
19), zlib, kerberos.
* Add pkg-config support to libnl3, libnl2 and libnl1 (but tested only
with libnl3).
This makes it easier to do 32-bit builds on 64-bit hosts by just setting
`PKG_CONFIG_LIBDIR`. Due to how HINTS work, it is still fragile though:
missing 32-bit libraries will cause a fallback to 64-bit libraries.
A future patch could check for `<PREFIX>_FOUND` and remove the manual
`find_path` and `find_library` hackery since the paths are already
known.
Change-Id: Ieb4fb74695c96afb1a4c70168e84abb1fa4612c6
Reviewed-on: https://code.wireshark.org/review/4292
Reviewed-by: Anders Broman <a.broman58@gmail.com>
If we have pcap_set_tstamp_precision(), use it to request nanosecond
time stamp resolution *if* we're writing a pcap-ng file; any code that
reads those files and can't handle nanosecond time stamp resolution is
broken and needs to be fixed.
If we're writing a pcap file, don't ask for nanosecond resolution time
stamps, as that requires a different magic number for pcap files, and
not all code that reads pcap files can handle that. (Unlike pcap-ng,
where the ability to have non-microsecond time stamp resolution was
present from Day One, it's a relatively recent addition to pcap.) We
could add a command-line option/GUI option for that, like the option
recent versions of tcpdump have, if it matters.
Change-Id: I8fa464eb929feecb9a70be70712502c9f0cc5270
Reviewed-on: https://code.wireshark.org/review/4355
Reviewed-by: Guy Harris <guy@alum.mit.edu>
If it is used, there is a modified file in git.
Fix this by only including the file if it exists.
Other changes:
- Rename the existing Custom files to CMakeListsCustom.txt.example.
- Move the plugins custom file to the top level (same level as its
including parent).
- Optionally allow a list of custom includes instead of the default one.
Change-Id: I8960eac6222f741c045055d43d1d5a2d4979caf6
Reviewed-on: https://code.wireshark.org/review/4163
Reviewed-by: Jörg Mayer <jmayer@loplof.de>
because of different timezones.
Change-Id: I93809447db29c2cc5f848edb438ee16372b57453
Reviewed-on: https://code.wireshark.org/review/3824
Reviewed-by: Jörg Mayer <jmayer@loplof.de>
Use a common set of SVG files for AsciiDoc / DocBook admonition
graphics. Put them in a common directory. According to
http://caniuse.com/svg all common browsers have had SVG support for
a while now.
The graphics themselves were created with Inkscape. If you would like
to refine them further you are more than welcome.
Use variables to assemble xsltproc commands in Autotools and Nmake
while we're here.
Try to update Debian rules to reflect ga92c3fb.
Change-Id: If82647af27a60117c517125dff0aca81c033be72
Reviewed-on: https://code.wireshark.org/review/3206
Petri-Dish: Gerald Combs <gerald@wireshark.org>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Gerald Combs <gerald@wireshark.org>
Create graphics directories and copy files into them similar
to Autotools and Nmake. Adjust some xsltproc arguments. Fix the
--asciidoc-opts flag.
Note: Admon graphics are broken pending change 3206.
Change-Id: I94d498de36150a7cb4ffd080581523300b222bd4
Reviewed-on: https://code.wireshark.org/review/3805
Reviewed-by: Gerald Combs <gerald@wireshark.org>
Add the "asciidoc.conf" compatibility configuration file from
Asciidoctor. Although we don't use Asciidoctor it gives us macros that
keep us from losing some useful DocBook elements. Update various CMake
files to support multiple AsciiDoc configuration files.
Leave most of the content intact for now. Hopefully the other chapters
aren't as laden with markup.
Change-Id: Id69757342b86abb2b3130cb61e90f5695a26ea8a
Reviewed-on: https://code.wireshark.org/review/3680
Petri-Dish: Gerald Combs <gerald@wireshark.org>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Gerald Combs <gerald@wireshark.org>
other liberaries. Also: After set( variable 0 ) the variable is defined.
Change-Id: Id26a771875b7643c19f173ce77bb29a4059a1bd0
Reviewed-on: https://code.wireshark.org/review/3793
Reviewed-by: Jörg Mayer <jmayer@loplof.de>
Change-Id: If55498167d5e357258841f2194962749c0dfc81e
Reviewed-on: https://code.wireshark.org/review/3787
Reviewed-by: Graham Bloice <graham.bloice@trihedral.com>
Petri-Dish: Graham Bloice <graham.bloice@trihedral.com>
Reviewed-by: Roland Knall <rknall@gmail.com>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
The shell script is not used in the CMake build system.
Change-Id: I97d3d750f2521c626594200da8fc9a4453414576
Reviewed-on: https://code.wireshark.org/review/3530
Reviewed-by: Evan Huus <eapache@gmail.com>
Reviewed-by: Anders Broman <a.broman58@gmail.com>
Look for wpcap.lib in WpdPack/Lib/x64 when our target platform is Win64.
Change-Id: I9a1bac22106bcb6a1f155ce83c02e344e6ec0d55
Reviewed-on: https://code.wireshark.org/review/3331
Reviewed-by: Gerald Combs <gerald@wireshark.org>
Note that the cache file (CMakeCache.txt) must be cleared when the
include or library paths change, otherwise the cached result of
CHECK_FUNCTION_EXISTS will be re-used.
Change-Id: Ia18ed4f1b6b162499aeae6e66f612a4945390c2a
Reviewed-on: https://code.wireshark.org/review/3245
Reviewed-by: Jörg Mayer <jmayer@loplof.de>
I have ***NO*** idea why this makes a difference, but, without this
change, APPLE_CORE_FOUNDATION_LIBRARY is apparently *not* set correctly
for wsutil/CMakeLists.txt, and, with this change, it is. I guess
there's something magic involved here with "global" CMake variables or
something crazy such as that.
Change-Id: I7a0046b9c249568cd666720838104f48e854e203
Reviewed-on: https://code.wireshark.org/review/2612
Reviewed-by: Guy Harris <guy@alum.mit.edu>
For -f and -m flags that don't apply to the target architecture(?),
Clang doesn't fail but does print a warning "argument unused during
compilation: '-{flag}'". Catch that and treat it as a failure
indication, so we don't use that flag.
Change-Id: I65948ada41fa44fb62f9a2b8b320f137a5902b8f
Reviewed-on: https://code.wireshark.org/review/2399
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Protecting HAVE_PCAP_CREATE was not enough for the moc run in qt,
maybe it doesn't have _WIN32 set? Don't even try to detect this
function on WIN32 for now.
Change-Id: I0d8a8b5b110cec164f86fe11f26a7add558eee1b
Reviewed-on: https://code.wireshark.org/review/2370
Reviewed-by: Jörg Mayer <jmayer@loplof.de>
of finding the package don't allow the first method to fail terminally.
Todo: Fix this in all other places as well.
Change-Id: I5a343fac33f6a5d6e50ff353d739459b2e36711a
Reviewed-on: https://code.wireshark.org/review/2300
Reviewed-by: Jörg Mayer <jmayer@loplof.de>
Fix detection of Homebrew on APPLE
Change-Id: I96506bb57d4772c5c90b1117c37e8350cda376ee
Reviewed-on: https://code.wireshark.org/review/2257
Reviewed-by: Jörg Mayer <jmayer@loplof.de>
Not all system have HtmlViewer, but we support that case
by inform user about file/website, so we "have" HtmlViewer.
Change-Id: I46e16a86b5c7f9dd47e1d1ded9d10fd4f565660f
Reviewed-on: https://code.wireshark.org/review/1875
Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com>
(Using sed : sed -i '/^\# \$Id\$/,+1 d') (start with dash)
Change-Id: Ia4b5a6c2302f6a531f6a86c1ec3a2f8205c8c2dd
Reviewed-on: https://code.wireshark.org/review/881
Reviewed-by: Anders Broman <a.broman58@gmail.com>
(Using sed : sed -i '/^\$Id\$/,+1 d') (No space or star before $Id$)
Change-Id: I0801bd7cf234d32487008a8b6dcee64875b07688
Reviewed-on: https://code.wireshark.org/review/876
Reviewed-by: Evan Huus <eapache@gmail.com>