summaryrefslogtreecommitdiff
path: root/external
AgeCommit message (Collapse)Author
2017-06-12iOS patch librevengejan Iversen
Support for arm64 Change-Id: Ibba1a9486b8eaee4bf3c0cf673dd4cc112d084e0
2017-06-12iOS hyphen patchjan Iversen
Added support for arm64 Change-Id: Ie48a463b0f0a8af98001fed592d59a8e7b0d0225
2017-06-12iOS, update hunspell "revert commit" patchjan Iversen
Added support for arm64, by modifying the patch that reverts an earlier commit. Change-Id: If0d1920c1a91b3ad44c4ae9c299270b7806db811
2017-06-12iOS, harfbuzz libcoretext not foundjan Iversen
changed --with-libcoretext from "yes" to "auto" for IOS and MACOSX Change-Id: I032ad9975413709fdfaead745b63e04f0e0db27e
2017-06-12iOS patch liborcusjan Iversen
Support for arm64 Change-Id: Ie2289e4df9f90b7c31357ecfe859f087a7df9c5a
2017-06-12iOS, patch langtagjan Iversen
Support for arm64 Change-Id: I1f5d8a11f42026b08aa01a8a19059c18ff116c1c
2017-06-12Fix typosAndrea Gelmini
Change-Id: Ib2ba32d48d3df16b0b20deea84416fe15a2d7176 Reviewed-on: https://gerrit.libreoffice.org/38650 Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: Michael Stahl <mstahl@redhat.com>
2017-06-12iOS, patch harfbuzzjan Iversen
Add support for arm64 Change-Id: Icfc47b0c3b600ea3d26c40741c933028e8e9c47d
2017-06-11iOS, solved coinmp buildbreakerjan Iversen
Patch was added in wrong sequence. Change-Id: I4d700d0f9591dbb43c738ef69f823488b9c3162c
2017-06-11iOS, coinmp patch (addon)jan Iversen
support for arm64, in submodules as well Change-Id: Idc13b03d2fcc9bcf11fe4dadc0fc298a9871848f
2017-06-11iOS, exttextcat patchjan Iversen
support for arm64 Change-Id: I4e92c55253fba95944972401bbb948caf1ea903e
2017-06-11iOS coinmp patchjan Iversen
add support for arm64 Change-Id: I76dc00058abd27b8e022ffcd0d68ff00446ebeb6
2017-06-11iOS, patch for expatjan Iversen
Support for arm64 Change-Id: Iedac1ee7d1b2ba7976ae56fdf24dd66d99f16264
2017-06-11iOS, patch for libxml2jan Iversen
Added support for arm64 Change-Id: Ibd13dc2c56e2caffd97b1f3b78caece2d331b51c
2017-06-11iOS lcms2 support for arm64jan Iversen
Added patch to support arm64 Change-Id: Ie5d9ea3f3bed6e8f3142f0209a0068bb698a3960
2017-06-11iOS, patch to jpeg-turbojan Iversen
Patch allowing cputype arm64 Change-Id: Idca7d40a4186cd724599af8b4b9891f9f876aa31
2017-06-10iOS icu4c patch file addedjan Iversen
Added support for arm64 in the icu4c configuration (done in form of a patch to the original) Change-Id: I96508c29261531a24ae946b247b18c55d6259f51
2017-06-09external: drop mingw bits from ExternalPackage_xmlsecMiklos Vajna
Change-Id: Icd0526c0bf0183b80bc4c098e3307574b8b8bb8b Reviewed-on: https://gerrit.libreoffice.org/38592 Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk> Tested-by: Jenkins <ci@libreoffice.org>
2017-06-07external/graphite: Avoid -fsanitize=pointer-overflowStephan Bergmann
...that was recently introduced into Clang trunk with <https://reviews.llvm.org/D33305> "[ubsan] Add a check for pointer overflow UB". Here, _code is of type instr*, dist is of type ptrdiff_t, and sizeof(instr) is something like 8. My first impulse was to cast the result of the division (done with arguments promoted to size_t) back to ptrdiff_t, but that wouldn't help: When dist is a relatively small negative number (like -3293184), the division expression will promote it to a large unsigned (size_t) value (like 0xFFFF'FFFF'FFCD'C000), but the result (in our case, 0x1FFF'FFFF'FFF9'B800) would be small enough to fit into ptrdiff_t as a positive value. So assume that sizeof(instr) fits into int and ensure the division is done on signed values. (At least CppunitTest_sc_subsequent_filters_test started to fail with "workdir/UnpackedTarball/graphite/src/inc/Code.h:165:15: runtime error: pointer index expression with base 0x7fb90a3b4df0 overflowed to 0x7fb90a0a0df0".) Change-Id: Ie6698e38d6abec80f2fa817c42ebf20618496109
2017-06-07Remove gb_LinkTarget_add_generated_cxxobjectsStephan Bergmann
...in favor of gb_LinkTarget_add_generated_exception_objects. The former would have needed any flags to be passed in explicitly (but no call sites did), so e.g. StaticLibrary_graphite didn't have any debug information (when building with --enable-debug). I guess there is no downside to having C++ exception support enabled in these places, and using _add_generated_cxxobjects instead was likely an oversight in the first place (at least in the case of external/graphite/StaticLibrary_graphite.mk, it was that way ever since 1ceb47d96da9e7977c96241f49ad291ff0466970 "graphite: convert to gbuild", but for no apparent reason). Change-Id: I9986a6c5ec30a521095dbe5315e5ca649741a790
2017-06-06msc-externals: Package msvc_dlls only necessary in 32-bit buildsMichael Stahl
In 64-bit builds the 64-bit runtimes are installed as MSM so this hack isn't needed there. Change-Id: Id609d2beaa3de1176138bc206210820397a8b732
2017-06-06libstaroffice: use correct loop indexDavid Tardon
Change-Id: I1b862ed2d69c41361bc8c26d96c5329473fe64aa
2017-06-02more_fonts: stop building that fontconfig file on WNTMichael Stahl
Change-Id: I485b725ff824f113991900c1ea110aab1b0421af
2017-05-31libxmlsec: drop no longer needed xmlsec1-nssdisablecallbacks.patch.1Miklos Vajna
Before the gbuild conversion in commit ec6af4194e80f5f0b2e46ca59802ff397a2a4a24 (convert libxmlsec to gbuild, 2012-11-29) the makefile.mk had a comment for this patch: "Disable use of smime3 so don't need to package it". Today smime3 is packaged (see external/nss/ExternalPackage_nss.mk), so patching xmlsec is no longer needed. Change-Id: I73ec13ee92a1860f5dc3cbeed0e51d42a0320baa Reviewed-on: https://gerrit.libreoffice.org/38239 Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
2017-05-30sal,external: remove checks for obsolete VCVER=120Michael Stahl
Change-Id: I4d32b7c4b2e545a8d979bc516f64cfcbf66ecd07
2017-05-30python3: remove obsolete python-3.3.3-msvc2012-winxp.patch.1Michael Stahl
...and python-3.3.0-msvc2012.patch etc. They were actually never applied to Python 3.5. Change-Id: I5beb5a81d55ab1921411d2351bdb397ff02ba75c
2017-05-30openssl: remove Windows XP patchMichael Stahl
Change-Id: I6e80c33d134100235ac1007154ca7f6151b59c2f
2017-05-30nss: remove obsolete nss-winXP-sdk.patch.1Michael Stahl
Change-Id: I044be3b53134081bcbdbfd6fd252d6851d3dba41
2017-05-30curl: remove curl-xp.patch.1Michael Stahl
Change-Id: I329289c97b857370852c982a8b74b739b9ac3d78
2017-05-30stop spewing libxml2 warnings during filter detectionCaolán McNamara
Change-Id: Ifef3e75065e55aefba0f9498cf517efaf78ba6c1
2017-05-25upload poppler 0.55.0David Tardon
Change-Id: I19b7d324a2dc50d006b4d255ddb6a5c1f15bc616 Reviewed-on: https://gerrit.libreoffice.org/37998 Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: David Tardon <dtardon@redhat.com>
2017-05-25xmlsec: use configure flags instead of patching out apps/docsMiklos Vajna
These flags are new in xmlsec-1.2.24. Change-Id: If32faa6e1d3a5f51fa4d631973d8bd0962783a34 Reviewed-on: https://gerrit.libreoffice.org/38004 Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
2017-05-24liblangtag#10 support roundtripping ca@valenciaCaolán McNamara
Change-Id: I67932541fbb14316798fe94b64b4b32a3d1c29f6 Reviewed-on: https://gerrit.libreoffice.org/37984 Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: Caolán McNamara <caolanm@redhat.com> Tested-by: Caolán McNamara <caolanm@redhat.com>
2017-05-24external: update pdfium to 3094Miklos Vajna
This allows dropping the now upstreamed system-zlib.patch.1. Change-Id: I92668ff243fa7f14a418cec419c003d4a9bc96ce Reviewed-on: https://gerrit.libreoffice.org/37960 Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk> Tested-by: Jenkins <ci@libreoffice.org>
2017-05-18external/libxmlsec: Pass down gb_VISIBILITY_FLAGSStephan Bergmann
...otherwise ASan build started to fail in CppunitTest_services with > ==16664==ERROR: AddressSanitizer: odr-violation (0x7f4583587440): > [1] size=35 'xmlSecNs' strings.c:22:15 > [2] size=35 'xmlSecNs' strings.c:22:15 > These globals were registered at these points: > [1]: > #0 0x44f410 in __asan_register_globals.part.11 /data/sbergman/github.com/llvm-project/llvm-project-20170507/compiler-rt/lib/asan/asan_globals.cc:358 > #1 0x7f458295cded in asan.module_ctor (instdir/program/libxsec_gpg.so+0x197ded) > > [2]: > #0 0x44f410 in __asan_register_globals.part.11 /data/sbergman/github.com/llvm-project/llvm-project-20170507/compiler-rt/lib/asan/asan_globals.cc:358 > #1 0x7f45833cc28d in asan.module_ctor (instdir/program/libxsec_xmlsec.so+0x5da28d) > > ==16664==HINT: if you don't care about these errors you may set ASAN_OPTIONS=detect_odr_violation=0 > SUMMARY: AddressSanitizer: odr-violation: global 'xmlSecNs' at strings.c:22:15 > ==16664==ABORTING after cae5f2a543b31552ccd9765aca5eb514fa694e07 "gpg4libre: initial GPG signature generation" Change-Id: I689241df52a7e878bfbd2d3dd6409bf0c104638b
2017-05-17xmlsecurity: use xmlsec API instead of patching out cert verificationMiklos Vajna
This flag does exactly what we need since xmlsec-1.2.24. Change-Id: I3ae052d4bfe564c3234aef2511ef82ebdb452ebe Reviewed-on: https://gerrit.libreoffice.org/37700 Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk> Tested-by: Jenkins <ci@libreoffice.org>
2017-05-10Upgrade libxmlsec to 1.2.24Miklos Vajna
Upstream changes interesting for us: - Added ECDSA-SHA1, ECDSA-SHA256, ECDSA-SHA512 support for xmlsec-nss, so we can drop 2 patches - Fixed XMLSEC_KEYINFO_FLAGS_X509DATA_DONT_VERIFY_CERTS handling, which allows dropping xmlsec1-noverify.patch.1 in the future Also backport a patch from xmlsec master that fixes signature creation on Windows (the release regressed in this regard). Change-Id: I2c14328283bf7d4f8af5595ea4c1efc29ee81f9e
2017-05-09Use BOOST_USE_WINDOWS_H to avoid errors when building with clang-clStephan Bergmann
> [build CXX] workdir/UnpackedTarball/boost/libs/locale/src/shared/date_time.cpp [...] > In file included from C:/lo64/core/workdir/UnpackedTarball/boost/libs/locale/src/shared/date_time.cpp:11: > In file included from C:/lo64/core/workdir/UnpackedTarball/boost\boost/thread/mutex.hpp:14: > In file included from C:/lo64/core/workdir/UnpackedTarball/boost\boost/thread/win32/mutex.hpp:9: > In file included from C:/lo64/core/workdir/UnpackedTarball/boost\boost/thread/win32/basic_timed_mutex.hpp:14: > C:/lo64/core/workdir/UnpackedTarball/boost\boost/thread/win32/thread_primitives.hpp(172,55): error: conflicting types for 'CreateMutexA' > __declspec(dllimport) void* __stdcall CreateMutexA(_SECURITY_ATTRIBUTES*,int,char const*); > ^ > C:/PROGRA~2/WI3CF2~1/10/Include/10.0.15063.0/um\synchapi.h(489,1): note: previous declaration is here > CreateMutexA( > ^ [...] Change-Id: I141ea196250dc855c780f237436c0f19f0594cb1
2017-05-05libxmlsec: remove xmlsec1-nssmangleciphers.patch.1Miklos Vajna
This was added in commit ebd1b95bb5f9235d1dba1b840fd746c9b53320d2 (INTEGRATION: CWS xmlsec08 (1.1.2); FILE ADDED, 2005-03-10). According to CWS history it was introduced in the 1.1.2.1 part, without any further comments. Before the gbuild conversion in commit ec6af4194e80f5f0b2e46ca59802ff397a2a4a24 (convert libxmlsec to gbuild, 2012-11-29) the makefile.mk had a comment for this patch: "Dubious, do we still need this ?". My best guess is that this was added as part of some effort to do ODF encryption (not just signing) in xmlsecurity, but code for that on the xmlsecurity part is already removed. Change-Id: I3a5f1fedd7ce10b8b874bb8a3c9e6260213fbd8f Reviewed-on: https://gerrit.libreoffice.org/37261 Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
2017-05-02Make some externals' libtool copies cope with -mllvmStephan Bergmann
...as needed e.g. for <https://bugs.llvm.org/show_bug.cgi?id=32349> "r294897 + NewGVN cause build failure with LibreOffice", by applying <http://git.savannah.gnu.org/cgit/libtool.git/commit/ ?id=d9a35fe9d3508b5c0d56e7f2ec80fc05e8415fa3> "libtool: Discard '-mllvm $arg' options when linking." Change-Id: Id2afc3c8af3c6c9595e7cb33cef5084a74f78cb0
2017-05-02Adapt bin/gen-boost-headers to Clang 5Stephan Bergmann
...since <http://llvm.org/viewvc/llvm-project?view=revision&revision=301647> "Use the -Wunknown-warning-option group for the 'unknown warning group' diagnostic in #pragma diagnostic". * external/boost/include/boost/{locale.hpp,locale/gnu_gettext.hpp} would have been removed by 'make cmd cmd=bin/gen-boost-headers' as they are still unused from c25eee44966703cb27d632bccb39b20978341ffd "build boost::locale library", but there's reportedly a patch in Gerrit to actually use them, so I fixed them manually for now. * The deviating comment style is to keep lines no wider than 80 characters. Change-Id: I64603ae8d8a82781eda46f12c9dd5c68dcf395b9
2017-04-28try without iconv, we probably don't need it in practiceCaolán McNamara
Change-Id: I14782971dd35f8d96f1e9b08fdf060aa917a1e37
2017-04-28build boost::locale libraryCaolán McNamara
Change-Id: I40b89a0df483645fc743fb092d3d39ea682c510c Reviewed-on: https://gerrit.libreoffice.org/37060 Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: Caolán McNamara <caolanm@redhat.com> Tested-by: Caolán McNamara <caolanm@redhat.com>
2017-04-27Add ICU ticket number for this patchEike Rathke
https://ssl.icu-project.org/trac/ticket/13176 Change-Id: I1bf7a0787f46186895e7ee9144d5f219ea59e2df
2017-04-27Add ICU ticket number for this patchEike Rathke
https://ssl.icu-project.org/trac/ticket/13175 Change-Id: Ib7756f3c41d2395f65d898e39808616c04ee58ee
2017-04-26Upgrade to ICU 59.1 follow-up: only add icu/source/data/misc/icudata.rcEike Rathke
... instead of the entire icu4c-59_1-data.zip content. Change-Id: Icfbf7d4103059f525b303b194212b78935fb00b1 Reviewed-on: https://gerrit.libreoffice.org/37001 Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: Eike Rathke <erack@redhat.com>
2017-04-26Upgrade to ICU 59.1Eike Rathke
Also regenerated all patches using make icu.genpatch (hence the .1 suffix that indicates the path level) as some hunks did not apply anyway and all now have the correct offset. Using genpatch may have the future benefit to yield smaller diffs between different versions of patches. Also prefixed all patch names with icu4c- for a cleaner listing. New patches introduced are prefixed with icu4c-59-... Change-Id: Ia83754b0823839887fce1a1d4ed04f8375b113c2 Reviewed-on: https://gerrit.libreoffice.org/36809 Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: Eike Rathke <erack@redhat.com>
2017-04-26external: update pdfium to 3064Miklos Vajna
This contains the new FPDFPath_GetFillColor / FPDFPageObj_GetType APIs I want to use in CppunitTest_vcl_pdfexport. Change-Id: I275ff761188c07dbbab27a1e0715ac250cd306c9 Reviewed-on: https://gerrit.libreoffice.org/36974 Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
2017-04-21external/epoxy: Work around missing clang-cl DLL-init CONSTRUCT supportStephan Bergmann
Looks like neither the __GNUC__ specific __attribute__((constructor)) definition of CONSTRUCT nor the _MSC_VER specific one (if the #ifs were reordered so that clang-cl would pick the latter) with __declspec(allocate(".CRT$XCU")) are supported by clang-cl, and both are rather silently ignored. That means that library_init is not called, library_initialized remains false, and the first call to get_dlopen_handle aborts. But this whole "verify that get_dlopen_handle isn't called too early" business is somewhat pointless here (and that's the only use of the CONSTRUCT macro, and DESTRUCT isn't used at all), so just short-circuit it for clang-cl for now. Change-Id: I5d50df3574d350f9591e807ef0fb6a1b02dc34ec
2017-04-21gbuild: Remove MSVC 2013 legacy codeDavid Ostrovsky
Uwinapi is discontinued. Change-Id: I063b4d0d8fab2d60de168e960a63b8181158ac01 Reviewed-on: https://gerrit.libreoffice.org/23198 Reviewed-by: David Ostrovsky <david@ostrovsky.org> Tested-by: David Ostrovsky <david@ostrovsky.org>
generated by cgit (git 2.34.1) at 2025-02-11 05:30:39 +0000