summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2020-09-23if allocating Vulkan surface fails, fall back to Skia raster surfaceLuboš Luňák
Occassionally there may be very large surfaces, such as in tdf#135952. Try to fall back to raster, which is more likely to succeed, given that it uses system RAM instead of video RAM. Change-Id: I81994b174e5e52066eacc5f8778e9469b042f9c8 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103170 Tested-by: Jenkins Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
2020-09-23assert that SkiaSalBitmap mImage is not deleted if it's the only dataLuboš Luňák
Change-Id: Idec2e0f3e852c37b37e220ac5059ef0b6accc77a Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103205 Tested-by: Jenkins Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
2020-09-23createSurface() should be enough if the surface doesn't exist yetLuboš Luňák
Change-Id: Ic390b94e6140169d878b428c782e03c29cce0ea2 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103184 Tested-by: Jenkins Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
2020-09-23Executable_pdfverify: use PDFSignatureHelper for signature verificationMiklos Vajna
Towards making vcl::filter::PDFDocument an implementation detail of PDFSignatureHelper during signature verification: so pdfverify, cppunit tests and the UI shares more code. Change-Id: Ibb68933d754e392bce0ebbf06be8916ab3f7efdc Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103214 Tested-by: Jenkins Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
2020-09-23tdf#128136: Build curl, nss, and xmlsec for iOS, tooTor Lillqvist
We must link nss statically, including the three dylibs that normally are loaded at run-time, because including bare dylibs in an iOS appp on the App Store is not OK. See https://developer.apple.com/forums/thread/125796 . For linking the softokn3 library statically, NSS already had code, behind NSS_STATIC_SOFTOKEN ifdefs. Introduce two more macros: NSS_STATIC_FREEBL for the freebl library and NSS_STATIC_PKCS11 for the nssckbi library. Turn off parallelism for the sub-make building nss. There seems to be race conditions or something when running simultaneous instances of the nsinstall.py script or the nsinstall program in nss (used when building nss for the build platform). When cross-compiling from macOS, use python3 to run the nsinstall.py script, as it is Python 3. Change-Id: Idd427b5ebf21f802b3feb0d5a3d259317ba8fc67 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103106 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com> Reviewed-by: Tor Lillqvist <tml@collabora.com> Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103218 Tested-by: Jenkins
2020-09-23WIN merge CWinClipbImpl into CWinClipboardJan-Marek Glogowski
Get rid of the already minimal pImpl. Change-Id: Ida6fedab6e779b649e546bae2cda5f14fd4090d2 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103211 Tested-by: Jenkins Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
2020-09-23Update git submodulesOlivier Hallot
* Update helpcontent2 from branch 'master' to ab4b270d35eb6db2f0e2408f33e7817a9a0e7ce0 - Refactor Relations Help page Change-Id: Iaf32149f3a546c04bc13206213c748bc6be01b4c Reviewed-on: https://gerrit.libreoffice.org/c/help/+/103222 Tested-by: Jenkins Reviewed-by: Olivier Hallot <olivier.hallot@libreoffice.org>
2020-09-23Update git submodulesOlivier Hallot
* Update helpcontent2 from branch 'master' to cd1253e8686bda4032ec658766eb8a103aca5d40 - tdf#136939 Make strings localizable Change-Id: Ie33cdda080ee06532170da5bc410ee660841bf40 Reviewed-on: https://gerrit.libreoffice.org/c/help/+/103221 Tested-by: Jenkins Reviewed-by: Olivier Hallot <olivier.hallot@libreoffice.org>
2020-09-23Update git submodulesOlivier Hallot
* Update helpcontent2 from branch 'master' to 2c59643f8b17e2f9e8d51f155687db86d7a26447 - tdf#136948 Fix broken link. Change-Id: I3867f97ce8dec389bae38b208c943588acbd46fb Reviewed-on: https://gerrit.libreoffice.org/c/help/+/103220 Tested-by: Jenkins Reviewed-by: Olivier Hallot <olivier.hallot@libreoffice.org>
2020-09-23Add Stephan's "Very bad HACK" to the iOS code, too, for iOS 14Tor Lillqvist
Change-Id: Ie05d2dc77bfdcd323037d2e94b523808df4e1c9c Reviewed-on: https://gerrit.libreoffice.org/c/core/+/102916 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com> Reviewed-by: Tor Lillqvist <tml@collabora.com> Reviewed-on: https://gerrit.libreoffice.org/c/core/+/102980 Tested-by: Tor Lillqvist <tml@collabora.com>
2020-09-22Use sed syntax that isn't GNU-specificTor Lillqvist
Fix fallout from 63972e79bbb9ea9654e755381641052632b0402c. Change-Id: Ie11514eb396fdc68e05ee180322df27e29ede03f Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103216 Tested-by: Jenkins Reviewed-by: Tor Lillqvist <tml@collabora.com>
2020-09-22Move MimeContentTypeFactory into vclJan-Marek Glogowski
The code is just used in vcl from LO's POV. This way we can drop the dtrans directory and get rid of yet an other library. Change-Id: Id77568e63a6fef4af30b49e035a9d76211b127a1 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103210 Tested-by: Jenkins Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
2020-09-22WIN move dtrans code into vcl/win/dtransJan-Marek Glogowski
There is nothing abstract about either the clipboard or data transfer code in that directory and it's just used on Windows. All other backends implement this code in VCL, so this moves almost all code, except for the common MimeContentTypeFactory, into the vcl Windows backend / vclplug_win. This also drops four DLLs: sysdtrans, dnd, dtrans and ftransl. Change-Id: I7018f50768bf221447b40487cc1f8a8586da33c4 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103209 Tested-by: Jenkins Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
2020-09-22Manually select JDK outside /Library/Java/JavaVirtualMachines on macOSStephan Bergmann
32bc8ddbf335dd26019edcf12758643b4cff9913 "tdf#94716 allow Oracle's JDK to be used on OS X 10.10 and 10.11" had restricted both addJavaInfosDirScan (jvmfwk/plugins/sunmajor/pluginlib/util.cxx, relevant when LO scans for available Java runtimes) and JvmfwkUtil_isLoadableJVM (which is more central and also relevant when manually adding a Java installation on the Advanced options page) to JDK installations under /Library/Java/JavaVirtualMachines. Recent versions of OpenJDK for macOS are provided as *.tar.gz files (e.g., <https://download.java.net/java/GA/jdk15/779bf45e88a44cbd9ea6621d33e33db1/36/GPL /openjdk-15_osx-x64_bin.tar.gz> avaiable at <http://jdk.java.net/15/>) that can potentially be installed anywhere, so it looks useful to remove the /Library/Java/JavaVirtualMachines restriction from JvmfwkUtil_isLoadableJVM. I'm not sure whether that would have any nagative impact, esp. for any scenario that would still be relevant today. When I unpacked the above openjdk-15_osx-x64_bin.tar.gz to ~/OpenJDK/ on my macOS 10.15.6, and manually added /Users/stephan/OpenJDK/jdk-15.jdk/Contents/Home on the Advanced options page, everything appeared to work fine. Change-Id: I15545cc786a4a423f3bfe22ba477a45d6ef7fdf1 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103212 Tested-by: Jenkins Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
2020-09-22Fix typosAndrea Gelmini
Change-Id: I2030700ea031c03242c13d1cb96babf2bc255bab Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103163 Tested-by: Jenkins Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
2020-09-22ofz#25855 overflow in nTmpOffsetCaolán McNamara
we already know nLength is >= 24 so just move the calc to the other term Change-Id: Ic52f1686ccf81e6b13d7eb7e74dbd9cb51c8ea01 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103183 Tested-by: Jenkins Tested-by: Caolán McNamara <caolanm@redhat.com> Reviewed-by: Caolán McNamara <caolanm@redhat.com>
2020-09-22Fix typoAndrea Gelmini
Change-Id: Ib7e7fe2181a23276eae25f0cf37f80be17dad588 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103181 Tested-by: Jenkins Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
2020-09-22Suppress PYTHONWARNINGS=error in system g-ir-scannerStephan Bergmann
With 315919306c7b6e95db6a280c4aa8d2203970e292 "Set PYTHONWARNINGS to error by default for --enable-werror" my build on Fedora 32 started to fail in CustomTarget_sysui/share with > Exception ignored in: <socket.socket fd=4, family=AddressFamily.AF_UNIX, type=SocketKind.SOCK_STREAM, proto=0> > Traceback (most recent call last): > File "/usr/lib/python3.8/site-packages/abrt_exception_handler3.py", line 93, in write_dump > response = send(data) > ResourceWarning: unclosed <socket.socket fd=4, family=AddressFamily.AF_UNIX, type=SocketKind.SOCK_STREAM, proto=0> > Traceback (most recent call last): > File "/usr/bin/g-ir-scanner", line 103, in <module> > from giscanner.scannermain import scanner_main > File "/usr/lib64/gobject-introspection/giscanner/scannermain.py", line 35, in <module> > from giscanner.ast import Include, Namespace > File "/usr/lib64/gobject-introspection/giscanner/ast.py", line 29, in <module> > from .sourcescanner import CTYPE_TYPEDEF, CSYMBOL_TYPE_TYPEDEF > File "/usr/lib64/gobject-introspection/giscanner/sourcescanner.py", line 24, in <module> > from .libtoolimporter import LibtoolImporter > File "/usr/lib64/gobject-introspection/giscanner/libtoolimporter.py", line 21, in <module> > import imp > File "/usr/lib64/python3.8/imp.py", line 31, in <module> > warnings.warn("the imp module is deprecated in favour of importlib; " > DeprecationWarning: the imp module is deprecated in favour of importlib; see the module's documentation for alternative uses Change-Id: I919caaad482e844ef1549f248a2faf2274c9efa0 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103177 Tested-by: Jenkins Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
2020-09-22Related: tdf#132970 handle more places with potentially utf16 bulletsCaolán McNamara
Change-Id: Iac6b319700d610b5a1debff0a633172b2411c40e Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103161 Tested-by: Jenkins Reviewed-by: Caolán McNamara <caolanm@redhat.com>
2020-09-22tdf#121323 avoid assert about duplicate listenersCaolán McNamara
attach the new one after the old one is disposed Change-Id: I0a020aeaa658832c03d18fcfaf341a56b8e65dfc Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103175 Tested-by: Jenkins Reviewed-by: Caolán McNamara <caolanm@redhat.com>
2020-09-22tdf#121323 avoid fatal exception with a11y enabledCaolán McNamara
not related to the problem described there, but worth avoiding Change-Id: I5a33425d856d1a2d31a4e735b9936a89576121b3 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103174 Tested-by: Caolán McNamara <caolanm@redhat.com> Reviewed-by: Caolán McNamara <caolanm@redhat.com>
2020-09-22tdf#136620: sw_ww8export: Add unittestXisco Fauli
Change-Id: Ib1d4512b4ac73976192e6f41362aba3e1e185629 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103168 Tested-by: Jenkins Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org>
2020-09-22Simplify those assertLayout checksStephan Bergmann
...recently introduced with 52a49f9e480ca03e231cfda82640a928393131c9 "static_assert that O[U]StringLiteral are layout compatible with rtl_[u]String" Change-Id: Ie89e92aa4230329aa582c4211c2001117d2287b5 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103138 Tested-by: Jenkins Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
2020-09-22tdf#104722: sd_import_tests: Add unittestXisco Fauli
Change-Id: I94b28074a773cec99227bbaae4a82bd085e17a94 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103159 Tested-by: Jenkins Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org>
2020-09-22this is an export test, move it to the right fileXisco Fauli
Change-Id: Idfdcc81d1a7ddb4ac78c68b9b0c07bd6b5fb5d31 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103160 Tested-by: Jenkins Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org>
2020-09-22tdf#93756 show highlight color in style previewCaolán McNamara
Change-Id: Ibddd97f8829d824667657177c7311680707df4e5 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103167 Tested-by: Jenkins Reviewed-by: Caolán McNamara <caolanm@redhat.com>
2020-09-22Added new command to resolve the comment threadPranam Lashkari
Change-Id: I8a4e5f63ee6ea5e560fae8a5d3602178f2b58b36 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/102827 Tested-by: Jenkins Reviewed-by: Pranam Lashkari <lpranam@collabora.com>
2020-09-22changed FN_RESOLVE_NOTE behaviour to resolve single notePranam Lashkari
Earlier it used to resolve the entire thread and there was no way to resolve a single comment in thread Change-Id: I471c5b575dd365d52653835be5b0d40009bb5cfa Reviewed-on: https://gerrit.libreoffice.org/c/core/+/102581 Tested-by: Jenkins Reviewed-by: Pranam Lashkari <lpranam@collabora.com>
2020-09-22do not use SkShader if not neededLuboš Luňák
Change-Id: I9913673c889e605d608e25dad9a36b05912447be Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103153 Tested-by: Jenkins Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
2020-09-22fix parsing of Vulkan version numbersLuboš Luňák
The Windows+OpenGL handling pads the numbers so that .98 > 0.978, but Vulkan uses "normal" numeric ordering for versions. Change-Id: I766baf50e3505a96aa85163962400bb69c0acea6 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103118 Tested-by: Jenkins Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
2020-09-22lok:sidebar: deactivate the shell with no default contextHenry Castro
When a different view deactivates to current shell, then it is set to default context that cause a side effect into another view with different panels in the sidebar. Change-Id: Ie99cde89bdfdb08e1ac93aed947cf13048800c0a Reviewed-on: https://gerrit.libreoffice.org/c/core/+/102887 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com> Reviewed-by: Henry Castro <hcastro@collabora.com> (cherry picked from commit e7c3678e309b59479be24be2b7c57f8b20dce4dd) Reviewed-on: https://gerrit.libreoffice.org/c/core/+/102943 Tested-by: Jenkins Reviewed-by: Andras Timar <andras.timar@collabora.com>
2020-09-22GtkMenuButton had two labelsCaolán McNamara
Change-Id: Icd1d2945c79e819fd1f3df5acb99d922f27c251f Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103123 Tested-by: Jenkins Reviewed-by: Caolán McNamara <caolanm@redhat.com>
2020-09-22tdf#136706 writerfilter: revert my info-gathering assertJustin Luth
The assert was failing in crashTesting for: wget http://bugs.documentfoundation.org/attachment.cgi?id=68205 -O tdf55725-1.docx ./instdir/program/soffice.bin --headless --convert-to docx ./tdf55725-1.docx and also for wget https://bugs.documentfoundation.org/attachment.cgi?id=127288 -O tdf102131-1.docx Both of these documents have SAX errors when loading. Running these interactively, saying "no, do not continue loading" parses a bit more to give the specific error and that is what was running into the assert. Actually importing the document itself does not hit the assert, nor does round-tripping it. Since this is a stupid enough situation, and because I can't think of a good way to word a SAL_WARN, and because I can't say that I have detailed knowledge about this area, I'll just remove the whole thing instead of leaving a warning. Change-Id: If50b06f0103adca348ca314ebc92e7e4045967dc Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103133 Tested-by: Jenkins Reviewed-by: Justin Luth <justin_luth@sil.org> Reviewed-by: Caolán McNamara <caolanm@redhat.com>
2020-09-22Restore .uno:ShowHiddenParagraphs to Writer View menuIlmari Lauhakangas
It was removed without explanation in adb5cdcf2352d3f1c78645feafc4c270b58b197a Restoring it makes our Help sensible again. Change-Id: I4fade9b58bd4cc010ae786bb3b2e58cb12ccebd6 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103154 Reviewed-by: Ilmari Lauhakangas <ilmari.lauhakangas@libreoffice.org> Tested-by: Ilmari Lauhakangas <ilmari.lauhakangas@libreoffice.org>
2020-09-22__coverity_tainted_data_sanitize__ can't be constCaolán McNamara
apparently, so use const_cast on its input instead Change-Id: Ib0dfd94c144a2509470ca7a9b3b8fbfacbfd7581 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103148 Tested-by: Jenkins Reviewed-by: Caolán McNamara <caolanm@redhat.com>
2020-09-22tdf#134227: sw_uiwriter: Add unittestXisco Fauli
Change-Id: I473ab1ea5ccc603c6086189075db49c224d33743 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103149 Tested-by: Jenkins Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org>
2020-09-22OUStringLiteral/OStringLiteral coverity PARSE_ERROR workaroundCaolán McNamara
do more like commit 121771e37f7e2de41cd5643475861062bf25627b Date: Mon Sep 21 09:17:54 2020 +0200 Make some OUStringLiteral vars constexpr cause coverity can live with that Change-Id: I9efd7f848289c4865997a44c6780373068422227 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103147 Tested-by: Jenkins Reviewed-by: Caolán McNamara <caolanm@redhat.com>
2020-09-22tdf#132970 SMP bullets mangledCaolán McNamara
working: a) bullet preview b) writer rendering c) save to odt a) load from odt Change-Id: I2f85576389fe4f0437f81799c14dfd98c8c40b2e Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103129 Tested-by: Jenkins Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
2020-09-22cid#1467075 silence Unchecked return valueCaolán McNamara
Change-Id: I77809ff271004a216131f26a26696d5717877a8d Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103145 Tested-by: Jenkins Reviewed-by: Caolán McNamara <caolanm@redhat.com>
2020-09-22Fix symbols already defined MSVC link errorMike Kaganski
... after af90b8089405d6f042207f5639e750f08798ae92, which made VCL to instantiate exported implementation of the template. Fixes the error: ivcl.lib(vcllo.dll) : error LNK2005: "public: virtual void __cdecl cppu::WeakImplHelper<class com::sun::star::frame::XStatusListener>::acquire(void)" (?acquire@?$WeakImplHelper@VXStatusListener@frame@star@sun@com@@@cppu@@UEAAXXZ) already defined in helpinterceptor.o Modelled after 013f84d06f7ad76d72b863170891589c3504508c. Change-Id: I40ba226d95ad2f1fda177839d9468d6861ca54c5 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103144 Tested-by: Mike Kaganski <mike.kaganski@collabora.com> Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
2020-09-22tdf#38885 Remove createTextFromNodedante
Change-Id: I85df73e498d61b976ddb908ab1b9c5e368d178cf Reviewed-on: https://gerrit.libreoffice.org/c/core/+/98795 Tested-by: Jenkins Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
2020-09-22tdf#136650 DOCX pie chart import: fix percentageBalazs Varga
Follow-up of commit e0da00d655ecca5986eea3812a8a670c6adbc40f (tdf#132174 Chart DOCX import: fix label number format). Change-Id: Ie0b2fe6305d7696de6dddc1f17d078fc2e749e4e Reviewed-on: https://gerrit.libreoffice.org/c/core/+/102753 Tested-by: László Németh <nemeth@numbertext.org> Reviewed-by: László Németh <nemeth@numbertext.org>
2020-09-22Avoid unique_ptr(nullptr_t) ctor for incomplete argument typeStephan Bergmann
...hoping that this will avoid > In file included from /home/tdf/lode/jenkins/workspace/lo_ubsan/sw/source/filter/ww8/wrtw8nds.cxx:25: > In file included from /home/tdf/lode/jenkins/workspace/lo_ubsan/sw/source/filter/ww8/docxexport.hxx:23: > In file included from /home/tdf/lode/jenkins/workspace/lo_ubsan/sw/source/filter/ww8/wrtww8.hxx:23: > In file included from /home/tdf/lode/jenkins/workspace/lo_ubsan/include/sot/storage.hxx:24: > In file included from /home/tdf/lode/jenkins/workspace/lo_ubsan/include/tools/stream.hxx:28: > In file included from /home/tdf/lode/opt_private/gcc-7.3.0/lib/gcc/x86_64-pc-linux-gnu/7.3.0/../../../../include/c++/7.3.0/memory:80: > /home/tdf/lode/opt_private/gcc-7.3.0/lib/gcc/x86_64-pc-linux-gnu/7.3.0/../../../../include/c++/7.3.0/bits/unique_ptr.h:76:16: error: invalid application of 'sizeof' to an incomplete type 'sax_fastparser::FastAttributeList' > static_assert(sizeof(_Tp)>0, > ^~~~~~~~~~~ > /home/tdf/lode/opt_private/gcc-7.3.0/lib/gcc/x86_64-pc-linux-gnu/7.3.0/../../../../include/c++/7.3.0/bits/unique_ptr.h:268:4: note: in instantiation of member function 'std::default_delete<sax_fastparser::FastAttributeList>::operator()' requested here > get_deleter()(__ptr); > ^ > /home/tdf/lode/opt_private/gcc-7.3.0/lib/gcc/x86_64-pc-linux-gnu/7.3.0/../../../../include/c++/7.3.0/bits/unique_ptr.h:233:45: note: in instantiation of member function 'std::unique_ptr<sax_fastparser::FastAttributeList, std::default_delete<sax_fastparser::FastAttributeList> >::~unique_ptr' requested here > constexpr unique_ptr(nullptr_t) noexcept : unique_ptr() { } > ^ > /home/tdf/lode/jenkins/workspace/lo_ubsan/include/sax/fshelper.hxx:33:34: note: forward declaration of 'sax_fastparser::FastAttributeList' > namespace sax_fastparser { class FastAttributeList; } > ^ (<https://ci.libreoffice.org/job/lo_ubsan/1767/>) with libstdc++ prior to <https://gcc.gnu.org/git/?p=gcc.git;a=commit; h=c3ba63c314d61362f7c48c4feeefa13ea3978344> "PR libstdc++/87704 fix unique_ptr(nullptr_t) constructors". Change-Id: Ie3b86b83d0b0e70fe44e2b2022d410643e265fbf Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103139 Tested-by: Jenkins Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
2020-09-22tdf#123610 DOCX shape import: keep text uprightSzabolcs Toth
Grab-bag attribute upright to keep text upright regardless of shape rotation, fixing the text direction in MSO after a round-trip. Co-authored-by: Balázs Regényi Change-Id: If18a12c7728317a7af093048b07bc392d2964389 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/102690 Tested-by: László Németh <nemeth@numbertext.org> Reviewed-by: László Németh <nemeth@numbertext.org>
2020-09-22tdf#136337 cairo canvas: fix missing image with negative heightMiklos Vajna
Regression from commit 78036f74fa74ee2552e79064660634e1342692ff (tdf#135094 cairo canvas: fix black slide containing a very small image, 2020-08-14), we try to predict when cairo would end up in an invalid state due to an invalid transformation matrix, but in this case we were too aggressive, so the image was missing while presenting. Fix the problem by allowing negative sizes, just require that neither of width/height is zero, because negative values are poor, but valid way of vertical/horizontal flip. [ No testcase, our tests are typically headless and currently SvpSalGraphics::SupportsCairo() reports false, so this would be tricky to test. ] Change-Id: Iaf843c0d40c108d5221c9b94b39d617e4d50f65c Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103127 Tested-by: Jenkins Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
2020-09-22Fix AVX2 cpuid checksDr. David Alan Gilbert
At the moment test_cpu_runtime_detection_AVX2.cxx is compiled with -mavx2 to allow it to use the intrinsics; however the compiler jumps at the chance to use newer instructions outside the actual test; in my case using AVX in the string manipulation in addTestsToSuite when my CPU doesn't actually have AVX. Swing the actual check into a separate file and only compile that with the extra flag. We probably need the same change for the SSE* checks as well. Change-Id: I1683231932fff264a87c96ac95ac1d24b921163a Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103075 Tested-by: Jenkins Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>
2020-09-22Update git submodulesGabor Kelemen
* Update helpcontent2 from branch 'master' to 9b4e021c4ba2cd74ebd81bf0a46efc762ad36931 - tdf#136071 (related) Remove Spacing submenu from Calc Format menu reference Change-Id: I474811cc3ea234a38537fa2bfed577ae6570ae90 Reviewed-on: https://gerrit.libreoffice.org/c/help/+/103126 Tested-by: Jenkins Reviewed-by: Ilmari Lauhakangas <ilmari.lauhakangas@libreoffice.org>
2020-09-21static_assert that O[U]StringLiteral are layout compatible with rtl_[u]StringStephan Bergmann
...as was suggested by Mike Kaganski in a comment at <https://gerrit.libreoffice.org/c/core/+/102222/10# message-aa8bcf42f04686766440e2848c7d1f76f474f2f8> "Turn OUStringLiteral into a consteval'ed, static-refcound rtl_uString". Doing so revealed that at least lopglugin:unusedmember needs to handle InjectedClassNameType as an alternative to RecordType in at least one place. (More places across compilerplugins might benefit from handling InjectedClassNameType too, but which did not lead to assertion failures for now and should be addressed in follow-up issues.) Change-Id: Icdb8b069324b5ce5f3c7c6b92989379ccb67fc8b Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103125 Tested-by: Jenkins Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
2020-09-21No need to make the OString(std::string_view) ctor explicitStephan Bergmann
<https://gerrit.libreoffice.org/c/core/+/101829> "Make std::u16string_view -> OUString construction explicit" (which has meanwhile been abandoned) had intended to make the OUString(std::u16string_view) ctor explicit, mostly not for the rationale given in the commit message there (it being "a rather expensive operation"), but rather to avoid anticipated ambiguity issues when introducing e6dfaf9f44f9939abc338c83b3024108431d0f69 "Turn OUStringLiteral into a consteval'ed, static-refcound rtl_uString"---but which has ultimately been added in a form that did not give rise to such ambiguity issues after all. 4b9e440c51be3e40326bc90c33ae69885bfb51e4 "Turn OStringLiteral into a consteval'ed, static-refcound rtl_String" had added the OString(std::string_view) ctor as explicit to match the corresponding explicit OUString(std::u16string_view) from the above <https://gerrit.libreoffice.org/c/core/+/101829>. But as that has now been abandoned, there is no good reason to keep this ctor explicit, either. Change-Id: I05742de5f3992ad5995149631bf8d55c8d448387 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103124 Tested-by: Jenkins Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
2020-09-21add an explicit --disable-qrcodegen configure optionCaolán McNamara
Change-Id: If8e965fa955aecdb9e7011bdddc690de9cad0c4d Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103120 Tested-by: Jenkins Reviewed-by: Caolán McNamara <caolanm@redhat.com>
generated by cgit (git 2.34.1) at 2024-07-19 04:14:58 +0000