summaryrefslogtreecommitdiff
path: root/configure.ac
AgeCommit message (Collapse)Author
2018-07-26Avoid configure error when warning with no TERMLuke Deller
When configure wants to show a warning, it tries to check whether the terminal supports colour. This check produces an error message when running with no TERM environment variable set (seen on a tinderbox) Fix this check so that it tolerates failure of the `tput colors` command Change-Id: I263136142120cf66fe900df0deb74d2fa6e5d1d0 Reviewed-on: https://gerrit.libreoffice.org/57009 Tested-by: Jenkins Reviewed-by: Michael Stahl <Michael.Stahl@cib.de>
2018-07-24Fix typosAndrea Gelmini
Change-Id: I07cf04cd2acc5b50d64224edb9f2c2de2c61e356 Reviewed-on: https://gerrit.libreoffice.org/57884 Reviewed-by: Julien Nabet <serval2412@yahoo.fr> Tested-by: Jenkins
2018-07-20Don't search for old (<= 2.0) .NET SDK anymoreAron Budea
Try to find >= 4.6 .NET SDK first. +allow .NET SDK 4.7.2. Change-Id: I64084e894ee829ecff091fe16a45f309a51eb4d5 Reviewed-on: https://gerrit.libreoffice.org/57550 Tested-by: Jenkins Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
2018-07-18configure: diagnose and report cygpath convert failureVasily Melenchuk
If Windows machine has short filenames disabled, it is not reported during configure run, but leads to unexpected and difficult to understand failures later. Checking of cygpath call conversion result should help to diagnose this problem earlier. Change-Id: Ia93ae612f4bc624bc3e71c6733838fd14b075f64 Reviewed-on: https://gerrit.libreoffice.org/57608 Tested-by: Jenkins Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
2018-07-16Upgrade to ICU 62.1Eike Rathke
Change-Id: I9426e77aa85cfe068df59db47b8ac50b59cd4eb3 Reviewed-on: https://gerrit.libreoffice.org/57500 Reviewed-by: Eike Rathke <erack@redhat.com> Tested-by: Jenkins
2018-07-12Bin support for Visual Studio 2015, leaving just 2017 for nowTor Lillqvist
Change-Id: I6e604a8a463ff453b82ca60f1bf92b7c73870482 Reviewed-on: https://gerrit.libreoffice.org/56455 Tested-by: Jenkins Reviewed-by: Tor Lillqvist <tml@collabora.com>
2018-07-07Drop TEST_FONTS_MISSING for HAVE_MORE_FONTSKhaled Hosny
TEST_FONTS_MISSING is just adding confusion since it tests for hardcoded font names, while what we really want (most of time) is to check whether bundled fonts are enabled or not, and HAVE_MORE_FONTS just does that. Change-Id: I1b00d359c37fa49584ca27c319d6d0e937c3b20b Reviewed-on: https://gerrit.libreoffice.org/57136 Tested-by: Jenkins Reviewed-by: Khaled Hosny <khaledhosny@eglug.org>
2018-07-07Fix typoAndrea Gelmini
Change-Id: I5c364f44ca86f536b490313f0cce88c6823d0abd Reviewed-on: https://gerrit.libreoffice.org/56965 Reviewed-by: Julien Nabet <serval2412@yahoo.fr> Tested-by: Jenkins
2018-07-04Make brittle SortedAutoCompleteStrings ownership handling more explicitStephan Bergmann
Change-Id: Ieaf2231a84d97528bb1b9a410c4ee0c38966dd27 Reviewed-on: https://gerrit.libreoffice.org/56950 Tested-by: Jenkins Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
2018-07-04Add font mapping test for Linux Libertine GJan-Marek Glogowski
The vcl_fontfeature test needs a valid "Linux Libertine G" font, not a fallback font like DejaVuSans.ttf. Also fixes the test to work with spaces in font names. Change-Id: I6acad55c912c2ecab4b17ac7ab31021ad22d0cbc Reviewed-on: https://gerrit.libreoffice.org/56928 Tested-by: Jenkins Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
2018-07-04add --enable-gdb-index for --gdb-index from gold/lldLuboš Luňák
https://lists.freedesktop.org/archives/libreoffice/2018-June/080437.html Change-Id: I66904333bf329e804025d4b229a0db573e21c3af Reviewed-on: https://gerrit.libreoffice.org/56566 Tested-by: Jenkins Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
2018-07-04add configure option --enable-ld to use GNU gold or LLVM lldLuboš Luňák
https://lists.freedesktop.org/archives/libreoffice/2018-June/080437.html Change-Id: I06214459fcebe5cc58fd7979f3cbe5ac3d97db7d Reviewed-on: https://gerrit.libreoffice.org/56417 Tested-by: Jenkins Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
2018-07-04add --enable-split-debug for -gsplit-dwarfLuboš Luňák
https://lists.freedesktop.org/archives/libreoffice/2018-June/080437.html Change-Id: I2a02e23e46d7a54083249408f09fba87932b1d44 Reviewed-on: https://gerrit.libreoffice.org/56416 Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk> Tested-by: Jenkins Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
2018-07-04drop usage of --dynamic-list-cpp-new/typeinfoLuboš Luňák
First of all, lld doesn't have these options, but there doesn't seem to be any point in using them anyway. They are supposed to block the effect of -Bsymbolic-functions, but: - --dynamic-list-cpp-new matters only if we'd create our own global operator new/delete, which we don't - --dynamic-list-cpp-typeinfo affects only the typeinfo (_ZTI*) and typeinfo name (_ZTS*) symbols, which are not functions, and so -Bsymbolic-functions shouldn't do anything with them. According to https://sourceware.org/bugzilla/show_bug.cgi?id=3831 my understanding is that --dynamic-list-cpp-typeinfo actually predates -Bsymbolic-functions and it was an attempt to do the same from the other direction ('bind locally everything except for this' instead of 'bind locally only functions'). Change-Id: Iadad2d78f32a2adfb9c2100fb4eb5abe75725545 Reviewed-on: https://gerrit.libreoffice.org/56739 Reviewed-by: Luboš Luňák <l.lunak@collabora.com> Tested-by: Luboš Luňák <l.lunak@collabora.com>
2018-07-03we still need a Archive::Zip test in configure.acRene Engelhard
since system-hsqldb uses it for checking the hsqldb version. This partly reverts 4e3dc8c141c2efd037c1f214b5edff071812b6a8 Change-Id: I60d333d5807404355dd8c8fc2763e76d3a7fa350 Reviewed-on: https://gerrit.libreoffice.org/56888 Reviewed-by: Rene Engelhard <rene@debian.org> Tested-by: Rene Engelhard <rene@debian.org>
2018-06-29Fix calculations about version for javaTomáš Chvátal
This solves problem where java prints ton of warnings about too many CPUs which turned out to empty lines that are now ommited by the added sed command. Change-Id: I552f40289e1d598f11f98509b9e9b0b867336c3b Reviewed-on: https://gerrit.libreoffice.org/55377 Tested-by: Jenkins Reviewed-by: Tomáš Chvátal <tchvatal@suse.cz>
2018-06-29Avoid warnings in QT5_CLFAGS system headersStephan Bergmann
Change-Id: Ib22903c5168f6be3b0ebb973b0e7bab9d3daf93c Reviewed-on: https://gerrit.libreoffice.org/56670 Tested-by: Jenkins Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
2018-06-29python3: upgrade to release 3.5.5Michael Stahl
Fixes CVE-2017-1000158. Change-Id: I1871f031e3431253a3403f2ea2b586bcc4aa1016 Reviewed-on: https://gerrit.libreoffice.org/56609 Tested-by: Jenkins Reviewed-by: Michael Stahl <Michael.Stahl@cib.de>
2018-06-21Allow building KDE5 backend on HaikuKacper Kasper
Change-Id: I3138d26596af4c615fe2caaa2b160a76d8d2a352 Reviewed-on: https://gerrit.libreoffice.org/55548 Tested-by: Jenkins Reviewed-by: Thorsten Behrens <Thorsten.Behrens@CIB.de>
2018-06-20Make check for ant a bit more robustStephan Bergmann
...by cd'ing to an empty dir first. As reported in the mail thread starting at <https://lists.freedesktop.org/archives/libreoffice/2018-June/080400.html> "Ant not working ???", there appear to be Java toolchains that can be disturbed by filenames with "accented" (non-UTF8?) characters, which people may happen to put in their checked-out SRCDIR. Change-Id: I9cf3523f4816c3b0e2d6f67fc490c2514c8e1263 Reviewed-on: https://gerrit.libreoffice.org/56158 Tested-by: Jenkins Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
2018-06-18[Try to] get rid of dependency on X11Katarina Behrens
also kill KDE5SalDisplay with fire, it is not used anywhere anymore Change-Id: I4fe439b01322a6e3a809cf518e32d844c303daca
2018-06-15Don't croak on --without-langTor Lillqvist
Change-Id: I23e21b5d108fcc4bfc4422bebbc50389565e36fb Reviewed-on: https://gerrit.libreoffice.org/55864 Reviewed-by: Tor Lillqvist <tml@collabora.com> Tested-by: Tor Lillqvist <tml@collabora.com>
2018-06-13Respect --bindir..orbea
Change-Id: I2ed73a296b813de722156c69ac64cf9d6cff199d Reviewed-on: https://gerrit.libreoffice.org/55197 Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: Rene Engelhard <rene@debian.org>
2018-06-07Bump iOS SDK version to the current one, 11.4Tor Lillqvist
Not sure why we at some stage lost the possibiliy to build against any of several recent versions, but require one specific. But yeah, no big deal, anybody working on iOS code is expected to keep ther Xcode (and thus SDKs) updated.
2018-06-07tdf#116916 - Icon themes missing on LibreOffice Vanilla from macOS appstoreheiko tietze
Update of blacklist for $WITH_THEMES Fallback to Tango for ancient/unknown DE, Colibre only on Windows MPL vs. non-MPL on macOS Change-Id: Ibea9e9429a79911d632b54fa4aa9649003830aa3 Reviewed-on: https://gerrit.libreoffice.org/55295 Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: Heiko Tietze <tietze.heiko@gmail.com> Reviewed-on: https://gerrit.libreoffice.org/54794
2018-06-04Fix typo in the JAVA_HOME warningMuhammet Kara
Change-Id: I7ca28487f780e8d28198f98dc3702aa6cbdb1a11 Reviewed-on: https://gerrit.libreoffice.org/55262 Reviewed-by: Julien Nabet <serval2412@yahoo.fr> Tested-by: Jenkins <ci@libreoffice.org>
2018-06-02avoid configure warning about <codecvt> not usable by preprocessorLuboš Luňák
Change-Id: I710579f343eab38f576311bd15d931e5521105b2
2018-06-01--enable-gio also for --disable-gtk --enable-gtk3Stephan Bergmann
...as is used by the Flatpak build. Other settings only checked in that block controlled by ENABLE_GTK, and thus potentially erronously left empty for --disable-gtk --enable-gtk3 builds, are GTHREAD- and GTK_PRINT-related settings. GTHREAD_LIBS is explicitly used in vcl/Library_vclplug_gtk3.mk even, but it being empty apparently has no negative effects, so leave it at that for now. On the GTK_PRINT-related settings, Caolan commented on IRC: "ENABLE_GTK_PRINT is probably best removed entirely I guess, it didn't work out and they're redoing the dialog in upstream gtk again I'm told." Change-Id: I7e026c0ac9b23e7ace2c2e92390bdcc3be7d39a3 Reviewed-on: https://gerrit.libreoffice.org/55127 Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
2018-05-29micro-optimization: do not check for ant if we don't need itRene Engelhard
Change-Id: I84644cddd8a318a2af23780ada9afc26fe23a9fe Reviewed-on: https://gerrit.libreoffice.org/54719 Tested-by: Jenkins <ci@libreoffice.org> Tested-by: Rene Engelhard <rene@debian.org> Reviewed-by: Rene Engelhard <rene@debian.org>
2018-05-29add --enable-build-opensymbol option to build opens___.ttf from sourceRene Engelhard
... and handle opens___.ttf like unowinreg.dll (-> download.lst/$(TARFILE_LOCATION)) Change-Id: Id4723fb15c4b4b04133dedf97b1262441baa4756 Reviewed-on: https://gerrit.libreoffice.org/54938 Tested-by: Rene Engelhard <rene@debian.org> Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: Rene Engelhard <rene@debian.org>
2018-05-29WITH_HELPPACK_INTEGRATION isn't substituted anywhereStephan Bergmann
(just used via SCPDEFS) Change-Id: I23a14b0e76399b5e6303ad11ad7d09966d951b6f Reviewed-on: https://gerrit.libreoffice.org/54915 Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
2018-05-24bump product version to 6.2.0.0.alpha0+Christian Lohmaier
Change-Id: Id6f2b9036832e581aa6ccb45fd3f7a56191db980
2018-05-23actually add AC_MSG_CHECKING/_RESULTs for --enable/disable-postgresql-sdbcRene Engelhard
Change-Id: I6b02998db5624960ba78ce8b1b0356601a80fc5e Reviewed-on: https://gerrit.libreoffice.org/54721 Reviewed-by: Rene Engelhard <rene@debian.org> Tested-by: Rene Engelhard <rene@debian.org>
2018-05-23remove Linux ("UNX") systray "Quickstarter"Rene Engelhard
Change-Id: Ie0e8b8b7ad59ee640d6b195dfae1a7cf745056fd Reviewed-on: https://gerrit.libreoffice.org/54543 Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: Michael Stahl <Michael.Stahl@cib.de>
2018-05-23rename config_extension_update.h.in to config_extensions.h.inRene Engelhard
... and move HAVE_FEATURE_EXTENSIONS there, too Change-Id: I37fe52071e1db32cf5cc1e51dbd89bb4f32a39c1 Reviewed-on: https://gerrit.libreoffice.org/54072 Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: Rene Engelhard <rene@debian.org>
2018-05-22tdf#117342 - Add Karasa Jaga to LibreOffice coreRizal Muttaqin
Change-Id: Id121df0ddc763a299c5714cadd0d61740876e3d6 Reviewed-on: https://gerrit.libreoffice.org/54512 Reviewed-by: Heiko Tietze <tietze.heiko@gmail.com> Tested-by: Heiko Tietze <tietze.heiko@gmail.com>
2018-05-22Fix misleading comment: We do check HAVE_FEATURE_READONLY_INSTALLSET...Tor Lillqvist
...at least in a couple of places, and at least for a sandboxed macOS build that check does its job, it prevents write attempts into the app bundle. Change-Id: I6d243d3e8a4c77f03a8265125aefd75becbf0998
2018-05-13Icon themes clean-upheiko tietze
Industrial and Oxygen removed because of no maintainer, limited coverage, outdated design, and no SVG support Legacy themes will be available as extension Change-Id: I8f9c7bff6a180e3c151404b644a514eae4efcba8 Reviewed-on: https://gerrit.libreoffice.org/54158 Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: Heiko Tietze <tietze.heiko@gmail.com>
2018-05-11(try to) fix dbaccess firebird-related testsRene Engelhard
after 1c5c9b84df270e41c68e68bb69fbf183fc20eed2 Move headless check into test and add missing component file for firebird... Also check for ENABLE_FIREBIRD_SDBC... Change-Id: I27ea9fb08ebcce701a54686a8986e5071a8bcf96 Reviewed-on: https://gerrit.libreoffice.org/53940 Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: Michael Stahl <Michael.Stahl@cib.de>
2018-05-11Also for SDK 10.0.17134 the winsdklibsubdir needs to have a .0 tacked onTor Lillqvist
Also remove questionable old "Hack needed at least by tml". Change-Id: I478358ea114cee2f8a181b98cb3433447b3ed560 Reviewed-on: https://gerrit.libreoffice.org/54081 Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: Tor Lillqvist <tml@collabora.com>
2018-05-10sal: log windows traces to debugger console, take twoThorsten Behrens
This reverts most of commit 4c5b4752786ae2c174cd8fa8aa42b27a0994f34a and just keeps the OutputDebugStringA() call unconditionally. Change-Id: Ia838052b3147ca2d66c7d28b04eadd9700244669 Reviewed-on: https://gerrit.libreoffice.org/54029 Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
2018-05-10configure: fix gdrive,alfresco and onedrive id/secret diagnosticsRene Engelhard
e.g. GDRIVE_CLIENT_ID was checked with -z but it actually is "", so this always was outputted as "set". Change-Id: I9ee3e742655af721ba2b9d56e52d86b17331c65c Reviewed-on: https://gerrit.libreoffice.org/54057 Tested-by: Rene Engelhard <rene@debian.org> Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: Rene Engelhard <rene@debian.org>
2018-05-08sal: log windows trace output to debugger consoleThorsten Behrens
Change-Id: Ic8fea70fd3b0b2d4881cd30e3616f5bbf7c0c533 Reviewed-on: https://gerrit.libreoffice.org/53776 Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: Thorsten Behrens <Thorsten.Behrens@CIB.de>
2018-05-08Prevent potential issues mixing system/non-system libwps and libnumbertextStephan Bergmann
Change-Id: I1e47b3cc01df3d204ef7ef36f2432830aa18e65e Reviewed-on: https://gerrit.libreoffice.org/53978 Reviewed-by: Stephan Bergmann <sbergman@redhat.com> Tested-by: Stephan Bergmann <sbergman@redhat.com>
2018-05-07configure option to disable libnumbertextNoel Grandin
Because we get a crash in CppunitTest_sw_odfexport when compiling with clang >= 6.0.0 Also had to fix the ENABLE_LIBNUMBERTEXT being propogated, and the way the code checked it. Change-Id: I68e4a0078016d176f478ca4c1110fc296b6ca921
2018-05-07Allow building Qt5 backend on HaikuKacper Kasper
Change-Id: I7e928e9e29076bdfaaeffb83791bdc35f1952055 Reviewed-on: https://gerrit.libreoffice.org/53892 Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: Thorsten Behrens <Thorsten.Behrens@CIB.de>
2018-05-05Replace Galaxy icon theme by Colibreheiko tietze
Galaxy goes into an extension, fallback updated Change-Id: I4b6998035857f01368b39d4f16b2639d8a665065 Reviewed-on: https://gerrit.libreoffice.org/53092 Reviewed-by: andreas_kainz <kainz.a@gmail.com> Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: Heiko Tietze <tietze.heiko@gmail.com>
2018-05-04set -DENABLE_LIBNJUMBERTEXT after pkg-configRene Engelhard
so that we don't "loose" the pkg-config --cflags info. That also makes the #ifdef in lingucomponent/source/numbertext/numbertext.cxx not needed anymore Change-Id: Ia5771859a4a181f1a8a450d9ad4130b1e53abf8e
2018-05-03fix system-libnumbertext buildRene Engelhard
actually export SYSTEM_LIBNUMBERTEXT Change-Id: I758ef7ee868c5b06fc4e8ae43e6f2091925232aa
2018-05-03Remove unused --with-help=commonStephan Bergmann
...see mail thread starting at <https://lists.freedesktop.org/archives/libreoffice/2018-April/080050.html> "Anybody using --with-help=common?". Instead, make configure fail for unknown --with-help=... arguments. Change-Id: I51b8b9087a88ff3a3681453cdc830aa7e499e8df Reviewed-on: https://gerrit.libreoffice.org/53793 Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: Stephan Bergmann <sbergman@redhat.com>