summaryrefslogtreecommitdiff
path: root/vcl
AgeCommit message (Collapse)Author
2020-11-05Make findChild more robust in the face of disposed windowsStephan Bergmann
<https://ci.libreoffice.org/job/lo_ubsan/1815/> had once failed UITest_sw_styleInspector with > /vcl/source/window/window.cxx:3920:26: runtime error: member access within null pointer of type 'WindowImpl' > #0 0x2b883bedef80 in vcl::Window::get_id() const /vcl/source/window/window.cxx:3920:26 > #1 0x2b883f186817 in (anonymous namespace)::findChild(vcl::Window*, rtl::OUString const&, bool) /vcl/source/uitest/uiobject.cxx:399:18 > #2 0x2b883f185f86 in WindowUIObject::get_child(rtl::OUString const&) /vcl/source/uitest/uiobject.cxx:451:28 > #3 0x2b883f20e117 in UIObjectUnoObj::getChild(rtl::OUString const&) /vcl/source/uitest/uno/uiobject_uno.cxx:41:45 > #4 0x2b883f20e4c4 in non-virtual thunk to UIObjectUnoObj::getChild(rtl::OUString const&) /vcl/source/uitest/uno/uiobject_uno.cxx > #5 0x2b8879c1ceab in gcc3::callVirtualMethod(void*, unsigned int, void*, _typelib_TypeDescriptionReference*, bool, unsigned long*, unsigned int, unsigned long*, double*) /bridges/source/cpp_uno/gcc3_linux_x86-64/callvirtualmethod.cxx:77:5 > #6 0x2b8879c171c2 in cpp_call(bridges::cpp_uno::shared::UnoInterfaceProxy*, bridges::cpp_uno::shared::VtableSlot, _typelib_TypeDescriptionReference*, int, _typelib_MethodParameter*, void*, void**, _uno_Any**) /bridges/source/cpp_uno/gcc3_linux_x86-64/uno2cpp.cxx:233:13 > #7 0x2b8879c13d0d in unoInterfaceProxyDispatch /bridges/source/cpp_uno/gcc3_linux_x86-64/uno2cpp.cxx:413:13 > #8 0x2b887efde9fa in binaryurp::IncomingRequest::execute_throw(binaryurp::BinaryAny*, std::__debug::vector<binaryurp::BinaryAny, std::allocator<binaryurp::BinaryAny> >*) const /binaryurp/source/incomingrequest.cxx:235:13 > #9 0x2b887efd861a in binaryurp::IncomingRequest::execute() const /binaryurp/source/incomingrequest.cxx:78:26 > #10 0x2b887f0b5367 in request /binaryurp/source/reader.cxx:85:9 > #11 0x2b880b8b0144 in cppu_threadpool::JobQueue::enter(void const*, bool) /cppu/source/threadpool/jobqueue.cxx:100:17 > #12 0x2b880b8d3f6b in cppu_threadpool::ORequestThread::run() /cppu/source/threadpool/thread.cxx:165:31 > #13 0x2b880b8dfbcf in threadFunc /include/osl/thread.hxx:185:15 > #14 0x2b88025206e8 in osl_thread_start_Impl(void*) /sal/osl/unx/thread.cxx:264:9 > #15 0x2b880438be64 in start_thread (/lib64/libpthread.so.0+0x7e64) > #16 0x2b8804cc188c in clone (/lib64/libc.so.6+0xfe88c) apparently because pParent was non-null but already disposed (so mpWindowImpl had already been cleared). I don't know whether this situation can legitimately happen, or was fallout from an actual error that happened earlier (and went unnoticed), but this is apparently uitest-only code, so just make it more robust and be done with it. Change-Id: I178fcf2fb70bf29bc6bbb246ebff5b1933579e5a Reviewed-on: https://gerrit.libreoffice.org/c/core/+/105326 Tested-by: Jenkins Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
2020-11-05vcl pdf tokenizer: add real numbers testcaseMiklos Vajna
Fails with commit 0d68738d67eacdfebdca3c9183dc11f953b38174 (Correctly parse real numbers in PDFs, 2020-09-25) reverted. Change-Id: I449ed781a3dd4b78e27a3fd2e135495e7d6c35ae Reviewed-on: https://gerrit.libreoffice.org/c/core/+/105297 Reviewed-by: Miklos Vajna <vmiklos@collabora.com> Tested-by: Jenkins
2020-11-05SpinButtonEntry is unusedNoel
since commit f2517e82904b92989ed7e38a070c18234f460b33 Author: Noel Grandin <noel.grandin@collabora.co.uk> Date: Tue Sep 8 22:11:11 2020 +0200 improve loplugin:unusedvarsglobal Change-Id: Ibfa5a651eae01aeecd77a46d9f949ae6e48e17fe Reviewed-on: https://gerrit.libreoffice.org/c/core/+/105304 Tested-by: Jenkins Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
2020-11-05INPUTENABLE is unusedNoel
since commit be3b1a8393ca3e77f31bea5d9caa7271fd53e9e4 Author: Caolán McNamara <caolanm@redhat.com> Date: Mon Aug 17 17:28:58 2020 +0100 weld OAppDetailPageHelper Change-Id: I97afdbd536c587a360b8f72b9faca71e8fa32ba3 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/105306 Tested-by: Jenkins Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
2020-11-05N8BitTcMask is unusedNoel
since commit 77809fba7d4bf5e0b604ffa3937c18d5530c2d56 Author: Luboš Luňák <l.lunak@collabora.com> Date: Fri Oct 9 19:28:49 2020 +0200 implement ImplFastBitmapConversion() for 8bit gray source Change-Id: I730aef10c1705ae6d3141014228cf79b8bfa3c63 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/105305 Tested-by: Jenkins Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
2020-11-04Fix typosAndrea Gelmini
Change-Id: If310f08c8df54356b52bb9c0ac614c374c750814 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/105268 Tested-by: Jenkins Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
2020-11-04Fix typosAndrea Gelmini
Change-Id: I8a7a313b54aafc5347e00c5afec4648e09dbc981 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/105295 Tested-by: Jenkins Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
2020-11-04sfxlo-SidebarToolBox not used in any current .ui filesCaolán McNamara
map sfxlo-SidebarToolBox to sfxlo-NotebookbarToolBox in case it appears in any stale customized profiles Change-Id: I6035b9042450630b23b48a895e4e9d0ff8d730a6 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/105294 Tested-by: Jenkins Reviewed-by: Caolán McNamara <caolanm@redhat.com>
2020-11-04negative returnsCaolán McNamara
Change-Id: I6f710f1aecc2e242b6006a3360e31bf2a9438fe7 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/105286 Tested-by: Jenkins Reviewed-by: Caolán McNamara <caolanm@redhat.com>
2020-11-04resource leakCaolán McNamara
Change-Id: I1abd8acef55d5bdb4744ecf1a62d8e1396de0e3f Reviewed-on: https://gerrit.libreoffice.org/c/core/+/105287 Tested-by: Jenkins Reviewed-by: Caolán McNamara <caolanm@redhat.com>
2020-11-04loplugin:unusedmethodsNoel
Change-Id: I58fb6477222a26455f5db0a7d972c46f290c085c Reviewed-on: https://gerrit.libreoffice.org/c/core/+/105261 Tested-by: Jenkins Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
2020-11-03vclplug_qt5 doesn't appear to need to link to vclplug_x11Caolán McNamara
Change-Id: I7f4c3fb210996c0ce66d3d9448a3954ce1aaa525 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/105256 Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de> Reviewed-by: Caolán McNamara <caolanm@redhat.com> Tested-by: Jenkins
2020-11-03loplugin:unusedfieldsNoel
Change-Id: Ia8e2a7ce75bfded98e85fa583d1404710069d335 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/105249 Tested-by: Jenkins Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
2020-11-02pdf: deduplicate resources when copying from external PDF streamTomaž Vajngerl
When using external PDF stream/data (from PDF graphic objects), make sure to copy the content of external PDF resources (fonts, bitmaps, forms) only one time (by sharing the map between calls) and every other use, just use the reference to the objects. Change-Id: Ibaa632c8f74806eb195e69404551db6fd077a986 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/104935 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com> Reviewed-by: Miklos Vajna <vmiklos@collabora.com> Reviewed-on: https://gerrit.libreoffice.org/c/core/+/105163 Tested-by: Jenkins Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>
2020-11-02pdf: test pdf re-export that bitmaps aren't duplicatedTomaž Vajngerl
Change-Id: I6956648bf0f9ebb1221b7e36d5fc1616c6b8b804 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/105162 Tested-by: Tomaž Vajngerl <quikee@gmail.com> Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>
2020-11-02Print dialog alignment fixesandreas kainz
Change-Id: Ife471e062745ce8aa7dd9b806612c00794e4d878 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/105159 Tested-by: Jenkins Reviewed-by: Andreas Kainz <kainz.a@gmail.com>
2020-11-02Better types for SystemGraphicsData::hDrawable, cairo::X11SysData::hDrawableStephan Bergmann
Both apparently only hold X11 Drawable values. But include/vcl/sysdata.hxx does not include the X11 include files that would define Drawable, so use sal_uIntPtr there, which is also used for other similar SystemEnvData and SystemParentData aWindow members there. Change-Id: Ia136ad1937e2009eb409c3ef8cc41dc81bd338a4 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/105156 Tested-by: Jenkins Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
2020-11-02toolbox always has flat styleNoel Grandin
ever since commit c1cd6af623e86b5b1b45f9d09dc17d6fbb907f02 Author: Jens-Heiner Rechtien <hr@openoffice.org> Date: Mon May 10 14:51:02 2004 +0000 INTEGRATION: CWS nwf (1.64.66); FILE MERGED 2004/03/31 09:28:33 ssa 1.64.66.1: #i25130# force flat toolbox buttons except for a completely and utterly undocumented hack of a registry key, introduced in commit 736dc0956a50315ec72ad126406556657a750d37 Author: Rüdiger Timm <rt@openoffice.org> Date: Thu Apr 17 14:19:46 2003 +0000 INTEGRATION: CWS vcl08 (1.57.2.4.18); FILE MERGED 2003/04/14 17:46:27 ssa 1.57.2.4.18.1: #108699# disabled flat toolbox buttons not exported anymore which only seems to apply to Windows. So just remove this. Change-Id: Idf315b8c89c3119883a5e6880d003d379fe6faec Reviewed-on: https://gerrit.libreoffice.org/c/core/+/105155 Tested-by: Jenkins Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
2020-11-01Transport Pixmap from GetNativeSurfaceHandle via css::uno::Any as sal_Int64Stephan Bergmann
The values returned from GetNativeSurfaceHandle are only used in cairocanvas::CanvasBitmap::getFastPropertyValue (canvas/source/cairo/cairo_canvasbitmap.cxx), whose consumers in turn are (at least the ones I could identify): * OGLTransitionerImpl::setSlides (slideshow/source/engine/opengl/TransitionerImpl.cxx), passing the values on to X11SalObject::SetLeaveEnterBackgrounds, which extracts the Pixmap value to pass it on to XSetWindowBackgroundPixmap. * X11SalBitmap::Create (vcl/unx/generic/gdi/salbmp.cxx), which extracts the Pixmap value to pass it on to X11SalBitmap::ImplCreateFromDrawable (as a Drawable) and to XFreePixmap. While Pixmap XIDs are apparently 32-bit in the X11 protocol, see e.g. <https://www.x.org/releases/X11R7.7/doc/xproto/x11protocol.html>, Xlib and the underlying /usr/include/X11/Xdefs.h appear to be somewhat confused whether the typedef for XID should be unsigned long (presumably stemming from times when long was universally 32-bit) or CARD32, see e.g. <https://gitlab.freedesktop.org /xorg/proto/xorgproto/-/blob/master/include/X11/Xdefs.h>. So conservatively stick to a 64-bit type here, even if it should only ever contain 32-bit values. Change-Id: I26c3a2ff74cef092042a6e3648cd9a6f4a9e3aac Reviewed-on: https://gerrit.libreoffice.org/c/core/+/105144 Tested-by: Jenkins Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
2020-10-31X11Pixmap::mhDrawable only ever holds a PixmapStephan Bergmann
...(passed in in X11Surface::getSimilar at vcl/unx/generic/gdi/cairo_xlib_cairo.cxx:226), so use that for its type. (Using an appropriate UNO type when passing it into an any in X11SalGraphics::GetNativeSurfaceHandle will be addressed in a later commit; lets stick to tools::Long there for now.) Change-Id: I047a5f078c4b3c45a8fdc0fb09d544c73a6a4e29 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/105087 Tested-by: Jenkins Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
2020-10-30remove pimpl from HyperLabelNoel Grandin
Change-Id: Iaf156065b85a0a809946236148a85d768ca5f51b Reviewed-on: https://gerrit.libreoffice.org/c/core/+/105026 Tested-by: Jenkins Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
2020-10-30convert some tools::Long->sal_Int32Noel
in places where it is obvious we only need a sal_Int32, because we are dealing with rows and columns, and not even calc needs more than 32 bits for that. Change-Id: I114417e639c224d45bfd9fc6838122ab195eefa3 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/104584 Tested-by: Jenkins Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
2020-10-29tdf#137693 fix sal tree net line renderingJim Raykowski
makes tree net lines for sal backends work as intended Change-Id: Ib7f32a147215635c89f3e4073541c9059765c995 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/104709 Tested-by: Jenkins Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
2020-10-29Speed improments191919
Improve drawing performance on 10-bit displays by avoiding multiple color space conversions. Make all drawing context bitmaps have the colour space and byte order of the host window. Fix serious CoreGraphics-related memory leak (existed no later than version 7.0) when changing the window size. Change-Id: Ia7b7e88d47b728bd1d10dedc1ca222215de41e73 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/104858 Tested-by: Jenkins Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
2020-10-29ofz#26708: Direct-leakCaolán McNamara
since... commit cd8dd48bf89d800c012afb04b16281ed8b261de9 Date: Tue Aug 21 16:07:55 2018 +0200 tdf#119282 Area tab, Pattern color change Change-Id: If0c988f7532786a4089edef39929f924d8e2ec2d Reviewed-on: https://gerrit.libreoffice.org/c/core/+/104976 Tested-by: Jenkins Reviewed-by: Caolán McNamara <caolanm@redhat.com>
2020-10-29tdf#137643 alternative solution to activate embedded fonts in one batchCaolán McNamara
Change-Id: Ib5ffb2b8a31f237d5d2e465bf3f777590e0bfade Reviewed-on: https://gerrit.libreoffice.org/c/core/+/104947 Tested-by: Jenkins Reviewed-by: Caolán McNamara <caolanm@redhat.com>
2020-10-29tdf#137556 max-content-height needs gtk >= 3.22Caolán McNamara
where the 7.0 baseline is 3.18 and 7.1 is 3.20 Change-Id: I73adb9bc3cf25112925e663f8fc0c3e5191bbf6a Reviewed-on: https://gerrit.libreoffice.org/c/core/+/104971 Reviewed-by: Heiko Tietze <heiko.tietze@documentfoundation.org> Reviewed-by: Caolán McNamara <caolanm@redhat.com> Tested-by: Jenkins
2020-10-29tdf#137506: Move single control in 'More Options' to print dialogAyhan Yalçınsoy
Change-Id: Iacefe276bf174b96118015b48e746d1591b93c1d Reviewed-on: https://gerrit.libreoffice.org/c/core/+/104805 Tested-by: Jenkins Reviewed-by: Heiko Tietze <heiko.tietze@documentfoundation.org>
2020-10-28print dialog has one expander "more" and one "More"Caolán McNamara
since... commit 8eff280bc08ec3d7b2312ae4ee48df4d7328b7de Date: Fri Oct 2 12:35:26 2020 +0200 ui files: fix some capitalisation issues so make the other one capitalized to match Change-Id: I7eeac239ea99b85b310354e3a0c6dddf15ceff04 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/104973 Tested-by: Jenkins Reviewed-by: Caolán McNamara <caolanm@redhat.com>
2020-10-28tdf#137643 Revert "lock refreshing font data when loading a document"Caolán McNamara
from tdf#69060, to replace with an alternative solution This reverts commit 98d71c4e0847797a4ba9229a8e6d832a8a3d5e0f. and This reverts commit 64d8e5f8db70f4f913abb902b41f4cff8dd1cdad. Change-Id: I384e994b54aa1bfc735c6ab591b9b1410058451f Reviewed-on: https://gerrit.libreoffice.org/c/core/+/104716 Tested-by: Jenkins Reviewed-by: Caolán McNamara <caolanm@redhat.com>
2020-10-28Related: tdf#137620 focus-in IMHandler on grab_focusCaolán McNamara
Change-Id: I9fa980f500cd11b34e349866e62f9f2399279e5b Reviewed-on: https://gerrit.libreoffice.org/c/core/+/104932 Reviewed-by: Justin Luth <justin_luth@sil.org> Reviewed-by: Caolán McNamara <caolanm@redhat.com> Tested-by: Jenkins
2020-10-28Use UCB instead of cURL to download https filesStephan Bergmann
Since 7dc6fc32eb618da6defb8a9f330978fa019677b8 "uitest: Check the more icons dialog opens" started to test the AdditionsDialog code, some ASan builds started to report use-after-poison from within SearchAndParseThread::execute -> curlGet -> curl_easy_perform -> ... -> https_connecting -> ... -> secmod_ModuleInit -> pemC_Initialize -> ..., see the comments starting at <https://gerrit.libreoffice.org/c/core/+/98226/ 10#message-2a55980ab2477e41dc7515e4aeabc7234afc2053> "tdf#133026: Tight integration of extensions - Adding thread structure". The exact cause for that ASan use-after-poison is not quite clear to me. First, I thought it was due to liberal use of curl_easy_init (without a central curl_global_init) in a multi-threaded process; see <https://curl.haxx.se/libcurl/c/curl_easy_init.html> for this being considered "lethal". But then, another issue could be that most of the nss libraries like instdir/program/libnss3.so (implementing the "... -> secmod_ModuleInit" part) come from LO's bundled --without-system-nss while /lib64/libnsspem.so (implementing the "pemC_Initialize -> ..." part) comes from the system, and there could be some mismatch when mixing these (esp. if the former are built with ASan). So whatever the actual cause, the right fix should be to use LO's UCB instead of directly calling into cURL anyway. That required a version of utl::UcbStreamHelper::CreateStream that uses a given XInteractionHandler (which may be null to indicate no interaction handling, instead reporting all interaction requests as exceptions) instead of internally creating an XInteractionHandler using the GUI, and which would cause deadlock in 7dc6fc32eb618da6defb8a9f330978fa019677b8's UITest_sw_options. (And introducing that additional utl::UcbStreamHelper::CreateStream overload required css::awt::XWindow to be a complete type now in vcl/source/graphic/GraphicLoader.cxx, for > include/com/sun/star/uno/Reference.h:277:18: note: in instantiation of variable template specialization 'std::is_base_of_v<com::sun::star::task::XInteractionHandler, com::sun::star::awt::XWindow>' requested here > std::is_base_of_v<interface_type, derived_type> > ^ ) (The removed code in cui/source/dialogs/AdditionsDialog.cxx should have been the last remaining use of curl in Library_cui. Apparently, e1e9e2aa16f08a8fd9e38db706d63e8eceeda8d1 "Kill Mozilla personas" had forgotten to remove it from cui/Library_cui.mk the last time Library_cui had become curl-free, before the introduction of the AdditionsDialog code.) (I did not remove the #undef ABSOLUTE FIXME from cui/source/dialogs/AdditionsDialog.cxx. c4bee547b02fbe3d07b1e9ee203c73e48f86e6bf "tdf#133026: Additions: Better Search Function" does not tell whether it had been added to mitigate a macro definition from the (now removed) #include <curl/curl.h>.) Change-Id: I1fec7488d36df81c3543d12d97da1291f77712fc Reviewed-on: https://gerrit.libreoffice.org/c/core/+/104938 Tested-by: Jenkins Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
2020-10-28Fix build for UNX x86Thorsten Behrens
Change-Id: I971aa787a33fa3a6bd9913abb94c90ddf53109cf Reviewed-on: https://gerrit.libreoffice.org/c/core/+/104937 Tested-by: Jenkins Reviewed-by: Thorsten Behrens <Thorsten.Behrens@CIB.de>
2020-10-28convert some more long -> tools::LongNoel
Change-Id: Ide9811c1a7582454b3fcf655b70ea106ed56509a Reviewed-on: https://gerrit.libreoffice.org/c/core/+/104914 Tested-by: Jenkins Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
2020-10-28pdfium: add an FPDF_RenderPageBitmap() wrapperMiklos Vajna
Change-Id: I2238aeb4b2f7483440ded2600a328436f861f178 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/104917 Reviewed-by: Miklos Vajna <vmiklos@collabora.com> Tested-by: Jenkins
2020-10-28pdfium: add an FPDFBitmap_FillRect() wrapperMiklos Vajna
Change-Id: I4201a9ead756cf814c8452ef504acd054c4fe38b Reviewed-on: https://gerrit.libreoffice.org/c/core/+/104908 Tested-by: Jenkins Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
2020-10-28convert some more long -> tools::LongNoel
grepping for stuff in template params this time Change-Id: Ia37bfd85480b3a72c3c465489581d56ad8dde851 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/104855 Tested-by: Jenkins Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
2020-10-28std::set->o3tl::sorted_vector in vclNoel Grandin
Change-Id: Ie486b629b4074da5121b55c76965aeb8ea057f31 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/104904 Tested-by: Jenkins Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
2020-10-27Qt5 std::abort, if we're unable to paintJan-Marek Glogowski
If this crashes LO too often, we can revert it. It helped me to debug tdf#137804. Then we might need to work with exceptions or some other way to notifiy the creator of the Qt5Painter, that the constructor has failed. Much more work to handle all sites using Qt5Painter... Change-Id: I81cd18770dc93d8dd1542972cd4fc220dd0dce09 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/104894 Tested-by: Jenkins Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
2020-10-27tdf#133950 scale hotspot to sync with scaled cursorsCaolán McNamara
keep aspect ratio as well Change-Id: Ia3c7f84e377ebb952be0a56e9e532fb260f83459 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/104890 Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk> Reviewed-by: Caolán McNamara <caolanm@redhat.com> Tested-by: Jenkins
2020-10-27tdf#137804 Qt5 don't paint empty stringsJan-Marek Glogowski
For a test I added a SAL_WARN, but that created many more debug messages then expected, just when opening the print dialog. So this just handles the crash like the CairoTextRender variant of DrawTextLayout does, by returning early. Change-Id: Ie5b58ebbba02dace0e777b60fcbc3b9f46ea9d66 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/104893 Tested-by: Jenkins Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
2020-10-27Resolves: tdf#137620 add DeleteSurroundingText at vcl::Window levelCaolán McNamara
a) give it a default implementation based on the current one b) re-use code introduced for WeldEditView::DeleteSurroundingText for the EditView containing vcl::Window in impress/draw and various similar Annotation windows Change-Id: I55547c70e90ee394795b5545450cf8131538fad8 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/104781 Tested-by: Jenkins Reviewed-by: Caolán McNamara <caolanm@redhat.com>
2020-10-27tdf#137413 EMF import: fix transparency in the PDF fallback caseMiklos Vajna
Commit d75c5b38911557173c54a78f42ff220ab3918573 (tdf#136836 emfio: speed up import of EMF import when the orig PDF is available, 2020-09-17) improved both performance and correctness of the EMF import, in case it had a PDF fallback. It turns out that PDF fallback can be nominally non-transparent, and still the EMF equivalent supports transparency. Fix the problem by enabling transparency in the PDF-in-EMF case. Change-Id: I4d1585a5db6f28bd9c9cb380b5f193f4d5edcc8d Reviewed-on: https://gerrit.libreoffice.org/c/core/+/104849 Reviewed-by: Miklos Vajna <vmiklos@collabora.com> Tested-by: Jenkins
2020-10-27tdf#136555 apply window bg color for button boxesJan-Marek Glogowski
The default background color of the frame is APPBACKGROUND, which is matched to GetWorkspaceColor(). That color is visible in Writer as the default document workspace background. Originally I just saw the dark background of the button box in the start center, and thought the workspace color in Qt is wrong and changed it in the reverted commit, missing the change of the color around the Writer document. Now that the button boxes's background color isn't fixed via the "branding" setting, we also must react to changed style settings, so we override the DataChanged function. This also reverts commit e0c72e31c1d455c26110c35e8780d420e17cdea6. Regressed-by: a927e0964ba0442d53fffb22c577e54bcf183ed7 Change-Id: I62396355054523baef1197a8a8af61c2f0d29ef3 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/104740 Tested-by: Jenkins Reviewed-by: Thorsten Wagner <thorsten.wagner.4@gmail.com> Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
2020-10-25Related: tdf#137620 use existing SalEvent::SurroundingTextRequestCaolán McNamara
for signalIMRetrieveSurrounding Change-Id: Ic3280918283e2e1cf63a6d30767949d1c59cd1c2 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/104780 Tested-by: Jenkins Reviewed-by: Caolán McNamara <caolanm@redhat.com>
2020-10-25Related: tdf#137620 factor out working IMDeleteSurrounding hunkCaolán McNamara
Change-Id: I7504e8670fb464fe3e76a3d38f1cac39ba2e9208 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/104779 Tested-by: Caolán McNamara <caolanm@redhat.com> Reviewed-by: Caolán McNamara <caolanm@redhat.com>
2020-10-24only sfxlo and svtlo still existCaolán McNamara
Change-Id: Id1138c2eb9f7c551d9d2855a68a364e3e21306cd Reviewed-on: https://gerrit.libreoffice.org/c/core/+/104748 Tested-by: Jenkins Reviewed-by: Caolán McNamara <caolanm@redhat.com>
2020-10-24crashtesting: avoid assert on loading tdf98681-3.docxCaolán McNamara
these limits are possibly totally archaic and maybe should be removed entirely given the number of freetype releases since their introduction. Just widen the range here. Change-Id: Ic6648bbcd0a996ab342ca4c310850cf81bd504eb Reviewed-on: https://gerrit.libreoffice.org/c/core/+/104733 Tested-by: Jenkins Reviewed-by: Caolán McNamara <caolanm@redhat.com>
2020-10-24Show animated GIFs in TotDHeiko Tietze
marchingants.gif added as an example Change-Id: I555aebc4a4e191b8afb4be06dfc1e5f3e6d1a911 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/104718 Tested-by: Jenkins Reviewed-by: Heiko Tietze <heiko.tietze@documentfoundation.org>
2020-10-23log why a FreetypeFont was flagged as !mbFaceOkCaolán McNamara
Change-Id: I6f474c1761e7b0d77a0c5516dd77a60e98c75996 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/104732 Tested-by: Caolán McNamara <caolanm@redhat.com> Reviewed-by: Caolán McNamara <caolanm@redhat.com>
generated by cgit (git 2.34.1) at 2025-02-07 07:16:46 +0000