summaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2020-01-16unoxml: CDocumentBuilder::parseURI should handle non-file:/// URIs.Michael Meeks1-4/+19
The proximate symptom of this is of only some of the slide layouts applying in impress on Android. This is caused by not parsing the file:///assets/.../layoutlist.xml - which needs to use UCB and its cleverer osl/ file APIs. Change-Id: I22ed77170891c0ec136caaa29da69987a0e51a73 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86899 Reviewed-by: Ashod Nakashian <ashnakash@gmail.com> Tested-by: Ashod Nakashian <ashnakash@gmail.com>
2020-01-15lok: ui: more files to increase the 'step-increment'Henry Castro49-94/+94
property to 5. In mobile view, touch spin button should increment by 5 Change-Id: Ib3c8966443d053ab0588fa0c124f0daae31b7ed8 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86844 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com> Reviewed-by: Henry Castro <hcastro@collabora.com>
2020-01-15lok: ui: increase the 'step-increment' property to 5Henry Castro1-1/+1
Change-Id: I1090a9f6435619f095ca22ef401c5e1bc9741d06 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86798 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com> Reviewed-by: Henry Castro <hcastro@collabora.com>
2020-01-14fix crash in draw Format->Bullets and Numbering dialogNoel Grandin1-6/+9
regression from commit a2b78c131db2bbe3555a89dfae32aefcd00949f5 make Format->Bullets and Numbering dialog async Change-Id: Ia726612a492e644f5e40e78a3b537c5ed539aef8 Reviewed-on: https://gerrit.libreoffice.org/66423 Reviewed-by: Katarina Behrens <Katarina.Behrens@cib.de> Tested-by: Jenkins (cherry picked from commit 84acbfd2ef8a895d8689fccfffce03df02240b4e) Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86784 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com> Reviewed-by: Aron Budea <aron.budea@collabora.com> Tested-by: Aron Budea <aron.budea@collabora.com>
2020-01-13lok: Unlock chart contollers after insertion on mobile viewMuhammet Kara1-0/+2
The chart controllers were locked until the dialog finishes its execution, but at some point, dialog is disabled for mobile view, and seems like it was forgotten to unlock the controllers. Change-Id: Ib1e5799f5c84ef8b3ebc0f538909b4f8b586ff1a Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86716 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com> Reviewed-by: Michael Meeks <michael.meeks@collabora.com>
2020-01-13mobile: escape JSON for function wizard, and emit only on mobile.Michael Meeks1-5/+16
Otherwise typing "-> wa" into a cell gives JS exceptions. Change-Id: I5303186fe9f84e055d65cf1929ccf5a7a20cae35 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86705 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com> Reviewed-by: Michael Meeks <michael.meeks@collabora.com>
2020-01-13Fix FontColor does not change selected cells on calcMert Tumer1-2/+24
Change-Id: Iae2f72bff4fd6986fc8cc07ba09996b1af4eb140 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86670 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com> Reviewed-by: Michael Meeks <michael.meeks@collabora.com>
2020-01-13WaE: unused code in calc.Michael Meeks1-3/+0
Change-Id: I42a895573f8871841c9a33a23c0b826be814ddae
2020-01-13Merge both x64 and x86 CRT merge modules into x64 MSIMike Kaganski1-39/+16
... because it contains x86 binaries in addition to x64: e.g., twain32shim, spsupp_x86. The opposite (inclusion of x64 MSM into x86 MSI) is not possible ( see https://stackoverflow.com/questions/58181986 ), so do it just for x64 MSI. Change-Id: I3935fce751b1b6d04291fede6b82be25fe541582 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86667 Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com> Tested-by: Mike Kaganski <mike.kaganski@collabora.com> (cherry picked from commit a4681e8b74a59439110af42ff18cce021512056d) Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86680 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
2020-01-13WaE: unused pLockData.Michael Meeks1-0/+1
Change-Id: I3c074405f8645caf889f145a6f09f9f680d5a6b8
2020-01-12tdf#107604 Handle predefined table styles.Gülşah Köse5-139/+950
Handle table styles that we have just style-id and have not corresponding style defination in tableStyle.xml Change-Id: Id06f5508d5eb1330b5bac1634c520c3066c09139 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86068 Tested-by: Jenkins Reviewed-by: Andras Timar <andras.timar@collabora.com> Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86327 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
2020-01-11Use registry key to track installed state of gm_SharePointSupport_SubstMSOMike Kaganski1-4/+15
Without that, after initial installation, the state of the feature is undefined, and following uninstallation can't define that it needs to be uninstalled, unregistering SharePoint.OpenDocuments class. Change-Id: Ib7455833fb397c332735eb4c8ab63f763b4e469b Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86598 Tested-by: Jenkins Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com> (cherry picked from commit 01fa2b022e4d5b7392b02181d9bb9bfc76272d62) Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86601 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
2020-01-11Improve logging of feature statesMike Kaganski1-0/+9
Change-Id: I8b304bbefe861abd7503204f0ae8efb0aea203ac Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86528 Tested-by: Jenkins Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com> (cherry picked from commit 2bb064943444cd5bac155ad367378b674bceb11a) Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86545 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
2020-01-11Register spsupp*.dll during installationMike Kaganski8-6/+432
This registers SharePoint integration libraries using regsvr.exe. Both 32-bit and 64-bit libraries are registered; registration of LOSPSupport.OpenDocuments is unconditional. This introduces a new hidden MSI feature, which is disabled for installation: gm_SharePointSupport_SubstMSO. When installed, it registers SharePoint.OpenDocuments class in registry, thus overriding registration of this component by MS Office, allowing LibreOffice to serve as MS Office replacement working in IE with SharePoint. To install the feature, either a transform is needed setting the feature's level <= 100, or a command line: msiexec path-to-msi ADDLOCAL=gm_SharePointSupport_SubstMSO Change-Id: I5517bbb68dcc6db8bcb2bbc2368394ee4a62d741 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86452 Tested-by: Jenkins Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com> Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86462 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
2020-01-10lok: jsdialog creation for sidebar panels cleanup.Michael Meeks1-6/+13
Use UNO panel names as ids on new Panel nodes. This avoids problems with over-writing child details, items like sd::PanelLayout don't have an empty parent to overwrite, but have an immediate valueset node. Change-Id: I00bab7f0d6a4fb247e0509bce7548b2da164bd23 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86571 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com> Reviewed-by: Michael Meeks <michael.meeks@collabora.com>
2020-01-10tdf#128207: DOCX import: fix chart positioningBakos Attila5-0/+17
Embedded graphic objects had got 0 values for vertical and horizontal positioning before, resulting overlapping, hidden charts, but now they are positioned according to the values in the document. (cherry picked from commit d9c535ead688e9f156dbcf43948df08a69e218be) Conflicts: sw/qa/extras/ooxmlexport/ooxmlexport14.cxx [ Miklos: reworked the testcase to use the UNO API, the other way would only work on master. ] Change-Id: Ia5403ac65ff7192d61072e8a9d8a7f80c7178b9b Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86557 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com> Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
2020-01-10lok: avoid emission storms of un-necessary invalidations.Michael Meeks2-3/+8
Common when constructing widgets with VclBuilder - which avoids the 'Show' detection Pranav introduced in 8de98e61fbc. This saves ~80% of the ~100k mostly bogus calls I get to: desktop::CallbackFlushHandler::processWindowEvent when opening and closing a few windows. Change-Id: Ie508d6e19274472b85543275aee33f078ddcbbb2 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86533 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com> Reviewed-by: Michael Meeks <michael.meeks@collabora.com>
2020-01-10Fix --enable-werror buildMike Kaganski1-51/+51
... producing this error in MSVC2017: C:/lo/src/core2/desktop/source/lib/init.cxx(3486): error C2220: warning treated as error - no 'object' file generated C:/lo/src/core2/desktop/source/lib/init.cxx(3486): warning C4190: 'getSidebarWindow' has C-linkage specified, but returns UDT 'VclPtr<vcl::Window>' which is incompatible with C C:\lo\src\core2\include\vcl/cursor.hxx(49): note: see declaration of 'VclPtr<vcl::Window>' ... because getSidebarWindow was inside 'extern "C"' block. Change-Id: If7364555d09f0955ac90ce127b19ce4bbdb5cc3c Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86518 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com> Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
2020-01-09lok: simplify jsdialog creation for sidebar panels.Michael Meeks2-0/+39
Remove some layers of nested and/or un-necessary content, also simplifies client-side JS. Change-Id: I67347035ceb9dbee9c62c99624b5084883d4e61a Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86498 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com> Reviewed-by: Michael Meeks <michael.meeks@collabora.com>
2020-01-09Bump version to 6.2-3cp-6.2-3CODE-4.2.0-3Andras Timar1-1/+1
Change-Id: Iab6ee2f21a0d13230faa2b510ad3af9d6ba3c519
2020-01-09sw: tolerate null ShadowOverlay object as the rest of the code does.Michael Meeks1-1/+1
Change-Id: I552c0f11b22212d36bcaee33432e2a10d4df58ac Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86472 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com> Reviewed-by: Michael Meeks <michael.meeks@collabora.com>
2020-01-09tdf#129205 DOCX import: handle the <w:shd w:val="nil" ...> paragraph propertyMiklos Vajna4-1/+77
Reading the spec, "nil" is the opposite of "clear": i.e. if the (background) color is red and the fill (color) is green, then "clear" means green. And you would expect "nil" means red, but it's just nothing in Word. Fix the problem by doing the same: don't set any paragraph property for the "nil" case and keep doing it for the common "clear" case. (cherry picked from commit fbe7612d654be9dfe1ea6f2e67900eb4eec4202a) Change-Id: I30af8a7fb55fb9bab2d12e120069a479fc7ab0a9 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86098 Tested-by: Jenkins Reviewed-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com> (cherry picked from commit dfd75ec6d4dcfec57607a8cf7c7a509c33bf2caa) Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86344 Reviewed-by: Andras Timar <andras.timar@collabora.com> (cherry picked from commit 62eee51aeaee380139126e21ac550e6e367164ab) Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86479 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
2020-01-09DOCX import: fix lost page break when footer ends with a tableMiklos Vajna6-5/+99
Regression from commit 7d3778e0ef9f54f3c8988f1b84d58e7002d6c625 (bnc#816593 DOCX import: ignore page breaks in tables, 2013-09-02), the page break was ignored because the preceding footer ended with a table (no empty paragraph at the end of the footer stream). Fix the problem by saving/loading the table state around header/footers, that way the page break is not ignored. Adjust testTdf102466 to test the page number from Word. (cherry picked from commit a86a2a1c1ceb7203857d4317913c5b1bb9feb4aa) Conflicts: writerfilter/source/dmapper/DomainMapper_Impl.hxx Change-Id: Ia4c22452ee2c37f7f941dfd922db04c851644d0c Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86451 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com> Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
2020-01-08tdf#129353, tdf#129402: fix node creation on index importMike Kaganski8-78/+189
ToC, bibliography, and index sections import code changed to closely follow what Word does, make sure that pre-rendered entries don't get imported as standalone paragraphs outside of the index sections, and paragraph count is accurate (no missing or added paragraphs as much as possible). In Word, an index may start and end in the middle of a paragraph: <w:p> <w:r> <w:t>Some text before index</w:t> </w:r> <w:r> <w:fldChar w:fldCharType="begin"/> </w:r> <w:r> <w:instrText> TOC ...</w:instrText> </w:r> <w:r> <w:fldChar w:fldCharType="separate"/> </w:r> <w:r> <w:t>First pre-rendered index entry</w:t> </w:r> </w:p> ... <w:p> <w:r> <w:t>Last pre-rendered index entry</w:t> </w:r> <w:r> <w:fldChar w:fldCharType="end"/> </w:r> <w:r> <w:t>Some text after index</w:t> </w:r> </w:p> However, normally it looks like either no runs precedig index, or no runs of pre-rendered contents will be present. When no Std elements are used, the typical situation is that there's a normal paragraph (possibly with some user text), which ends with index start marker, without any pre-rendered contents in the same paragraph; and all pre- rendered contents goes in following paragraphs. Such index normally ends with index end marker in the *first* run of a paragraph, which then might have normal text runs. When Stds are used, then no leading/trailing out-of-index runs in paragraphs with marks are usually present; and in this case, when paragraphs with index marks don't contain pre-rendered entries, they still are treated as part of the index. In Writer, indexes are node sections (and so cannot be inline with other paragraph contents). When there was some paragraph content already before the start-of-index mark, the paragraph is assumed to end before the index; in this case, when current <w:p> element ends, importer decides if a separate starting paragraph is needed or not, depending on if there was some runs after the mark. When there was no text runs before the starting mark, then the paragraph is treated as leading paragraph of the index. This allows to not miss empty paragraphs before index; and not have two paragraphs where there was one in Word. Only in cases when user had manually typed text both in and outside of the index in the same paragraph in Word, we would have the paragraph split into two in Writer. For end marks, the behaviour depends on whether it's inside Std. When inside, the ending paragraph starting with index end mark is considered part of the index. For out-of-Std case, it's considered normal paragraph (and measures are taken to make sure it's not dropped even if empty, because sometimes such paragraphs don't have other content, and have section settings, which is usually treated by Writer as "drop this paragraph" sign). A special problem is multi-column index. It's wrapped into a continuous section by Word; and in Writer, we also wrap it into a section. It would be possibly useful to detect somehow if this section is part of index definition, and in this case, drop the section and put its properties into the Writer's index section. That would avoid an explicit section in the imported document. This is TODO, for someone who figures how to detect reliably if the section belongs to index definition. See comment in DomainMapper_Impl::appendTextSectionAfter. By the way, current export code is wrong, producing an index that is single-column in Word; this change doesn't touch that. Several existing tests needed to be fixed, which used to test wrong results. Change-Id: I9597c8ab13f31ded9abcc24054d3478d3e3a3b40 Reviewed-on: https://gerrit.libreoffice.org/85089 Tested-by: Jenkins Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com> Reviewed-on: https://gerrit.libreoffice.org/c/core/+/85278 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
2020-01-08tdf#129373: Make table column width and row height controls look better on iOSTor Lillqvist1-2/+3
For some reason the space allocated for the measurement in a svxlo-SvxRelativeField is quite large, and it looks weird when the plus button of the row height control bumps into the icon of the column width control. (Changing the hexpand element to False does not fully help.) It looks much better to have these controls on separate rows. Hopefully the change in how it looks in web-based Online or in desktop LibreOffice will not cause endless bikeshedding. Change-Id: Ia8fbbbfe035a92fc5c89b27dc8c8985707a5cb0b Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86358 Reviewed-by: Tor Lillqvist <tml@collabora.com> Tested-by: Tor Lillqvist <tml@collabora.com>
2020-01-08sw: don't send LOK notifications about redlines during saveMiklos Vajna6-1/+31
SwXMLWriter::Write_() sets redline flags to show insertion and hide deletion, but it resets those flags before the function returns. So LOK notifications for redline changes during save is not useful. Conflicts: sw/qa/extras/tiledrendering/tiledrendering.cxx Change-Id: I4bf963bbe9c7003cbe85ea6c5538be733a3e3cdf Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86369 Reviewed-by: Miklos Vajna <vmiklos@collabora.com> Tested-by: Miklos Vajna <vmiklos@collabora.com>
2020-01-07android: avoid expensive load of un-used sidebar icons on mobile.Michael Meeks1-0/+6
Change-Id: I34c24c40009eeb4e8edbac785f9367f1dc3d56c1 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86367 Reviewed-by: Michael Meeks <michael.meeks@collabora.com> Tested-by: Michael Meeks <michael.meeks@collabora.com>
2020-01-07ofz#19803 check for negative valuesCaolán McNamara1-1/+1
Change-Id: I30036a16cf1651a7e27c7aefa086b3725d58933a Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86250 Tested-by: Jenkins Reviewed-by: Caolán McNamara <caolanm@redhat.com> Tested-by: Caolán McNamara <caolanm@redhat.com> (cherry picked from commit 0ab8ae4d98122f7de50365a062675d001ef0ca6a) Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86349 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com> Reviewed-by: Andras Timar <andras.timar@collabora.com>
2020-01-07tdf#129845 vcl: avoid expensive system caching of trivial lines & polygonsMichael Meeks1-33/+64
Interestingly the cache map lookup is rather expensive; why: polyline output for some trivial impress edits: count # of polylines 2 2134 3 141 4 41 9 4 polypolygon output for some trivial impress edits: count # of polypolygons geometry. 3 54 all single polygon 4 583 ~all single 9 52 ~ all paired with a 4 node polygon 13 2 both single 32 22 all single Change-Id: I15c0053a84399eaf153b2119b2c28d1f168f16b1 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86313 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com> Reviewed-by: Michael Meeks <michael.meeks@collabora.com>
2020-01-06tdf#125444 Add pptx import test for percentageTamas Bunth2-0/+23
Add test for imporing custom label field of type "PERCENTAGE" from a pptx document. Change-Id: Ide6baa95d07836cafb97efc9c00ec2ab224a1d1b Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86109 Tested-by: Jenkins Reviewed-by: Tamás Bunth <btomi96@gmail.com> Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86127 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com> Reviewed-by: Andras Timar <andras.timar@collabora.com>
2020-01-06tdf#125444 Percentage as custom chart labelTamas Bunth3-2/+10
Support importing ooxml charts with a field of type "PERCENTAGE" as custom label. Change-Id: Ie8931f77e3b6199d98635422d11d776e675f6e5c Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86067 Tested-by: Jenkins Reviewed-by: Tamás Bunth <btomi96@gmail.com> Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86118 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com> Reviewed-by: Andras Timar <andras.timar@collabora.com>
2020-01-06tdf#123206 Add test: import custom-label-fieldTamas Bunth3-2/+22
Add test for deprecated custom-label-field attribute of the data-point element. Change-Id: I9a7619898fa72900323e5b41728d2146c14203e3 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86103 Tested-by: Jenkins Reviewed-by: Tamás Bunth <btomi96@gmail.com> Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86117 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com> Reviewed-by: Andras Timar <andras.timar@collabora.com>
2020-01-06tdf#123206 Store custom label as chart:data-labelTamas Bunth10-46/+186
Use the chart:data-label element instead of using the loext:custom-label-field attribute. chart:data-label stores can be a child of chart:data-point and it may contain a text:o element for holding one or more paragraphs of custom label text. This commit aims to export and import chart:data-label with paragraphs put into different text:span elements. These span elements may hold a text:style-name attribute in order to achieve formatted text. This structure is already in the ODF format. Change-Id: I0bea7ce1a16af9c47b33555e18545bdaae7e95ca Reviewed-on: https://gerrit.libreoffice.org/c/core/+/85659 Tested-by: Jenkins Reviewed-by: Tamás Bunth <btomi96@gmail.com> Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86116 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com> Reviewed-by: Andras Timar <andras.timar@collabora.com>
2020-01-06sidebar: improve invalidation tracking for Panels.Michael Meeks2-4/+32
Only emit an invalidation on panels that change position. Ensure we invalidate the ScrollContainerWindow for updated separators. Emit a single vcl::Region for an invalidation rather than per panel. Change-Id: I5452791ac9a7d1a9b8604c7704d24641160c275b Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86234 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com> Reviewed-by: Michael Meeks <michael.meeks@collabora.com>
2020-01-06sidebar: allow panels to lurk around instead of being disposed.Michael Meeks5-41/+72
Creating and destroying sidebar panels is done remarkably often - on changes of context eg. The process is remarkably expensive - loading UI XML files, processing them etc. and is un-necessary. Instead let panels lurk around for future use - particularly in the Properties deck which gets the most thrash. This gives a big speedup particularly noticable on mobile where it could take several seconds to load switch between panels for eg. shape vs. slide properties when tapping to edit text. Change-Id: I497e77432c11bbd1e35a8a8716519cabc3730e61 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86233 Tested-by: Michael Meeks <michael.meeks@collabora.com> Reviewed-by: Michael Meeks <michael.meeks@collabora.com>
2020-01-03Use only the 'MacOSXSpell' component on iOS, not the 'spell' oneTor Lillqvist6-11/+16
Using only the system spell checker (through MacOSXSpell) is what we have been doing anyway. Do not build the hunspell or mythes externals for iOS. Do not build the lnth or spell components for iOS. Change-Id: I2e2abc268d7719e540072e5daff3f7960e04ed27 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86172 Reviewed-by: Tor Lillqvist <tml@collabora.com> Tested-by: Tor Lillqvist <tml@collabora.com>
2020-01-03android: fixup - the menu file was meant to go to the non-extracted assetsChristian Lohmaier1-7/+7
fixup from 93ee955c5359e283e2fe88e45b4c02a60525ed11 Change-Id: I94fa357e982bdf6a74bc182a51fb562ac39c0090 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86167 Reviewed-by: Michael Meeks <michael.meeks@collabora.com> Tested-by: Michael Meeks <michael.meeks@collabora.com>
2020-01-03tdf#128101 android: fix documents with manual page breakChristian Lohmaier2-7/+24
code reads a .ui file to show a menu to edit/delete that pagebreak. That file was not packaged in the Android viewer and causes an exception that is not handled and ultimately results in a crash. Change-Id: Ie73d886daf9202ba12e1b5a241bc7b6d184ae770 Reviewed-on: https://gerrit.libreoffice.org/83104 Tested-by: Jenkins Reviewed-by: Christian Lohmaier <lohmaier+LibreOffice@googlemail.com> Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86165 Reviewed-by: Michael Meeks <michael.meeks@collabora.com> Tested-by: Michael Meeks <michael.meeks@collabora.com>
2020-01-03android: file-cache to improve performance.Michael Meeks1-65/+98
The transition to java - interestingly to free the passed buffer was showing on profiles. Change-Id: Id1f4f6e60896c3f42fcbf761e535b68318e0a0a2 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86164 Reviewed-by: Michael Meeks <michael.meeks@collabora.com> Tested-by: Michael Meeks <michael.meeks@collabora.com>
2020-01-03android: don't try to copy the apk if it does not existMiklos Vajna1-1/+1
Linking obj/local/armeabi-v7a/liblo-native-code.so cp: cannot stat 'android/source/build/outputs/apk/strippedUI/release/*-release-unsigned.apk': No such file or directory make[1]: *** [android/CustomTarget_lo_android.mk:19: workdir/CustomTarget/android/source/done] Error 1 Change-Id: I90818d432cab0ef8ef7049b6be175db9e9028261 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86157 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com> Reviewed-by: Jan Holesovsky <kendy@collabora.com>
2020-01-02android: Avoid building the apk when configured with --enable-android-lok.Jan Holesovsky1-6/+4
Change-Id: I6e82d1c1c00d67f5ff370a7625e33efbb172ca38 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86110 Reviewed-by: Jan Holesovsky <kendy@collabora.com> Tested-by: Jan Holesovsky <kendy@collabora.com>
2020-01-01tools: define DbgTestSolarMutex() unconditionallyMiklos Vajna2-7/+3
See the discussion at <https://gerrit.libreoffice.org/#/c/58170/>, and this came up on IRC today again. The above change broke the invariant that you can mix product and debug (but not dbgutil) builds. Restore this, without mandating dbgutil for the solar mutex assert code, which is useful for plain debug builds as well. Change-Id: I1f8bdb114b129fc4f39f186ba917e35e346a16b5 Reviewed-on: https://gerrit.libreoffice.org/85369 Tested-by: Jenkins Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
2019-12-31tdf#129375: Avoid crash when inserting bibliography entry on iOSTor Lillqvist6-15/+27
I don't really know how the vivliography functionality works and how it is connected to database stuff. Until now the Library_bib for instance was excluded for iOS because it was seen to be part of the "DBCONNECTIVITY" feature. Change that now. Also, build the dba and dbahsql libraries also in the non-DBCONNECTIVITY case. This at least avoids the crash and avoids new warnings about missing constructors or factories. Change-Id: I8a8c62a895fcd43e7fa725a4707ac5ad428a64b8 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86043 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com> Reviewed-by: Tor Lillqvist <tml@collabora.com>
2019-12-27Remove broken MergeDataHashMap optimizationStephan Bergmann2-156/+12
...which apparently didn't take into account that inserting into a std::unordered_map may invalidate iterators, which now started to cause build failures like > [PRP] dictionaries/hu_HU/dialog/hu_HU_de > warn:sal.osl:25175:25175:sal/osl/unx/thread.cxx:1026: RTL_TEXTENCODING_DONTKNOW -> _ASCII_US > .../gcc/trunk/inst/include/c++/9.0.1/debug/safe_iterator.h:307: > In function: > __gnu_debug::_Safe_iterator<_Iterator, _Sequence, _Category>::pointer > __gnu_debug::_Safe_iterator<_Iterator, _Sequence, > _Category>::operator->() const [with _Iterator = > std::__detail::_Node_iterator<std::pair<const rtl::OString, > std::unique_ptr<MergeData> >, false, true>; _Sequence = > std::__debug::unordered_map<rtl::OString, std::unique_ptr<MergeData> >; > _Category = std::forward_iterator_tag; > __gnu_debug::_Safe_iterator<_Iterator, _Sequence, _Category>::pointer = > std::pair<const rtl::OString, std::unique_ptr<MergeData> >*] > > Error: attempt to dereference a singular iterator. > > Objects involved in the operation: > iterator "this" @ 0x0x5a8228 { > type = std::__detail::_Node_iterator<std::pair<rtl::OString const, std::unique_ptr<MergeData, std::default_delete<MergeData> > >, false, true> (mutable iterator); > state = singular; > references sequence with type 'std::__debug::unordered_map<rtl::OString, std::unique_ptr<MergeData, std::default_delete<MergeData> >, std::hash<rtl::OString>, std::equal_to<rtl::OString>, std::allocator<std::pair<rtl::OString const, std::unique_ptr<MergeData, std::default_delete<MergeData> > > > >' @ 0x0x5a81c8 > } > /bin/sh: line 1: 25175 Aborted (core dumped) LD_LIBRARY_PATH=${LD_LIBRARY_PATH:+$LD_LIBRARY_PATH:}"$I/program:$I/program" $W/LinkTarget/Executable/propex -i $S/dictionaries/hu_HU/dialog/hu_HU_en_US.properties -l de -m ${MERGEINPUT} -o $W/PropertiesTranslateTarget/dictionaries/hu_HU/dialog/hu_HU_de.properties > make[1]: *** [.../solenv/gbuild/Dictionary.mk:88: .../workdir/PropertiesTranslateTarget/dictionaries/hu_HU/dialog/hu_HU_de.properties] Error 134 with trunk libstdc++ in debug mode. Both classes MergeData and MergeDataHashMap became redundant. Reviewed-on: https://gerrit.libreoffice.org/69395 Tested-by: Jenkins Reviewed-by: Stephan Bergmann <sbergman@redhat.com> (cherry picked from commit dc2329fb66dc38f62a21b1afaca38529d7e40fa9) Change-Id: I771548a6155e14037d84acfd74cd13566a26a8d7 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/85866 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com> Reviewed-by: Aron Budea <aron.budea@collabora.com> Tested-by: Aron Budea <aron.budea@collabora.com>
2019-12-26sw: paragraph-sign: process signatures in one pass on loadingAshod Nakashian4-12/+17
This combines the metadata field restoration logic with validation/update of said field in one pass upon loading docs. This cuts down overheads and makes the code more readable. In addition, now that paragraph signature validation is quite cheap (separate commits), esp. when no signatures exist, we validate and update signatures (where applicable) when fields are updated. Change-Id: I4adcea579c591f9be457ed742d2cf54fa308163d Reviewed-on: https://gerrit.libreoffice.org/63008 Tested-by: Jenkins Reviewed-by: Ashod Nakashian <ashnakash@gmail.com> (cherry picked from commit ad9902f34dd0b5837ab518445ea9f3a24352738b) Reviewed-on: https://gerrit.libreoffice.org/c/core/+/85809 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
2019-12-26sw: paragraph-sign: avoid re-validating paragraphs when classifyingAshod Nakashian1-0/+6
When setting the classification watermark, every edit to the document is triggering a re-validation of signatures, which is excessive and excessively expensive. Change-Id: I7808877263f553a871f2c31856da1a2f9e50c4bb Reviewed-on: https://gerrit.libreoffice.org/63007 Tested-by: Jenkins Reviewed-by: Ashod Nakashian <ashnakash@gmail.com> (cherry picked from commit 10a0693a60d11cfd2401f9f96041bc3d5d7c4c5b) Reviewed-on: https://gerrit.libreoffice.org/c/core/+/85808 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
2019-12-26sw: paragraph-sign: skip invalid graph namesAshod Nakashian1-0/+3
Change-Id: Iaff61723497466521eddb2333656981f976d3fa6 Reviewed-on: https://gerrit.libreoffice.org/63006 Tested-by: Jenkins Reviewed-by: Ashod Nakashian <ashnakash@gmail.com> (cherry picked from commit ffc9efb5c770a674c781bfedb3b91e912b6d0136) Reviewed-on: https://gerrit.libreoffice.org/c/core/+/85807 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
2019-12-26sw: paragraph-sign: formattingAshod Nakashian1-18/+23
Change-Id: I9bf1b58aa6e18202a5f7eba010cc5b914e1d33d4 Reviewed-on: https://gerrit.libreoffice.org/63005 Tested-by: Jenkins Reviewed-by: Ashod Nakashian <ashnakash@gmail.com> (cherry picked from commit 89c351a2610cc24720f391cf933f8af170de03ed) Reviewed-on: https://gerrit.libreoffice.org/c/core/+/85806 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
2019-12-26sw: paragraph-sign: validate current SwTextNode directlyAshod Nakashian3-13/+21
When invoking undo, it turns out that the cursor position is updated after the text modification, which triggers the paragraph signature validation. Relying on the cursor position, then, results in the wrong (previous) paragraph to be validated (if the undo is in a different paragraph). Since we have the correct SwTextNode when it's modified (due to undo or otherwise), there is no reason why we shouldn't use it and try to deduce it from the cursor. Change-Id: I4c3283d59738988dcc1c592a9f3ef2c818ce675d Reviewed-on: https://gerrit.libreoffice.org/63004 Tested-by: Jenkins Reviewed-by: Ashod Nakashian <ashnakash@gmail.com> (cherry picked from commit 25446d1c7d021ba99201f36b04d59c29c85b7733) Reviewed-on: https://gerrit.libreoffice.org/c/core/+/85805 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
2019-12-26sw: paragraph-sign: get graph-names only onceAshod Nakashian1-27/+31
And reduce other overheads to make the paragraph signature validation significantly faster. Change-Id: I631bd0f15200bb8f0b85b40c0d4819a7bc4d562e Reviewed-on: https://gerrit.libreoffice.org/63003 Tested-by: Jenkins Reviewed-by: Ashod Nakashian <ashnakash@gmail.com> (cherry picked from commit 9ae364330ddbc7cec836d42221ebc9f7b4c40240) Reviewed-on: https://gerrit.libreoffice.org/c/core/+/85804 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
generated by cgit v1.2.3 (git 2.39.1) at 2024-06-06 19:46:39 +0000