http://cia.vc/stats/author/mkruisselbrink1293NoneNoneNoneMarijn KruisselbrinkNoneNoneSubversion CIA Bot client shell script1.0from svn.kde.org (svn.kde.org [195.135.221.69])
by cia.vc (Postfix) with ESMTP id 78A421D82B1
for <cia@cia.navi.cx>; Tue, 7 Dec 2010 05:25:17 -0800 (PST)by svn.kde.org (Postfix, from userid 50782)
id C6D0FAC8A6; Tue, 7 Dec 2010 14:25:06 +0100 (CET)sysadmin@kde.org<20101207132506.C6D0FAC8A6@svn.kde.org>Tue, 7 Dec 2010 14:25:06 +0100 (CET)KDEinteroperability1204473mkruisselbrinktrunk/tests/kofficetests/interoperability/kspread/MSExcel2003/me03_webpage_copied.xlsAdd one more file10http://websvn.kde.org/?view=revision&revision=12044731291728317Subversion CIA Bot client shell script1.0from svn.kde.org (svn.kde.org [195.135.221.69])
by cia.vc (Postfix) with ESMTP id 702981D83C0
for <cia@cia.navi.cx>; Thu, 2 Dec 2010 01:27:00 -0800 (PST)by svn.kde.org (Postfix, from userid 50782)
id 637E7AC8B2; Thu, 2 Dec 2010 10:26:49 +0100 (CET)sysadmin@kde.org<20101202092649.637E7AC8B2@svn.kde.org>Thu, 2 Dec 2010 10:26:49 +0100 (CET)KDE1202778mkruisselbrinkbranches/work/koffice-essen/kspread/CMakeLists.txtlink to threadweaver, and optionally define KSPREAD_MT to enable all the multithreaded stuff22http://websvn.kde.org/?view=revision&revision=12027781291282031Subversion CIA Bot client shell script1.0from svn.kde.org (svn.kde.org [195.135.221.69])
by cia.vc (Postfix) with ESMTP id 0E81D1D83C7
for <cia@cia.navi.cx>; Thu, 2 Dec 2010 01:26:58 -0800 (PST)by svn.kde.org (Postfix, from userid 50782)
id 23BE1AC8B0; Thu, 2 Dec 2010 10:26:45 +0100 (CET)sysadmin@kde.org<20101202092645.23BE1AC8B0@svn.kde.org>Thu, 2 Dec 2010 10:26:45 +0100 (CET)KDE1202776mkruisselbrinkbranches/work/koffice-essen/kspread/ui/SheetView.cppbranches/work/koffice-essen/kspread/ui/SheetView.hprotect cache with a mutex, and don't return cellviews by reference, as the references can cease to be valid at any point due to the nature of QCache181http://websvn.kde.org/?view=revision&revision=12027761291282027Subversion CIA Bot client shell script1.0from svn.kde.org (svn.kde.org [195.135.221.69])
by cia.vc (Postfix) with ESMTP id 948CA1D83C0
for <cia@cia.navi.cx>; Thu, 2 Dec 2010 01:26:58 -0800 (PST)by svn.kde.org (Postfix, from userid 50782)
id 5F74DAC8A7; Thu, 2 Dec 2010 10:26:47 +0100 (CET)sysadmin@kde.org<20101202092647.5F74DAC8A7@svn.kde.org>Thu, 2 Dec 2010 10:26:47 +0100 (CET)KDE1202777mkruisselbrinkbranches/work/koffice-essen/kspread/ui/PixmapCachingSheetView.cppbranches/work/koffice-essen/kspread/ui/PixmapCachingSheetView.hand add the real background painting logic to PixmapCachingSheetView; no proper cleaning up yet, and I do probably want to rewrite it to use QFuture/QtConcurrent instead of threadweaver, but at least it seems to work quite well already204http://websvn.kde.org/?view=revision&revision=12027771291282026Subversion CIA Bot client shell script1.0from svn.kde.org (svn.kde.org [195.135.221.69])
by cia.vc (Postfix) with ESMTP id 3C6681D83C7
for <cia@cia.navi.cx>; Thu, 2 Dec 2010 01:26:53 -0800 (PST)by svn.kde.org (Postfix, from userid 50782)
id C2BF6AC8AD; Thu, 2 Dec 2010 10:26:42 +0100 (CET)sysadmin@kde.org<20101202092642.C2BF6AC8AD@svn.kde.org>Thu, 2 Dec 2010 10:26:42 +0100 (CET)KDE1202775mkruisselbrinkbranches/work/koffice-essen/kspread/ui/SheetView.cppbranches/work/koffice-essen/kspread/ui/SheetView.hmake visibleRect a parameter of the paint methods, to support multiple parallel paints with different visible rects206http://websvn.kde.org/?view=revision&revision=12027751291282018Subversion CIA Bot client shell script1.0from svn.kde.org (svn.kde.org [195.135.221.69])
by cia.vc (Postfix) with ESMTP id 279631D83C0
for <cia@cia.navi.cx>; Thu, 2 Dec 2010 01:26:51 -0800 (PST)by svn.kde.org (Postfix, from userid 50782)
id 678D3AC8AB; Thu, 2 Dec 2010 10:26:40 +0100 (CET)sysadmin@kde.org<20101202092640.678D3AC8AB@svn.kde.org>Thu, 2 Dec 2010 10:26:40 +0100 (CET)KDE1202774mkruisselbrinkbranches/work/koffice-essen/kspread/ui/CellView.cppsort of try to handle QTextDocument in a thread-safe manner; while it works correctly, it is impossible to use QTextDocument in any non-trivial multi-threaded manner without qWarnings getting printed all the time, because you can't do anything to a QTextDocument in a thread that is different from the one it is created in/associated to. And this also includes cloning it (cloning a document creates some internal QObjects with the original document as parent, which gives warnings because threads associated to the objects and their parents don't match).
The only way to use QTextDocuments in multiple threads is by cloning them in the GUI thread, and then moveToThread them to the thread where they will be used, but that can never work here because we have know way of knowing which ThreadWeaver (or QtConcurrent) worker thread will need access to which QTextDocument, so conclusion: QTextDocument is broken by design.88http://websvn.kde.org/?view=revision&revision=12027741291282014Subversion CIA Bot client shell script1.0from svn.kde.org (svn.kde.org [195.135.221.69])
by cia.vc (Postfix) with ESMTP id 791551D83C0
for <cia@cia.navi.cx>; Thu, 2 Dec 2010 01:26:48 -0800 (PST)by svn.kde.org (Postfix, from userid 50782)
id CBD73AC8AA; Thu, 2 Dec 2010 10:26:37 +0100 (CET)sysadmin@kde.org<20101202092637.CBD73AC8AA@svn.kde.org>Thu, 2 Dec 2010 10:26:37 +0100 (CET)KDE1202773mkruisselbrinkbranches/work/koffice-essen/kspread/StyleStorage.cppalso protect style-storage cache with a mutex114http://websvn.kde.org/?view=revision&revision=12027731291282011Subversion CIA Bot client shell script1.0from svn.kde.org (svn.kde.org [195.135.221.69])
by cia.vc (Postfix) with ESMTP id E44551D83C0
for <cia@cia.navi.cx>; Thu, 2 Dec 2010 01:26:46 -0800 (PST)by svn.kde.org (Postfix, from userid 50782)
id 7679EAC8A7; Thu, 2 Dec 2010 10:26:35 +0100 (CET)sysadmin@kde.org<20101202092635.7679EAC8A7@svn.kde.org>Thu, 2 Dec 2010 10:26:35 +0100 (CET)KDE1202772mkruisselbrinkbranches/work/koffice-essen/kspread/RectStorage.hprotect rect-storage internal cache with a mutex; for the other thread safety the protection inside CellStorage should be enough46http://websvn.kde.org/?view=revision&revision=12027721291282009Subversion CIA Bot client shell script1.0from svn.kde.org (svn.kde.org [195.135.221.69])
by cia.vc (Postfix) with ESMTP id C9EE81D83C0
for <cia@cia.navi.cx>; Thu, 2 Dec 2010 01:26:38 -0800 (PST)by svn.kde.org (Postfix, from userid 50782)
id 272EAAC8A5; Thu, 2 Dec 2010 10:26:28 +0100 (CET)sysadmin@kde.org<20101202092628.272EAAC8A5@svn.kde.org>Thu, 2 Dec 2010 10:26:28 +0100 (CET)KDE1202771mkruisselbrinkbranches/work/koffice-essen/kspread/CellStorage.cppfirst step towards a more multi-threaded kspread: protect access through the CellStorage (and thus Cell) interfaces with one big read/write lock per sheet733http://websvn.kde.org/?view=revision&revision=12027711291281999Subversion CIA Bot client shell script1.0from svn.kde.org (svn.kde.org [195.135.221.69])
by cia.vc (Postfix) with ESMTP id 6EE4F1D83C3
for <cia@cia.navi.cx>; Fri, 26 Nov 2010 05:05:19 -0800 (PST)by svn.kde.org (Postfix, from userid 50782)
id 827DFAC8A3; Fri, 26 Nov 2010 14:05:08 +0100 (CET)sysadmin@kde.org<20101126130508.827DFAC8A3@svn.kde.org>Fri, 26 Nov 2010 14:05:08 +0100 (CET)KDE1200992mkruisselbrinkbranches/work/koffice-essen/kspread/ui/SheetView.cppnow CellView has an explicitly shared data pointer, there is no reason anymore to keep references here (and in fact references can cause crashes because the CellView might get deleted at any point)23http://websvn.kde.org/?view=revision&revision=12009921290776739Subversion CIA Bot client shell script1.0from svn.kde.org (svn.kde.org [195.135.221.69])
by cia.vc (Postfix) with ESMTP id 7CD041D83D3
for <cia@cia.navi.cx>; Fri, 26 Nov 2010 02:10:41 -0800 (PST)by svn.kde.org (Postfix, from userid 50782)
id 1B33CAC8AE; Fri, 26 Nov 2010 11:10:28 +0100 (CET)sysadmin@kde.org<20101126101028.1B33CAC8AE@svn.kde.org>Fri, 26 Nov 2010 11:10:28 +0100 (CET)KDE1200934mkruisselbrinkbranches/work/koffice-essen/kspread/ui/CellView.cppanother nice catch by thorsten: this check never actually causes a return, as SheetView already takes care to never call this for cells that are not visible23http://websvn.kde.org/?view=revision&revision=12009341290766641Subversion CIA Bot client shell script1.0from svn.kde.org (svn.kde.org [195.135.221.69])
by cia.vc (Postfix) with ESMTP id A17A61D83CC
for <cia@cia.navi.cx>; Fri, 26 Nov 2010 02:10:37 -0800 (PST)by svn.kde.org (Postfix, from userid 50782)
id 484AAAC8AB; Fri, 26 Nov 2010 11:10:26 +0100 (CET)sysadmin@kde.org<20101126101026.484AAAC8AB@svn.kde.org>Fri, 26 Nov 2010 11:10:26 +0100 (CET)KDE1200932mkruisselbrinkbranches/work/koffice-essen/kspread/ui/CellView.cppbranches/work/koffice-essen/kspread/ui/CellView.hseems I also need assignment operator...41http://websvn.kde.org/?view=revision&revision=12009321290766627Subversion CIA Bot client shell script1.0from svn.kde.org (svn.kde.org [195.135.221.69])
by cia.vc (Postfix) with ESMTP id A3A9D1D83D5
for <cia@cia.navi.cx>; Fri, 26 Nov 2010 02:10:36 -0800 (PST)by svn.kde.org (Postfix, from userid 50782)
id 3D5E0AC8A8; Fri, 26 Nov 2010 11:10:24 +0100 (CET)sysadmin@kde.org<20101126101024.3D5E0AC8A8@svn.kde.org>Fri, 26 Nov 2010 11:10:24 +0100 (CET)KDE1200931mkruisselbrinkbranches/work/koffice-essen/kspread/CMakeLists.txtbranches/work/koffice-essen/kspread/ui/PixmapCachingCellView.cppbranches/work/koffice-essen/kspread/ui/PixmapCachingCellView.hbranches/work/koffice-essen/kspread/ui/PixmapCachingSheetView.cppbranches/work/koffice-essen/kspread/ui/PixmapCachingSheetView.hdon't do double pixmap caching; with the caching in the sheetview we can get rid of the caching in the cellview201http://websvn.kde.org/?view=revision&revision=12009311290766621Subversion CIA Bot client shell script1.0from svn.kde.org (svn.kde.org [195.135.221.69])
by cia.vc (Postfix) with ESMTP id 7F33A1D83D3
for <cia@cia.navi.cx>; Fri, 26 Nov 2010 02:10:32 -0800 (PST)by svn.kde.org (Postfix, from userid 50782)
id A5574AC8A6; Fri, 26 Nov 2010 11:10:21 +0100 (CET)sysadmin@kde.org<20101126101021.A5574AC8A6@svn.kde.org>Fri, 26 Nov 2010 11:10:21 +0100 (CET)KDE1200930mkruisselbrinkbranches/work/koffice-essen/kspread/ui/CellView.cppbranches/work/koffice-essen/kspread/ui/CellView.hbranches/work/koffice-essen/kspread/ui/SheetView.cppcache some of the calculations in SheetView::paintCells
patch by Thorsten Zachmann129http://websvn.kde.org/?view=revision&revision=12009301290766616Subversion CIA Bot client shell script1.0from svn.kde.org (svn.kde.org [195.135.221.69])
by cia.vc (Postfix) with ESMTP id 5A57A1D83C3
for <cia@cia.navi.cx>; Fri, 26 Nov 2010 02:10:18 -0800 (PST)by svn.kde.org (Postfix, from userid 50782)
id 56BF8AC8A5; Fri, 26 Nov 2010 11:10:07 +0100 (CET)sysadmin@kde.org<20101126101007.56BF8AC8A5@svn.kde.org>Fri, 26 Nov 2010 11:10:07 +0100 (CET)KDE1200929mkruisselbrinkbranches/work/koffice-essen/kspread/ui/CellView.cppbranches/work/koffice-essen/kspread/ui/CellView.hmake CellView explicitly shared instead of implicitly, so it is possible to pass it around by value, yet still have the caching done inside it work correctly57http://websvn.kde.org/?view=revision&revision=12009291290766607Subversion CIA Bot client shell script1.0from svn.kde.org (svn.kde.org [195.135.221.69])
by cia.vc (Postfix) with ESMTP id E07971D83C3
for <cia@cia.navi.cx>; Fri, 26 Nov 2010 01:45:44 -0800 (PST)by svn.kde.org (Postfix, from userid 50782)
id 720D6AC8A4; Fri, 26 Nov 2010 10:45:33 +0100 (CET)sysadmin@kde.org<20101126094533.720D6AC8A4@svn.kde.org>Fri, 26 Nov 2010 10:45:33 +0100 (CET)KDEfilters1200921mkruisselbrinktrunk/koffice/filters/kspread/excel/sidewinder/excel.cppdon't crash when unexpected data is found in the CompObj stream (and whatever would be read there was ignored anyway).
BUG: 25793824http://websvn.kde.org/?view=revision&revision=12009211290764963Subversion CIA Bot client shell script1.0from svn.kde.org (svn.kde.org [195.135.221.69])
by cia.vc (Postfix) with ESMTP id 70B3D1D83C3
for <cia@cia.navi.cx>; Wed, 24 Nov 2010 01:05:44 -0800 (PST)by svn.kde.org (Postfix, from userid 50782)
id B9188AC8A3; Wed, 24 Nov 2010 10:05:33 +0100 (CET)sysadmin@kde.org<20101124090533.B9188AC8A3@svn.kde.org>Wed, 24 Nov 2010 10:05:33 +0100 (CET)KDE1200199mkruisselbrinkbranches/work/koffice-essen/filters/kspread/excel/sidewinder/worksheetsubstreamhandler.cppfix handling of notes/comments. The existing assert was not only incorrect, but also should not have been there as invalid user-input should never trigger assertions. Also actually making use of the NoteObject object makes it work more correct.43http://websvn.kde.org/?view=revision&revision=12001991290589545Subversion CIA Bot client shell script1.0from svn.kde.org (svn.kde.org [195.135.221.69])
by cia.vc (Postfix) with ESMTP id DB2571D83AA
for <cia@cia.navi.cx>; Tue, 23 Nov 2010 05:17:25 -0800 (PST)by svn.kde.org (Postfix, from userid 50782)
id 31845AC8A4; Tue, 23 Nov 2010 14:17:15 +0100 (CET)sysadmin@kde.org<20101123131715.31845AC8A4@svn.kde.org>Tue, 23 Nov 2010 14:17:15 +0100 (CET)KDE1199997mkruisselbrinkbranches/work/koffice-essen/kspread/ui/PixmapCachingSheetView.cppdebug--13http://websvn.kde.org/?view=revision&revision=11999971290518248Subversion CIA Bot client shell script1.0from svn.kde.org (svn.kde.org [195.135.221.69])
by cia.vc (Postfix) with ESMTP id E6FAF1D83AA
for <cia@cia.navi.cx>; Tue, 23 Nov 2010 05:10:55 -0800 (PST)by svn.kde.org (Postfix, from userid 50782)
id 2BE44AC8A3; Tue, 23 Nov 2010 14:10:45 +0100 (CET)sysadmin@kde.org<20101123131045.2BE44AC8A3@svn.kde.org>Tue, 23 Nov 2010 14:10:45 +0100 (CET)KDE1199992mkruisselbrinkbranches/work/koffice-essen/kspread/ui/PixmapCachingSheetView.cppfix the sheetview pixmap caching to also work when rotations exist in the transformation matrix47http://websvn.kde.org/?view=revision&revision=11999921290517858Subversion CIA Bot client shell script1.0from svn.kde.org (svn.kde.org [195.135.221.69])
by cia.vc (Postfix) with ESMTP id C21DE1D83AA
for <cia@cia.navi.cx>; Mon, 22 Nov 2010 07:51:52 -0800 (PST)by svn.kde.org (Postfix, from userid 50782)
id 113DBAC8A2; Mon, 22 Nov 2010 16:51:42 +0100 (CET)sysadmin@kde.org<20101122155142.113DBAC8A2@svn.kde.org>Mon, 22 Nov 2010 16:51:42 +0100 (CET)KDE1199748mkruisselbrinkbranches/work/koffice-essen/kspread/CMakeLists.txtinstall a bunch of kspread headers to make it somewhat possible to write a document viewer based on kspread outside the koffice source tree51http://websvn.kde.org/?view=revision&revision=11997481290441113