summaryrefslogtreecommitdiff
path: root/solenv
AgeCommit message (Collapse)AuthorFilesLines
2011-11-04gbuild: document more user defined variablesMichael Stahl2-2/+9
2011-11-03Give Mac master builds a nice background .dmg image, tooThorsten Behrens2-0/+16
Added a new DevVersionCopy action to scp2, that fires if PRODUCTNAME equals "LibO-dev" - in that case, we use the DS_Store_Dev one.
2011-11-03need to quote command hereMarkus Mohrhard1-1/+1
2011-11-03Check $OS, not the build platformTor Lillqvist1-1/+1
2011-11-03Don't attempt any stripping when cross-compilingTor Lillqvist1-1/+2
It uses a hardcoded plain "strip" command anyway, which hardly will work for foreign executable file formats.
2011-11-03Use DYLD_LIBRARY_PATH when OS_FOR_BUILD == MACOSXTor Lillqvist1-0/+4
2011-11-02pushd is a bashism, use #!/bin/bash in solenv/bin/localizeRene Engelhard1-1/+1
2011-11-02Add gbuild support for .components files.Stephan Bergmann3-0/+80
2011-11-02Add gbuild support for .jar in .oxt.Stephan Bergmann1-1/+2
2011-11-02.DELETE_ON_ERROR is a global flag (that should always be set).Stephan Bergmann2-4/+2
2011-11-01Made ooinstall more deterministic, fixed Mac OS X Python module dependencies.Stephan Bergmann1-1/+2
2011-11-01wrong lib name for old-style makefile.mksCaolán McNamara1-1/+1
2011-10-27Avoid Perl warning: Use of uninitialized valueTor Lillqvist1-1/+1
2011-10-27Undo basis/brand split: removed obsoleted BASISROOTNAME.Stephan Bergmann1-6/+0
2011-10-26WaE: possible change in behavior, change in UDT return calling conventionTor Lillqvist2-1/+3
Ignore also MSVC warning C4686. From http://msdn.microsoft.com/en-us/library/ae77984s%28v=VS.90%29.aspx it seems that this is just a warning that the compiler behaviour has changed, but that change was already in MSVC 2003. So it should be irrelevant to us.
2011-10-25WaE: ignore also MSVC warning C4706: assignment within conditional expressionTor Lillqvist2-1/+3
Sad, but it's too tedious to refactor code like: while( (pLayout = GetLayout(pLayout)) ) { ... }
2011-10-25Ignore also MSVC warning C4555, as generated by the <list> header in MSVC2008Tor Lillqvist2-1/+3
2011-10-24Undo basis/brand split: removed RPATH BRAND (same as OOO now).Stephan Bergmann14-19/+3
2011-10-24WinResTarget: remove realpath callsMichael Stahl1-3/+3
2011-10-24JavaClassSet: fix partial rebuildsMichael Stahl1-1/+1
Partial rebuilds can fail because the reubilt classes may import other classes that are not rebuilt; putting the target directory on the classpath seems to fix it (it would also be possible to use -sourcepath but that seems more complicated).
2011-10-24assuming this is a typoCaolán McNamara1-1/+1
2011-10-24gbuild: remove remnants of gb_CHECKOBJECTOWNERMichael Stahl2-10/+0
2011-10-24Undo basis/brand split: Move SDK and ure-link, remove OOO_BASE_DIR and ↵Stephan Bergmann12-25/+24
BaseInstallation.
2011-10-24create links to gdb autoloaders in the right dirDavid Tardon1-2/+25
2011-10-24make --enable-layout have effect under gbuildCaolán McNamara1-2/+21
2011-10-23no need for SRCDIR parameter hereMatúš Kukan5-6/+6
2011-10-23No need to register repositoryMatúš Kukan1-13/+0
2011-10-23do not use gb_Helper_REPOSITORYNAMES anymoreMatúš Kukan4-25/+5
2011-10-22another fix for 68dd19a368fb4afbf3eba4cb722f56b87f8ea7c1 on OSXBjoern Michaelsen1-1/+1
2011-10-21restore accidentally removed lineBjoern Michaelsen1-0/+1
2011-10-21rename YaccObject to YaccTargetBjoern Michaelsen7-28/+28
the object is a GenCxxObject just like those from CustomTargets, we only need to get the generation right
2011-10-21cleanup YaccObjectBjoern Michaelsen2-25/+16
2011-10-21only enable gb_CHECKOBJECTOWNER for explicit check runsBjoern Michaelsen1-0/+4
2011-10-21Fix for YaccObject, also: YaccObject should not generate stuff directly in ↵Bjoern Michaelsen1-2/+2
toplevel workdir
2011-10-21SetupLocal is gone, also fix unbalanced endifBjoern Michaelsen1-6/+0
2011-10-21gbuild should use SRCDIR as per GNU make conventionsBjoern Michaelsen7-23/+23
2011-10-21kill multi repo support: BuildDirs and gbuild.mk, GNUmakefile.mkBjoern Michaelsen2-12/+5
2011-10-21kill multi repo support: ExternalLibBjoern Michaelsen1-4/+2
2011-10-21kill multi repo support: WinResTargetBjoern Michaelsen1-3/+3
2011-10-21kill multi repo support: SetupLocal can only work with multiple repo supportBjoern Michaelsen2-82/+9
2011-10-21kill multi repo support: LinkTargetBjoern Michaelsen1-62/+28
2011-10-21kill multi repo support: UnoApiTargetBjoern Michaelsen1-8/+4
2011-10-21kill multi repo support: AllLangResTargetBjoern Michaelsen1-22/+9
2011-10-21LinkTarget: remove old object owner implementationMichael Stahl2-84/+0
2011-10-21LinkTarget: new check for duplicate objectsMichael Stahl1-0/+21
Always enabled, as this runs for me with an overhead in the range of measurement error (in tail_build): current implementation: time make -nsrj8 17.269s 16.470s 17.537s 16.754s 16.595s disabled: time make -nsrj8 gb_CHECKOBJECTOWNER= 13.212s 13.345s 13.470s 13.852s 13.712s new implementation: time make -nsrj8 gb_CHECKOBJECTOWNER= 13.477s 13.911s 13.624s 13.606s 13.535s
2011-10-21UnoApiTarget: replace realpath callsMichael Stahl1-5/+12
2011-10-21CustomTarget: remove unnecessary dependencyMichael Stahl1-2/+0
There already is a dependency from the library to the package, so this dependency should not be necessary. Furthermore, the command is executed spuriously always, because the .done file is always newer than the generated files.
2011-10-21Revert "pass -s flag to custom target's make process"Michael Stahl1-1/+1
This reverts commit 6c8a7e0e6ad223051442ef6ce1d1103086c3f4a0. Doing this explicitly is unnecessary because make does it automatically. Also, make will pass --jobserver-fds, which happens to match in findstring as well.
2011-10-21Warn if we're missing a RdbTarget from the check_targetsMichael Meeks2-4/+7
2011-10-21more helpful error messages for debugging when a unit test failsMichael Meeks1-1/+3