| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
| |
Change-Id: I47ac500922d7e834f5915cc8fb1e19d584675620
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Make builds in 2 phases: first it ensures that all targets that are
included by the makefiles (i.e. the .d files) are up to date, then
it restarts and builds everything else.
It would be possible to update the .d files in the same build by simply
having dependencies from the dep-targets to the corresponding real targets,
but that has undesirable effects (8b5a984d45005d3df1c89eae897d6e04612625d8).
For a successful build, the .d files can actually be updated as a side
effect of other rules in such a way that they will be up-to-date on the
next make invocation, thus the restart can be avoided.
This optimization will not always work in case of build failures.
Change-Id: I753039ed33f5a84104f35990769eddc83e1d0d7e
|
|
|
|
|
|
| |
Needed since e352abc2e60abd21e383ebf971e545ca510ac4ed .
Change-Id: I1f46a01f319703a2225386e89fc94ca286b301ca
|
|
|
|
| |
Change-Id: I40c1639356a0c7e60c276bfc7071c976ba04f314
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
...so that regressions like the missing libjava_uno.jnilib in fdo#66703
"REPORTBUILDER - report design wizard does not start on click (missing
libjava_uno.jnilib symlink)" are found earlier in the future. (It is unclear to
me why it had been demoted to a warning right from its inception in
5b6ee999707ef33bf8bbab6848f331df072543ae "INTEGRATION: CWS native99 (1.6.738);
FILE MERGED.")
Change-Id: I415477cff39b4a6b848701374c7f9dbb8ab2f6ba
|
|
|
|
|
|
|
| |
Change-Id: I7d0fbe865cd10d0a3ef690e521d4752d6d7f3939
Reviewed-on: https://gerrit.libreoffice.org/4829
Reviewed-by: Fridrich Strba <fridrich@documentfoundation.org>
Tested-by: Fridrich Strba <fridrich@documentfoundation.org>
|
|
|
|
|
|
|
| |
Change-Id: Ia2c1c6ed43c6a42fa1a65a22f4c8d68de216706c
Reviewed-on: https://gerrit.libreoffice.org/4822
Reviewed-by: Petr Mladek <pmladek@suse.cz>
Tested-by: Petr Mladek <pmladek@suse.cz>
|
|
|
|
| |
Change-Id: I86d01c3ec1267495bd664108fdcf2e5533d97430
|
|
|
|
| |
Change-Id: I341a7c3bdd00de5e4f3e097c32fb05cff62b10b7
|
|
|
|
| |
Change-Id: Iabebc288b68a6729aae05e2875f7bb39d923699a
|
|
|
|
| |
Change-Id: Ib7ef6fe87673c601ad3c24f3ed3f382c039551f4
|
|
|
|
|
|
| |
No more gbuild.XXXXXX files filling /tmp
Change-Id: I2f3cd388ccd9a09bc72e40c0d43d845bc8e38cd7
|
|
|
|
| |
Change-Id: I7cebb54777dc8c2c5331e7b693a1cdc407127acf
|
|
|
|
|
|
|
|
|
| |
Apparently causes problems for some people.
Unfortunately the scp2 entries are "non-standard" so i've had to add
another variable to let the installer find it.
Change-Id: I7a01160369ed8dc401625c06760008341afcee67
|
|
|
|
| |
Change-Id: I127f6756bae786c41122174b615c607b2c3b7ec4
|
|
|
|
|
|
|
| |
Change-Id: I97188235f0b72b5b54423e05deeaf17c5c681048
Reviewed-on: https://gerrit.libreoffice.org/4188
Reviewed-by: Andras Timar <atimar@suse.com>
Tested-by: Andras Timar <atimar@suse.com>
|
|
|
|
| |
Change-Id: Ice814ea226ddac24325770d038e019e19bc635ce
|
|
|
|
| |
Change-Id: I91393aa912bd69f3fbeb67d895ff7d8021effccb
|
|
|
|
|
|
|
|
|
|
| |
And move striplanguagetags.sh to bin/ - possibly it's used.
Change-Id: I68e23ef543f4e26f617d9860681273234ec9174c
Reviewed-on: https://gerrit.libreoffice.org/4239
Reviewed-by: Mathias M <matm@gmx.fr>
Reviewed-by: David Tardon <dtardon@redhat.com>
Tested-by: David Tardon <dtardon@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
| |
Compilers do not necessarily output all includes files when included
using a PCH, so when a PCH is being rebuilt, it might actually get
reused this way, causing its dependencies to be incomplete afterwards.
The PCH actually should not be used when being regenerated, but with MSVC
this apparently happens, even though there's only the option to create
PCH, not to use it.
Change-Id: Iee1340a60085550ab867304c44752daa04cbba31
|
|
|
|
| |
Change-Id: I390df82ebaa0b0e715e0cdfd8d39febe4f217d96
|
|
|
|
|
|
| |
...see new TODO comment in configure.ac
Change-Id: Id3ffd45357a3f539f3d3f068528019334d929654
|
|
|
|
| |
Change-Id: Icfdf8dddc4da20785e0881a95f50ee498e8405cd
|
|
|
|
|
|
|
| |
Change-Id: Ia765b37888b4095a735015e792f06fc89201d1a3
Reviewed-on: https://gerrit.libreoffice.org/4294
Reviewed-by: Noel Power <noel.power@suse.com>
Tested-by: Noel Power <noel.power@suse.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* this and Ib4762f5a260035f00b5e68cf45b687fdf02e9c02 reduces a default
build on my machine from 2min25sec to 2min12sec
* without unitchecks, its down to 1min50sec now
* it reduces the build time on a i7-4770 Windows tinderbox from 99min to
89min
* by now it also takes care of avoiding most string copying, although
this can certainly be considered overeager given that the file
creation on Windows took ~250ms per file before and still will take a
lot longer than any string operation with this change
Change-Id: I515432bdefe2b055c78b6ba97868adbde65d9165
|
|
|
|
| |
Change-Id: Ib4762f5a260035f00b5e68cf45b687fdf02e9c02
|
|
|
|
|
|
|
| |
Change-Id: I5f17a33bcd735b4655d350ce8ec2dd069e215a4b
Reviewed-on: https://gerrit.libreoffice.org/4306
Reviewed-by: Noel Power <noel.power@suse.com>
Tested-by: Noel Power <noel.power@suse.com>
|
|
|
|
|
|
| |
It causes some weird problems with solenv/bin/linkoo.
Change-Id: Ic1613394ec3f0fe508ca813c2600f9bc5439dce7
|
|
|
|
| |
Change-Id: I5e517c93c402dbbdc5dc09c77d537aec819d9a81
|
|
|
|
| |
Change-Id: I66734830ef228b6928b24194f278632fa9f2c70c
|
|
|
|
|
|
|
|
|
|
| |
When 918695a2bf8f95fd437071bfc6ac59e446c256f8 introduced this, we were
having a "foo-bar-baz-"-like hash, and removing the last char made
sense. Since we then switched to 'git log -n 1 --pretty=format:"%H"'
from './g -s log -n 1 --pretty=format:%h-', this is just no longer
necessary.
Change-Id: Icee0d38e5bd917163d678df5cb12c36902460c7b
|
|
|
|
|
|
|
| |
Unfortunately, new gb_Executable_get_target is not
possible to call when parsing platform/ makefiles.
Change-Id: I8528e2fb7f688b148857c24410511f539e1ff8ff
|
|
|
|
|
|
| |
It's simpler than creating a package for every executable.
Change-Id: I5f853ef2d5574b1a1c65232d1d09bc307c46f67e
|
|
|
|
| |
Change-Id: I935fb9fd2620447f9aba9142b7db896e6fe31133
|
|
|
|
|
|
|
| |
Helps to kill package in cpputools.
And autoinstall uno executable.
Change-Id: Ib1d546371ec8d2b06381d88eff98dbc5b70da400
|
|
|
|
| |
Change-Id: Id13e77fe890301a8510952994a91853568a7aea6
|
|
|
|
| |
Change-Id: I3ce56f505e5e011f592d9622665a9b7451dc51ae
|
|
|
|
| |
Change-Id: I60de04bba401632b01f0da7b772a91a98de5ff07
|
|
|
|
| |
Change-Id: Ifbe4a0b6b93c5be6dfd64b6c77ef7172d3f1c425
|
|
|
|
|
|
|
| |
...so that -Wstrict-aliasing, -Wstrict-overflow=1 included in -Wall do not only
produce warnings/errors in non-debug builds.
Change-Id: I4534427738fa7688853564e8c7255a0c3cd199e1
|
|
|
|
|
|
| |
...in com_GCC_defs.mk
Change-Id: I0a3b29e1a91acf3425afa44dea0353bebf88f58f
|
|
|
|
| |
Change-Id: Ia655865b621b1c8e37a52715b89534a00dcda1d7
|
|
|
|
|
|
|
|
|
| |
...but do not pretty-print its dereferenced value. This is in line with the
handling of css::uno::Reference, and avoids gdb Python exceptions when trying to
print uninitialized rtl::Reference instances (which can e.g. happen during a
"backtrace full").
Change-Id: I9a3c0a6441cf23cba748183226832d2ba23bd531
|
|
|
|
| |
Change-Id: I9f2a97715dc0d7fc5905afb1b908de97acdafc57
|
|
|
|
| |
Change-Id: I582a841ea3b2a882cbdc1556058b007a8667024a
|
|
|
|
| |
Change-Id: I7312d94f90758d450a69410819c5e8c142756a47
|
|
|
|
|
|
|
|
| |
gb_LinkTarget__add_internal_headers was used only for sdi headers now
anyway, hopefully for no good reason.
One headers target is enough for all.
Change-Id: Ia01fb7a0aac7f72915448424b3393ecd9ba0f8f5
|
|
|
|
| |
Change-Id: I68410bc1a8b77877b07f60e6ace0f54c255de95f
|
|
|
|
| |
Change-Id: Iaad3f195438369dc5d3d947995d60351bf39538c
|