Age | Commit message (Collapse) | Author |
|
this removes dmake completely out of the build for migrated modules
build.pl now assumes modules to be gbuild, unless there is a
prj/dmake file
Change-Id: I674a036b182ee13c5ec093e83cb3d38133112d3b
|
|
|
|
...by allowing our special @___... tokens anywhere within an install name,
so that external modules can configure --prefix=/@___... etc. This removes
the need for the special extshl and EXTRPATH=LOADER. Also, a new
OUT2BIN_NONE can be used for external modules where the generated libraries
need the default EXTRPATH=OOO, but generated executables are only used
during the build and such need RPATH=NONE.
|
|
|
|
|
|
|
|
|
|
|
|
SAL_UNUSED_PARAMETER (expanding to __attribute__ ((unused)) for GCC)
is used to annotate legitimately unused parameters, so that static
analysis tools can tell legitimately unused parameters from truly
unnecessary ones. To that end, some patches for external modules
are also added, that are only applied when compiling with GCC and
add necessary __attribute__ ((unused)) in headers.
|
|
|
|
|
|
|
|
The silly apkbuilder tool doesn't add extra native libs to an .apk
package unless their name ends with .so. It just silently ignores
them. So, force libtool to avoid versioning suffix for Android.
Yeah, unclear whether apkbuilder is what we will end up using to
construct the APKs of actual LO code using Android apps. But at this
stage when just trying to build a unit testing app, it seems to be the
simplest way to get the .apk properly signed and all to just use a
normal Ant project, and let Ant run apkbuilder.
|
|
|
|
|
|
In the spirit of 12759f67a36d52a345be6002d7017551e8414c59, change
external lib's config.sub to eat the arm-unknown-linux-androideabi
host os string. Also, permit shared libs again - seems Android can
handle those.
Added dictionaries to cross-build-toolset - idxdict is needed.
Should build up to sfx2 - some residual static lib issues there,
and in raptor.
|
|
This reverts commit 2dea0dab4fafda3c10a5bd03ad15ed39a4658b51.
|
|
|
|
|
|
|
|
http://lists.freedesktop.org/archives/libreoffice/2011-July/014970.html
|
|
For iOS, the LibreOffice unit tests that normally are built as
"plugins", i.e. dynamic libraries, are actually built as static
libraries, because we don't do any dynamic linking of own code on iOS.
Enable giving the "plugin" unit test entry point a different name than
a fixed "cppunitTestPlugIn", so that several unit tests can be linked
together into one executable, for simplicity.
Drop the stupid dummy main() function in cppunit plugins. I doubt any
Unix platform we care for needs it, but drop just for iOS and MacOSX
for now.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Makes it easier to build other stuff. Whether we will actually run
normal unit tests at build time, or ever, or not when cross-compiling
to some platform is a different matter...
|
|
|
|
|
|
|
|
|
|
Unfortunately it still untars the tarball. But at least nothing gets
built or delivered when cross-compiling.
|
|
|
|
Conflicts:
cppunit/ldflags.patch
hyphen/hyphen-2.4.patch
libegg/source/eggtrayicon.c
libtextcat/makefile.mk
|
|
Since the version of autotools used for the rolling of the tarball
and the version on a machine running the build might be different,
avoid breaking builds.
Notes:
split repo tag: libs-extern_LO-BASE-INTEGRATION-DEV300_M101
|
|
We don't define any dummy DllMain() any more so no need to include
<windows.h>. That was necessary just to get the types used by the
DllMain() prototype.
|
|
|
|
|
|
It caused compilation errors even thanks to our workarounds for BOOL
clashes.
|
|
|
|
Notes:
split repo tag: libs-extern_LO-BASE-INTEGRATION-DEV300_m98
|
|
|
|
|
|
reenable this after tweaking deliver to not remove a link unless symlink would
work
|
|
|
|
|