In other external projects using libtool, we fix that by patching configure,
resetting hardcode_libdir_flag_spec[_CXX] at the end of the linux*) case block
that sets the Linux-specific value. But here we run autoreconf in
ExternalProject_libassuan, so that patch in configure would be overwritten. The
relevant code in configure comes from autoconf boilerplate, so we cannot just do
the same patch in configure.ac. But we can reset hardcode_libdir_flag_spec
sufficiently late in configure.ac so that things still work as intended.
Disable tests that would build executabes linking against libgpgme.so, which in
turn links against the libassuan and libgpg-error libs, which would no longer be
found by the linker because of the dropped -rpath flags. (Alternatives might
be to pass in LD_LIBRARY_PATH or to link with --allow-shlib-undefined.)
Change-Id: I7e37abf802d213347bd80383b7980d85cf0762d4
Reviewed-on: https://gerrit.libreoffice.org/50960
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
In other external projects using libtool, we fix that by patching configure,
resetting hardcode_libdir_flag_spec[_CXX] at the end of the linux*) case block
that sets the Linux-specific value. But here we run autoreconf in
ExternalProject_libassuan, so that patch in configure would be overwritten. The
relevant code in configure comes from autoconf boilerplate, so we cannot just do
the same patch in configure.ac. But we can reset hardcode_libdir_flag_spec
sufficiently late in configure.ac so that things still work as intended.
Change-Id: Ic6c6123bcfe8bc2dac87812f919842519374abaa
Reviewed-on: https://gerrit.libreoffice.org/50959
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
The dependency of Library_vcl on qt5 libraries (via vcl/qt5/Qt5Font.cxx) is
there ever since b66a7cbd84 "QT5 first stab on
implementing CommonSalLayout", but no idea whether that is by design or is a
hack that should eventually be cleaned up.
Change-Id: I4cec702709b37069987ec68c035628b5172b3543
...containing replacements for global operator new/delete (that can be linked
into executables), but which is no longer used. The mail thread starting at
<https://lists.freedesktop.org/archives/libreoffice/2012-March/028690.html>
"operator new no longer routes through rtl_AllocMemory in libsalcpprt under
gbuild link rules" has the details of how this was used on some platforms (but
not on others) before the switch to gbuild, and has been "lost" ever since---but
apparently a loss not mourned much over the years.
For the SDK, c5f974287f "INTEGRATION: CWS jsc3:
#i62434# copy libsalcpprt.a" added the library (under Linux) and
6db9c5af96 "INTEGRATION: CWS jsc3: #i62434# extend
link options for executbales to link libsalcpprt.a, LINUX only" added its use to
odk/settings/settings.mk, but fc0ca57f2c
"INTEGRATION: CWS jsc21" removed that use again (for no documented reason). So
this is an incompatible change, but unlikely to actually affect any users of the
SDK.
Change-Id: Ia38b4c439f21fca3f5d9af7d1a34054e992054e9
Reviewed-on: https://gerrit.libreoffice.org/31810
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
It takes almost no time, but this way of doing it changes the
interpretation of the -d parameter; not sure how important that is.
Change-Id: I9d29cdf96d7006d5daea00bcc41e613fbd30aac8
Reviewed-on: https://gerrit.libreoffice.org/29707
Reviewed-by: Norbert Thiebaud <nthiebaud@gmail.com>
Tested-by: Norbert Thiebaud <nthiebaud@gmail.com>
using that and restricting the tests to the content of
instdir/program bring the runtime on out slave builder from
4 minutes+ to just above one second
added -d <path> to override INSTDIR
added -p to indicate parallelism (no limit for now, so be cautious
with the -d you point too
Change-Id: I14c140f54bea329b82960843a6db44bfaf6c7108
... so use readelf itself to check if it's an ELF file.
Also just skip the SDK documentation which is 2/3 of all files.
Change-Id: Icfd3023dd5b2ccc4f4e94fbb05cdd4eb05051f86
Stephan Bergmann