Age | Commit message (Collapse) | Author |
|
I am not sure this really fixes the problem, but it is the only
difference between dmake and gbuild builds I can see.
Change-Id: I96fa4120dc2a8221a75e150a62582aebda98f505
|
|
Change-Id: Idc838315696666c8fcb08d28ed4c0b991e185008
|
|
Change-Id: I12e98ac9fc49ef2007914324006a396d183b778c
|
|
ignoring null tag in filter/source/svg/presentation_engine.js
Change-Id: I84e077373d872031308b12c2b336e7e4972973f1
|
|
Some of the areas are guesses I've added after seeing them, whoever feels reponsible
for whichever part of the code feel free to adjust them.
Change-Id: I2192de84d51cc2bc7c28fa84019d38b465985d15
|
|
Change-Id: I18281f9cdf4c20ab4f939b2f7d3b0a6c3bf5387b
|
|
Also - verified that codemaker touches both .hdl and .hpp regardless
of whether the contents changed in anyway under gbuild.
Change-Id: I391e9cf952c61a6c59087cd7cd0ddc8c54a14df3
|
|
Change-Id: I72ed637ecec5aff7528ff724b2665e22650c224b
|
|
Factor out the common part from the ANDROID_*_GCC.mk files into an anrdoid.mk.
Make the DISABLE_DYNLOADING-specific bits conditional.
Change-Id: If41065fbc096dfae01819ac8fabf4b21e95545ca
|
|
IN this branch these changes are not conditional. Unclear yet whether
this is what we finally will want to use or not. Maybe should make
these changes conditional and do this stuff in master instead?
Change-Id: I379d570a0e00648d295c675fd90eba6594ba3182
|
|
Change-Id: I5f4f1a5b78333aa294c334870412599f76573aef
|
|
Change-Id: Ie05eeb89e79663674e223339918239facf49603a
|
|
Change-Id: I4c0d2d408c5de770bbad54294a109743fa35b3f6
|
|
The plugin is intentionally built using a custom Makefile,
because it's used by gbuild, so I don't want to build the plugin
using gbuild too. It is also intentionally not placed under workdir/,
as that is cleaned by 'make clean', the plugin is cleaned only
by 'make distclean', so that cleaning it doesn't cause ccache misses.
No actual functionality in the plugin itself yet.
Change-Id: Ic05eba8d6260eec123c9e699eb5385abfe1b832f
|
|
Change-Id: I3539bfcfbbb43f44a6fc8fae60bda84267938116
|
|
This reverts commit 3fc40953c635185f8bcd1b58a1d68d28526558d4.
Change-Id: I6d11212cc260e03e5c83a0e7be93ed66e1ab452f
|
|
Change-Id: I3e02afbd1b75641bdc90443e7c07018c113a971a
|
|
Change-Id: Ia532558eb25c3745f9d43303358b785994f389fe
|
|
Change-Id: I05d1c557967000be2fb7128b43d6e8c6005a7892
|
|
Change-Id: Ib86a5406c28560d766ecb48cb8f74ac9f9b43c4d
|
|
Change-Id: I7c38722c98300f419e6129b46c89b68f46ce4173
|
|
Change-Id: I0916051214d8631198723933799e8e27fadef993
|
|
Change-Id: Ic06deb22da77428ef88c1071af5d7b8bf9dc77f7
|
|
Change-Id: I6384a9d86123ab354414cf6e9366b4f9e46f1c1f
|
|
Change-Id: Ie5969a0f5772676f91c29fe946217d8d31bf514a
|
|
Change-Id: I1e4e701ba4883862ddf79ef45d4f7521346a067d
|
|
Or rather, re-purpose that for consistency (and rename original to
gb_ExternalProject_use_external_project), to abstract over the
system/internal status of dependencies of external projects.
Use it in libcdr and replace exisiting uses in apache-commons.
Change-Id: Ie144600688fa884b5b6faa986c6b95bdfc1ee15c
|
|
Change-Id: Idc8b0adf1a9ed26b2e0bec3454fed0ee34e277aa
|
|
Change-Id: I1674da35bf1d756bda5c4b991cc586462325c838
|
|
unfortunately the ExternalProject uses a StaticLibrary built in libwpg.
Change-Id: Ie4a8933247edbd2d53f697432a0848a05216237f
|
|
Change-Id: Ie6eed251a5034d9761abca75506b677af394d945
|
|
Change-Id: I02ca44cc33d1d8e1ebc9f158f859c9218d2ed832
|
|
Change-Id: I042269aabaa2e9952c5d2f41cf93e55d654f6313
|
|
There needs to be a way to depend on the output of other external
packages.
Change-Id: Ic0f200b6d6b6c0968a28434ba96f1a2f1efa527e
|
|
and also after stuff that references external symbols, as it only looks
for symbols it has encountered so far in the object files. Stuff that
comes after the specified library then results in symbols not found
despite the lib being on the commandline earlier.
(another solution than reordering the arguments would be to use
-Wl,-all_load)
Change-Id: If4add0d3f81b13874c1620406fa40fcbd25b7cf0
|
|
Change-Id: Iee09fc308097ab0a5f73e321da30b7e301b9f4dd
|
|
Change-Id: I9ba3739f1693f5427c9ea49f2b6016f364f444ff
|
|
Change-Id: I59ced6720bdfd2df7a8ed7d9247887a09beeb8fd
|
|
Change-Id: I6ef6a7dea7a5f594b907546e8dff39c3640d7509
|
|
Change-Id: I55257cc5dbdffdc2dd9f9222a120f0a322dda7c8
|
|
Change-Id: If84003c6a95cee0fd6e4e985ec0655f7074b0851
|
|
Besides the usual problem this causes with rpath (security), clang
also doesn't like empty arguments in -Wl.
Change-Id: I5c0896ae06ffcf6d2a268bf98a2b3f031cb8a70f
|
|
Change-Id: If6c607fc9f0be1f7a71eda8d555399451938c5fc
|
|
Change-Id: Iea8385aa9213ccde7e6650cb934361597d508250
|
|
Change-Id: Ice20b6903f7bf26547a7a3e2bf3cec6c831a6734
|
|
Change-Id: I02bd49c6f6b252c6797292b8e4a9bc319fe0947e
|
|
The whole idea of having two paths to look for schemas (one local, for
module, and one global, in $(OUTDIR)) is bad anyway, because there is no
way to do it in XSLT without warning if the checked schema does not
exist in the first path. I see two ways out of that:
1. rewrite schema_val.xsl in a decent language
2. split the node reference check in schema_val.xsl into two phases: the
first phase returns a list of subpaths to check, a helper script
processes that, finds the real paths and feeds it to the second phase
XSL, which does the internal check.
Change-Id: Icfe3f9ec0960cfd3423c1223ad8f495d7634b25a
|
|
Change-Id: Ie3c9bddcb4760d2fe2195c1ca0de7520e57d705f
|
|
Change-Id: Ie91376bf9c52238febdc15720bb7f4cd62ee2e08
|
|
Change-Id: I09a14fd50cf69cec8693b7058d105ee5f2998e02
|