| Commit message (Collapse) | Author | Age |
|
|
|
|
| |
--
MOS_MIGRATED_REVID=106597904
|
|
|
|
|
|
|
| |
Convert to Skylark values when destructuring a sequence or map.
--
MOS_MIGRATED_REVID=106591523
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=106588003
|
|
|
|
|
|
|
| |
Progress towards #445.
--
MOS_MIGRATED_REVID=106585570
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=106584736
|
|
|
|
|
|
|
| |
location. Then, if unavailable, look relative to the workspace.
--
MOS_MIGRATED_REVID=106536282
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=106533345
|
|
|
|
|
|
|
|
|
| |
This was using -i to replace WORKSPACE in-place, but OS X requires -i to take an
argument ('' in this case) and Linux expects `sed -i''` and OS X expects
`sed -i ''`.
--
MOS_MIGRATED_REVID=106530804
|
|
|
|
|
|
|
|
|
|
|
| |
attributes that would otherwise be skipped by
default policy.
This is the simplistic start to a user-controllable
enforcement policy API.
--
MOS_MIGRATED_REVID=106530210
|
|
|
|
|
|
|
| |
Makes the the inevitable null pointer from unboxing null easier to diagnose.
--
MOS_MIGRATED_REVID=106525458
|
|
|
|
|
|
|
| |
the BUILD documentation.
--
MOS_MIGRATED_REVID=106515647
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=106515187
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=106504113
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=106499960
|
|
|
|
|
|
|
| |
again on OS X.
--
MOS_MIGRATED_REVID=106498988
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=106493557
|
|
|
|
|
|
|
| |
Special-case the return statement to avoid throwing an exception.
--
MOS_MIGRATED_REVID=106488391
|
|
|
|
|
|
|
| |
e.g. for i in list: pass
--
MOS_MIGRATED_REVID=106488146
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=106487671
|
|
|
|
|
|
|
|
| |
There's no timeline right now, but it's unusable on OSX and Oracle has no
plans of fixing it.
--
MOS_MIGRATED_REVID=106486015
|
|
|
|
|
|
|
| |
--
Change-Id: Ic5b08394e042fee48bf079620ecb18426455d838
Reviewed-on: https://bazel-review.googlesource.com/2230
MOS_MIGRATED_REVID=106480667
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
RuleContext returned fragments for the target configuration, even when Skylark requested fragments for the host configuration.
This Cl solves this bug. Since injecting a custom BuildConfiguration into out tests is surprisingly difficult, I tested this fix manually:
I wrote a custom bzl file with
def custom_rule_impl(ctx):
print("target = {}, host = {}".format(ctx.fragments.cpp.cpu, ctx.host_fragments.cpp.cpu))
and built the rule with --cpu ppc.
Output before the fix: target = ppc, host = ppc.
Output after the fix: target = ppc, host = k8.
--
MOS_MIGRATED_REVID=106479714
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=106459854
|
|
|
|
|
|
|
| |
created for flags that were renamed, but which were specified on the command line using their new name.
--
MOS_MIGRATED_REVID=106458953
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=106444993
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
*** Reason for rollback ***
Breaking O(150) tests on the blaze-2015.10.27-1 nightly TGP []
*** Original change description ***
Make -DNS_BLOCK_ASSERTIONS=1 by default when c=opt on objc builds.
--
MOS_MIGRATED_REVID=106435189
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
declare required configuration fragments: from
*everything* to *nothing*.
Now that all builtin rules properly declare their fragments,
the "backwards compatibility" concern that inspired the
original behavior is no longer needed.
This impacts, for example, filegroup rules, which have
nothing to declare.
--
MOS_MIGRATED_REVID=106433791
|
|
|
|
|
|
| |
--
Reviewed-on: https://github.com/bazelbuild/bazel/pull/524
MOS_MIGRATED_REVID=106431355
|
|
|
|
|
|
|
|
|
| |
that we don't abort evaluation until badly behaved builder has started.
Also fix the same race in MemoizingEvaluatorTest#interruptAfterFailFails (although that race would lead to spurious passes not fails).
--
MOS_MIGRATED_REVID=106427172
|
|
|
|
|
|
|
| |
sandboxing.
--
MOS_MIGRATED_REVID=106426071
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=106418732
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=106418596
|
|
|
|
|
|
|
| |
Should fix build on OSX.
--
MOS_MIGRATED_REVID=106415882
|
|
|
|
|
|
|
| |
RELNOTES: Bazel does strict validation of include files now to ensure correct incremental builds. If you see compilation errors when building C++ code, please make sure that you explicitly declare all header files in the srcs or hdrs attribute of your cc_* targets and that your cc_* targets have correct "deps" on cc_library's that they use.
--
MOS_MIGRATED_REVID=106410969
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=106410378
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=106410161
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=106409855
|
|
|
|
|
|
|
| |
sandboxing was enabled, by adding "jni_dep.h" to "hdrs" in //examples/android/java/bazel:jni_dep.
--
MOS_MIGRATED_REVID=106408122
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=106407898
|
|
|
|
|
|
|
|
|
|
| |
philwo@philwo:/tmp/bazel$ ./output/bazel query 'labels(srcs, @androidndk//:arm-linux-androideabi-4.9-v7a-gnu-libstdcpp-toolchain_files)'
INFO: Empty results
This caused compiliation failures when building //examples/android/java/bazel:hello_world with sandboxing enabled.
--
MOS_MIGRATED_REVID=106405986
|
|
|
|
|
|
|
|
|
|
|
|
| |
//examples/android/java/bazel:hello_world with sandboxing enabled.
external/androidndk/ndk/toolchains/arm-linux-androideabi-4.9/prebuilt/linux-x86_64/bin/../lib/gcc/arm-linux-androideabi/4.9/../../../../arm-linux-androideabi/bin/ld: error: cannot open crtbegin_so.o: No such file or directory
external/androidndk/ndk/toolchains/arm-linux-androideabi-4.9/prebuilt/linux-x86_64/bin/../lib/gcc/arm-linux-androideabi/4.9/../../../../arm-linux-androideabi/bin/ld: error: cannot open crtend_so.o: No such file or directory
external/androidndk/ndk/toolchains/arm-linux-androideabi-4.9/prebuilt/linux-x86_64/bin/../lib/gcc/arm-linux-androideabi/4.9/../../../../arm-linux-androideabi/bin/ld: error: cannot find -lc
external/androidndk/ndk/toolchains/arm-linux-androideabi-4.9/prebuilt/linux-x86_64/bin/../lib/gcc/arm-linux-androideabi/4.9/../../../../arm-linux-androideabi/bin/ld: error: cannot find -ldl
--
MOS_MIGRATED_REVID=106405880
|
|
|
|
|
|
|
|
|
| |
all files that go through the plmerge routines. It turns out we use plmerge to convert string files from text to binary.
RELNOTES:none
--
MOS_MIGRATED_REVID=106403889
|
|
|
|
|
|
|
| |
aar does not contain the transitive closure of the target, as this is a common misconception.
--
MOS_MIGRATED_REVID=106396742
|
|
|
|
|
|
|
|
|
| |
This makes Android builds slightly faster and avoids the "Modification date is in the future" warnings by javac and removes the sources of devtools/common/options from the binary.
incrementaldeployment is not pre-compiled yet.
--
MOS_MIGRATED_REVID=106391321
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=106389517
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=106382513
|
|
|
|
|
|
|
|
|
| |
Constants.ANDROID_DEP_PREFIX.
This regrettably entails adding the options parser to the embedded tools in the Bazel binary, but that's only a temporary measure until the embedded binaries are compiled (and not provided in source from)
--
MOS_MIGRATED_REVID=106379028
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
During bootstrap a number of tools are linked manually. These
link steps should take LDFLAGS to bootstrap bazel with custom
toolchain.
These tools are build-runfiles, process-wrapper and
namespace-sandbox.
cc_link() routine in compile.sh already takes LDFLAGS but not
the link steps for the tools listed above. They should all do so.
A developer can now do this to build Bazel:
CC="some-gcc" CXX="some-g++" LDFLAGS="-Lsome-glibc" \
JNI_LD_ARGS="-Lsome-glibc" ./compile.sh
--
Change-Id: Ifd1a03e989266a7fe3f1f92a7a44093b135fdc18
Reviewed-on: https://bazel-review.googlesource.com/2211
MOS_MIGRATED_REVID=106378256
|
|
|
|
|
|
|
| |
($python_precompile, etc.)
--
MOS_MIGRATED_REVID=106352835
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=106333445
|