| Commit message (Collapse) | Author | Age |
... | |
|
|
|
|
|
|
|
| |
--
Change-Id: Ia6c94304fded25de48eda2060c5ec11368ff172c
Reviewed-on: https://cr.bazel.build/8991
PiperOrigin-RevId: 148286095
MOS_MIGRATED_REVID=148286095
|
|
|
|
|
|
| |
--
PiperOrigin-RevId: 148281228
MOS_MIGRATED_REVID=148281228
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
__file__ is not necessarily a absoulte path, fix and rename the function
for adding UNC prefix.
Since UNC prefix doesn't work with Python 2, skip adding it if running
in Python 2
--
Change-Id: I274dc0127c1427740c6824533b15299a066cdd83
Reviewed-on: https://cr.bazel.build/9050
PiperOrigin-RevId: 148281049
MOS_MIGRATED_REVID=148281049
|
|
|
|
|
|
|
|
| |
when detecting the installed Xcode versions.
--
PiperOrigin-RevId: 148277514
MOS_MIGRATED_REVID=148277514
|
|
|
|
|
|
|
|
| |
is true.
--
PiperOrigin-RevId: 148277221
MOS_MIGRATED_REVID=148277221
|
|
|
|
|
|
| |
--
PiperOrigin-RevId: 148275963
MOS_MIGRATED_REVID=148275963
|
|
|
|
|
|
|
|
|
|
| |
We currently have no need to discern between strings quoted with ' or ". While it could be
nice for something one day (and may have been in the past), it's yagni now. Removing the
distinction simplifies string concatenation.
--
PiperOrigin-RevId: 148273400
MOS_MIGRATED_REVID=148273400
|
|
|
|
|
|
|
|
| |
java.* and sun.* classes cannot be desugared directly because there are hard coded restrictions in the JVM that prevent using loading or using reflection on them. Add a --core_library flag that rewrites package names to have a __desugar__ prefix when reading them, and strips the prefix when writing them back out.
--
PiperOrigin-RevId: 148273386
MOS_MIGRATED_REVID=148273386
|
|
|
|
|
|
|
|
| |
compiler flags.
--
PiperOrigin-RevId: 148273368
MOS_MIGRATED_REVID=148273368
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
a boring naive synchronized implementation.
There's a race condition in my design of the lockless data structure. I haven't been able to come up with a lockless algorithm that actually works, and the naive one seems to be fine. In Blaze's usage, performance actually isn't super important so the naive implementation is fine.
Consider three threads and a MultisetSemaphore with 2 max unique values.
T1: acquireAll({a, b})
T2: acquireAll({a, c})
T3: acquireAll({a, d})
For the for-loop before the 'acquire' call, suppose:
-T1 wins the race to acquire 'a' [1] and also wants to acquire 'b' [1]
-T2 loses the race to acquire 'a' [2] and also wants to acquire 'c' [1]
-T3 loses the race to acquire 'a' [2] and also wants to acquire 'd' [1]
So then in [3] we have:
-T1 tries to acquire 2 permits
-T2 tries to acquire 1 permit
-T3 tries to acquire 1 permit
Suppose the execution order in [3] is T2, T3, T1. So that means we then have T1 still at [3] and both T2 and T3 at [4], which is a deadlock.
[1] https://github.com/bazelbuild/bazel/blob/fa96b04f6c8ca6b6b3464727672267133e852959/src/main/java/com/google/devtools/build/lib/concurrent/MultisetSemaphore.java#L184
[2] https://github.com/bazelbuild/bazel/blob/fa96b04f6c8ca6b6b3464727672267133e852959/src/main/java/com/google/devtools/build/lib/concurrent/MultisetSemaphore.java#L191
[3] https://github.com/bazelbuild/bazel/blob/fa96b04f6c8ca6b6b3464727672267133e852959/src/main/java/com/google/devtools/build/lib/concurrent/MultisetSemaphore.java#L199
[4] https://github.com/bazelbuild/bazel/blob/fa96b04f6c8ca6b6b3464727672267133e852959/src/main/java/com/google/devtools/build/lib/concurrent/MultisetSemaphore.java#L210
--
PiperOrigin-RevId: 148272171
MOS_MIGRATED_REVID=148272171
|
|
|
|
|
|
|
|
| |
android binaries
--
PiperOrigin-RevId: 148270755
MOS_MIGRATED_REVID=148270755
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Fixed failures by removing the (expected) failing XML reports from the path, to stop Jenkins from reporting a failure despite passing tests.
After removing the XMLs I see jenkins passing where it was previously failing: http://ci.bazel.io/job/Gerrit-bazel-tests/419/ (caused by an un-clean version of this change https://bazel-review.googlesource.com/c/9039/)
<Rollback history below>
*** Original change description ***
Automated [] rollback of commit 9822151a2ab0574b3d7a884f623b6b260258cd6b.
*** Reason for rollback ***
Jenkins reports the tests as failing. See #2554 for more details.
--
PiperOrigin-RevId: 148268951
MOS_MIGRATED_REVID=148268951
|
|
|
|
|
|
|
|
|
| |
e.g. Error Prone flags, since VanillaJavaBuilder doesn't
support them.
--
PiperOrigin-RevId: 148263963
MOS_MIGRATED_REVID=148263963
|
|
|
|
|
|
|
|
|
|
| |
classes
were converted to dex.
--
PiperOrigin-RevId: 148255629
MOS_MIGRATED_REVID=148255629
|
|
|
|
|
|
| |
--
PiperOrigin-RevId: 148253141
MOS_MIGRATED_REVID=148253141
|
|
|
|
|
|
|
|
|
|
|
| |
These targets are unneeded and do not build without an android_sdk_repository
set up.
This issue was identified in https://github.com/bazelbuild/bazel/issues/2559.
--
PiperOrigin-RevId: 148251416
MOS_MIGRATED_REVID=148251416
|
|
|
|
|
|
|
|
| |
there is one) worry about it.
--
PiperOrigin-RevId: 148249223
MOS_MIGRATED_REVID=148249223
|
|
|
|
|
|
| |
--
PiperOrigin-RevId: 148244816
MOS_MIGRATED_REVID=148244816
|
|
|
|
|
|
|
|
|
|
|
|
| |
And blog about it :)
Fixes #1390.
--
Change-Id: I138f1df0d18ab61bfa01c5bf3a3c483d33da1464
Reviewed-on: https://cr.bazel.build/8222
PiperOrigin-RevId: 148217151
MOS_MIGRATED_REVID=148217151
|
|
|
|
|
|
| |
--
PiperOrigin-RevId: 148211866
MOS_MIGRATED_REVID=148211866
|
|
|
|
|
|
|
| |
--
Reviewed-on: https://cr.bazel.build/8970
PiperOrigin-RevId: 148206918
MOS_MIGRATED_REVID=148206918
|
|
|
|
|
|
|
|
|
|
|
| |
This PR addresses comment by @gregestren in https://github.com/bazelbuild/bazel/pull/2531
Closes #2546.
--
Reviewed-on: https://github.com/bazelbuild/bazel/pull/2546
PiperOrigin-RevId: 148201730
MOS_MIGRATED_REVID=148201730
|
|
|
|
|
|
|
|
| |
found.
--
PiperOrigin-RevId: 148199034
MOS_MIGRATED_REVID=148199034
|
|
|
|
|
|
|
|
|
|
| |
Removed obsoleted TODOs and use more character matcher for package name validation
--
Change-Id: I7e8b69e34b0befe8a81c7c32924299790b6c56d0
Reviewed-on: https://cr.bazel.build/9052
PiperOrigin-RevId: 148195881
MOS_MIGRATED_REVID=148195881
|
|
|
|
|
|
| |
https://github.com/google/error-prone-javac/archive/9-dev-r3297-3.zip
Change-Id: I78b073c8db4a525d19b73650ed8d3a12b10ff86a
|
|
|
|
|
|
|
|
| |
apply on the top level.
--
PiperOrigin-RevId: 148155171
MOS_MIGRATED_REVID=148155171
|
|
|
|
|
|
|
|
| |
flaky
--
PiperOrigin-RevId: 148139717
MOS_MIGRATED_REVID=148139717
|
|
|
|
|
|
|
|
|
|
| |
Skylark modules.
With this change, we no longer have to manually register modules whose generation should be generated. Manual registration turned out to be prone to errors since engineers didn't know about it or simply forgot to register their modules. As a result, we had ~20 modules that featured nice documentation texts, but that were not being picked up by the documentation pipeline.
--
PiperOrigin-RevId: 148136776
MOS_MIGRATED_REVID=148136776
|
|
|
|
|
|
|
|
| |
Note that I also have to make some changes to cpp files to add the ability to specify a suffix for linked artifacts. This helps to avoid an artifact conflict between proto_library and J2ObjcAspect acting on proto_library, where both will create an archive file. J2ObjcAspect will simply generate a "_j2objc.a" archive instead of ".a" archive.
--
PiperOrigin-RevId: 148135630
MOS_MIGRATED_REVID=148135630
|
|
|
|
|
|
|
|
|
| |
- Print developer_dir when an error occurs during getting a specific xcode's SDK info
- Increase process timeouts from 5 seconds to 30 seconds. (There are some preliminary reports that users on slower Macs may be reaching this timeout on occasion)
--
PiperOrigin-RevId: 148135624
MOS_MIGRATED_REVID=148135624
|
|
|
|
|
|
|
|
| |
In Skylark, "Target" refers to TransitiveInfoCollection.class, which contains the real documentation string.
--
PiperOrigin-RevId: 148132811
MOS_MIGRATED_REVID=148132811
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
*** Reason for rollback ***
rules_closure needs to set go/be#java_plugin.generates_api to 1 for dagger
*** Original change description ***
Disable Java header compilation by default in Bazel.
Enabling this by default broke rules_closure at HEAD:
http://ci.bazel.io/view/Dashboard/job/rules_closure/BAZEL_VERSION=HEAD,PLATFORM_NAME=linux-x86_64/711/console
--
PiperOrigin-RevId: 148132758
MOS_MIGRATED_REVID=148132758
|
|
|
|
|
|
|
|
|
|
|
|
| |
*** Reason for rollback ***
Rollforward with fix for android_integration_test.sh. android_integration_test.sh should no longer make assumptions about the SDK api_level that is installed.
Note that it does still assume NDK api_level 24 and support libraries 24.0.0 and android_sdk_repository itself requires build_tools_version >= 24.0.3. I will try and loosen the first two of those dependencies of this test at some point.
--
PiperOrigin-RevId: 148130764
MOS_MIGRATED_REVID=148130764
|
|
|
|
|
|
| |
--
PiperOrigin-RevId: 148125885
MOS_MIGRATED_REVID=148125885
|
|
|
|
|
|
|
|
| |
flag.
--
PiperOrigin-RevId: 148115586
MOS_MIGRATED_REVID=148115586
|
|
|
|
|
|
| |
--
PiperOrigin-RevId: 148109339
MOS_MIGRATED_REVID=148109339
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
*** Reason for rollback ***
To deprecate that flag, we need to fix #2539 first.
*** Original change description ***
Deprecate the legacy_java_test flag, now that nolegacy_java_test has been the default for quite a while now.
RELNOTES[INC]: legacy_java_test flag will not have any effect now.
--
PiperOrigin-RevId: 148107667
MOS_MIGRATED_REVID=148107667
|
|
|
|
|
|
|
|
| |
and toolchain contexts.
--
PiperOrigin-RevId: 148104218
MOS_MIGRATED_REVID=148104218
|
|
|
|
|
|
|
|
|
|
| |
descriptors
these fields currently can't return null/None (since the @AutoValue will reject it) and there doesn't seem to be a reason to think that they might in the future so i think this is safe to remove.
--
PiperOrigin-RevId: 148099927
MOS_MIGRATED_REVID=148099927
|
|
|
|
|
|
| |
--
PiperOrigin-RevId: 148092963
MOS_MIGRATED_REVID=148092963
|
|
|
|
|
|
|
|
|
| |
Enabling this by default broke rules_closure at HEAD:
http://ci.bazel.io/view/Dashboard/job/rules_closure/BAZEL_VERSION=HEAD,PLATFORM_NAME=linux-x86_64/711/console
--
PiperOrigin-RevId: 148092235
MOS_MIGRATED_REVID=148092235
|
|
|
|
|
|
|
|
|
|
|
|
| |
...through the ExtendedEventHandler. The main use case of the extended
event handler is to allow posting of BuildEvents. So it is implicit that
they are all good to be posted. Declare this.
--
Change-Id: I9fcb2f33e57e05c3a05509b1121e44396f28c168
Reviewed-on: https://cr.bazel.build/9011
PiperOrigin-RevId: 148090465
MOS_MIGRATED_REVID=148090465
|
|
|
|
|
|
| |
Also, rename rules in third_party/protobuf/3.0.0/BUILD to match upstream protobuf.
Change-Id: I343982e752722ebb2461fb08e722558cea2f06b9
|
|
|
|
|
|
|
|
|
|
|
| |
Some projects fail [1] to build after the change that renamed extdir to extclasspath [2].
[1] http://ci.bazel.io/job/rules_appengine/BAZEL_VERSION=HEAD,PLATFORM_NAME=linux-x86_64/564/console
[2] 8153cb80f4a4424640e694cb6ed4792dcc193171
--
PiperOrigin-RevId: 148085503
MOS_MIGRATED_REVID=148085503
|
|
|
|
|
|
| |
--
PiperOrigin-RevId: 148083461
MOS_MIGRATED_REVID=148083461
|
|
|
|
|
|
|
|
| |
by the documentation generation in the future.
--
PiperOrigin-RevId: 148081562
MOS_MIGRATED_REVID=148081562
|
|
|
|
|
|
|
|
|
|
|
|
| |
Not in all situation, having an EventHandler allows detailed enough
reporting on events during the build. Therefore add a new interface
that additionally allows posting arbitrary objects on the event bus.
--
Change-Id: I6ff3eebcad5f36b52c38d3f7ac20df8af2706821
Reviewed-on: https://cr.bazel.build/8793
PiperOrigin-RevId: 148078891
MOS_MIGRATED_REVID=148078891
|
|
|
|
|
|
| |
--
PiperOrigin-RevId: 148075887
MOS_MIGRATED_REVID=148075887
|
|
|
|
|
|
|
|
| |
--
Change-Id: I9e1483b9e99751b03d332110db023d99c045cb52
Reviewed-on: https://cr.bazel.build/9012
PiperOrigin-RevId: 148069886
MOS_MIGRATED_REVID=148069886
|
|
|
|
|
|
| |
--
PiperOrigin-RevId: 148042501
MOS_MIGRATED_REVID=148042501
|