aboutsummaryrefslogtreecommitdiffhomepage
path: root/src/main/java/com/google/devtools/build/lib/rules
Commit message (Collapse)AuthorAge
* Automated rollback of commit 4260c30a03a9b83d48a5e8690aca19cd80be4c38.Gravatar ulfjack2018-02-06
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | *** Reason for rollback *** Try again with fixes. *** Original change description *** Automated rollback of commit 10b0d8aa6b73a024cc007c5e075cb329add878ef. *** Reason for rollback *** Breaks Google-internal targets, sadly. *** Original change description *** Ban middlemen from runfiles artifacts. Previous changes have removed all middlemen from runfiles artifacts. This CL locks it down and removes various now-redundant *WithoutMiddlemen() methods from Runfiles. I put a check for middlemen in ConflictChecker.put, which should be a chokepoint for runfiles arti... *** PiperOrigin-RevId: 184661375
* PiperOrigin-RevId: 184564534Gravatar Googler2018-02-05
|
* Tag a few aspect classes with @AutoCodec and make CcProtoAspect abstract, ↵Gravatar janakr2018-02-05
| | | | | | since it's never instantiated on its own. PiperOrigin-RevId: 184554483
* Simplify RuleClassProviders by making most RuleSets have their own class.Gravatar lpino2018-02-05
| | | | PiperOrigin-RevId: 184540561
* Fix AndroidDevice to never put the unified launcher's runfiles middleman in ↵Gravatar Benjamin Peterson2018-02-05
| | | | | | | | | | | the runfiles tree. As I understand it, this is only a theoretical issue today because $unified_launcher is generally a flat file. (Flat files never have runfiles middlemen.) However, it's good to be future proof. Change-Id: If77edfa9dd7475ab93b19c62b08f8d86a77acbe6 PiperOrigin-RevId: 184540188
* Add type assertion that TreeFileArtifact's parentTreeArtifact member is aGravatar cpeyser2018-02-05
| | | | | | SpecialArtifact. PiperOrigin-RevId: 184539696
* Fix manifest_merger documentation link.Gravatar ajmichael2018-02-05
| | | | | RELNOTES: None PiperOrigin-RevId: 184538771
* Automated rollback of commit 17830c25a816d67ba79e783acaa6a085b516bd43.Gravatar jmmv2018-02-05
| | | | | | | | | | | | | | | *** Reason for rollback *** Causes Blaze to crash when building Exoblaze as shown in b/72936965. Confirmed as root cause by rolling back this CL, building a Blaze from HEAD, and successfully using it to build Exoblaze. *** Original change description *** C++: Remove last instatiation of CppModel outside CcLibraryHelper. RELNOTES:none PiperOrigin-RevId: 184528551
* C++: Remove last instatiation of CppModel outside CcLibraryHelper.Gravatar plf2018-02-05
| | | | | RELNOTES:none PiperOrigin-RevId: 184510731
* Convert "-" to "" when generating module nameGravatar Googler2018-02-05
| | | | | RELNOTES: None. PiperOrigin-RevId: 184492828
* RELNOTES: stop using --no-locals in android coverage buildsGravatar kmb2018-02-02
| | | | PiperOrigin-RevId: 184369667
* Remove documentation of ↵Gravatar dannark2018-02-02
| | | | | | | android_local_test.{resource_files,assets,assets_dir}. They are not used and their usage is unsupported. RELNOTES: None PiperOrigin-RevId: 184354541
* Add allowReturnNone to skylark annotationsGravatar Googler2018-02-02
| | | | PiperOrigin-RevId: 184348602
* Stop allowing generic LateBoundDefault value types. Such types are always ↵Gravatar janakr2018-02-02
| | | | | | | | either a Label or a List<Label>. We can easily enforce this through static type checking, so do it. This will help with LateBoundDefault serialization, since we don't have to serialize an arbitrary object. PiperOrigin-RevId: 184347100
* C++: Stops generating DWO files when PER_OBJECT_DEBUG_INFO feature is not ↵Gravatar plf2018-02-02
| | | | | | | enabled. RELNOTES:none PiperOrigin-RevId: 184303490
* Implement Merge logic for all new providers in JavaInfoGravatar dbabkin2018-02-02
| | | | | | | In scope of b/69780248 three new provider had been added to JavaInfo. This CL implements merge logic for its. RELNOTES:none PiperOrigin-RevId: 184273903
* Change how the libraries_to_link are expanded for archiverGravatar hlopko2018-02-02
| | | | | | | | | This makes object_file_group work correctly for archiving (So far object_file_group was only possible for linking, but with the introduction of tree artifact support, it can appear also for archiving). RELNOTES: Fix how libraries_to_link are expanded in the archiver command line. PiperOrigin-RevId: 184258745
* PiperOrigin-RevId: 184203570Gravatar Googler2018-02-01
|
* Remove unused code from bazel.Gravatar kaipi2018-02-01
| | | | | RELNOTES: PiperOrigin-RevId: 184191201
* Expand the docs on hdrs to better indicate what they are.Gravatar thomasvl2018-02-01
| | | | | | | This borrows from what is on cc_library, to hopefully better capture/indicate what should and shouldn't be listed in their attribute. PiperOrigin-RevId: 184173403
* Remove setting of $DISPLAY variable by android_instrumentation_test.Gravatar jingwen2018-02-01
| | | | | | | | | | This makes the android_instrumentation_test silent by default. If the emulator GUI is required, the following flag is used: `--action_env=DISPLAY=$DISPLAY # or whatever port the display is on` RELNOTES: None. PiperOrigin-RevId: 184172016
* Add a codec for FeatureConfiguration using AutoCodec. This is necessary forGravatar cpeyser2018-02-01
| | | | | | CppCompileAction serialization. PiperOrigin-RevId: 184141676
* Automated rollback of commit aa8ce5653f92473b331e500a3614245dae38dfc4.Gravatar hlopko2018-02-01
| | | | | | | | | | | | | | *** Reason for rollback *** Breaks //third_party/msvc crosstool *** Original change description *** RELNOTES: Fix how libraries to link is specified to archiver actions. start-lib/end-lib should not be passed to ar. Fix how the libraries to link are passed to ar by not using the generic feature "libraries_to_link" PiperOrigin-RevId: 184140734
* Allow CustomCommandLine's mapFn to expand each object into multiple items.Gravatar tomlu2018-02-01
| | | | | | This is needed to migrate JavaCompileAction away from CustomMultiArgv. PiperOrigin-RevId: 184136486
* Renamed android_instrumentation_test.instrumentation to ↵Gravatar jingwen2018-01-31
| | | | | | | android_instrumentation_test.test_app RELNOTES: None PiperOrigin-RevId: 184063294
* Improve safety of NestedSetFingerprintCache by detecting multiple instances ↵Gravatar tomlu2018-01-31
| | | | | | | | | | | | of the same mapFn class. This code tries to add protection against the user creating new mapFn instances per-rule. This would cause the nested set cache to be computed per-rule instead of shared across rule instances, causing memory bloat and slowdowns. Since this can only happen in native code, we can get away with detecting this and crashing blaze. I think this is a better choice than silently allowing it / falling back to slow computations. The user can override this behaviour by inheriting from CommandLineItem.CapturingMapFn, in which case the user is explicitly saying they assume responsibility for the number of instances of the mapFn the application will use. PiperOrigin-RevId: 184061642
* Change ConfiguredAspectFactory.create signature to take in aGravatar mjhalupka2018-01-31
| | | | | | | ConfiguredTargetAndTarget instead of a ConfiguredTarget. This is to assist in deprecating ConfiguredTarget.getTarget(). PiperOrigin-RevId: 184043491
* Make CcToolchainProvider structField methods resilient to null fieldsGravatar cparsons2018-01-31
| | | | | | | | structField methods must not throw an exception in all cases. Null fields were technically used for incomplete toolchain objects (such as the EMPTY_TOOLCHAIN_IS_ERROR object). RELNOTES: None. PiperOrigin-RevId: 184031908
* Add neverlink support to java_common.compile.Gravatar elenairina2018-01-31
| | | | | | | Fixes #3735. RELNOTES: java_common.compile supports neverlink PiperOrigin-RevId: 184017410
* Order the input-output map keys before processing its elements. This fixes ↵Gravatar kaipi2018-01-31
| | | | | | non-determinism in how the protos get bundled into the BundledProtos_X archives. PiperOrigin-RevId: 184009755
* Move BazelRuleClassProvider.LIPO_DATA_TRANSITION to its own class.Gravatar lpino2018-01-31
| | | | PiperOrigin-RevId: 183990860
* Remove the CppConfiguration field from CppCompileAction. MakingGravatar cpeyser2018-01-30
| | | | | | CppCompileAction smaller improves serialization. PiperOrigin-RevId: 183871454
* Rename Transition to ConfigurationTransition.Gravatar gregce2018-01-30
| | | | PiperOrigin-RevId: 183859414
* Drop CppCompileAction.actionContextGravatar ulfjack2018-01-30
| | | | | | It always set to the same value. PiperOrigin-RevId: 183843518
* Stop warning about resource filters with "deprecated" qualifiersGravatar Googler2018-01-30
| | | | | | | | | | | | | | | Resource qualifiers of the form "en_US" are deprecated for resources (in favor of qualifiers of the form "en-rUS") but are apparently the recommended (if not only) way of specifying locales in resource filter strings. Removing the warning (and reverting any changes that tried following it) should solve this problem - the original filter strings from the rule are passed unmodified to aapt, but these filters are still "fixed" so that android_ide_common (which doesn't support the "deprecated" qualifier form) can still be used to filter resources in analysis. RELNOTES: none PiperOrigin-RevId: 183830253
* Remove DynamicTransitionMapper, ConfigurationTransitionProxy.Gravatar gregce2018-01-30
| | | | PiperOrigin-RevId: 183826311
* C++: Removes last calls to CCLibraryHelper.build().Gravatar plf2018-01-30
| | | | | | | | | | In a follow-up CL, CcLibraryHelper will be split in two classes, one for compilation and one for linking. SKIP_KOKORO=mac-flakes RELNOTES:none PiperOrigin-RevId: 183824395
* Add alternative interface for LIPO data transitions, opt in native rules.Gravatar gregce2018-01-29
| | | | | | | | | | | This removes the need for ConfigurationTransitionProxy.DATA by providing a way for the C++ rule defs to directly inject the transition for all rules to use. Skylark attributes work differently, so they'll be addressed in another change. PiperOrigin-RevId: 183721293
* Expose ApkInfo and AndroidInstrumentationInfo to Skylark.Gravatar ajmichael2018-01-29
| | | | | | | This will enable us to write testing rules in Skylark. RELNOTES: None PiperOrigin-RevId: 183719720
* Allows @AutoCodec to use factory methods.Gravatar shahan2018-01-29
| | | | | | Generalizes @AutoCodec.Constructor to @AutoCodec.Instantiator. PiperOrigin-RevId: 183702768
* Remove quotes from argv in android_instrumentation_test stub template and ↵Gravatar jingwen2018-01-29
| | | | | | | add the argv to the java args RELNOTES: None. PiperOrigin-RevId: 183685024
* Move BazelRuleClassProvider.TESTING_SUPPORT to its own class in rules/test.Gravatar lpino2018-01-29
| | | | PiperOrigin-RevId: 183672444
* Support Android testing with Skylark-defined device rules.Gravatar ajmichael2018-01-26
| | | | | | | | | Do this by exposing DeviceBrokerInfo and a constructor for it in android_common. See AndroidInstrumentationTestTest for an example. RELNOTES: None PiperOrigin-RevId: 183432674
* Add a type that explicitly limits the CppSemantics implementations that can beGravatar cpeyser2018-01-26
| | | | | | | a member of CcProtoAspect. This is to prevent a heavy implementation like ObjcCppSemantics from weighing down AspectKey. PiperOrigin-RevId: 183408339
* C++: Refactors CLIF rule to go through CcLibraryHelper.Gravatar plf2018-01-26
| | | | | | | | | | | TBD: finish pulling out CLIF logic from the C++ rules implementation. This involves refactoring ArtifactCategory and CppSource.Type. CLIF does PIC compilation now. This detail should be irrelevant to users since CLIF only cares about the matched protobuf output file. SKIP_KOKORO=flakes RELNOTES:none PiperOrigin-RevId: 183396902
* Reduce garbage generated from JavaCompileAction#getArtifactGeneralizedLabel.Gravatar tomlu2018-01-26
| | | | PiperOrigin-RevId: 183380025
* Make experimental_inmemory_dotd_files not fail with local executionGravatar ulfjack2018-01-26
| | | | | | If actions are executed locally, silently fall back to reading the on-disk .d file. PiperOrigin-RevId: 183376432
* Add a few options that tweak the host configuration to parallel those that ↵Gravatar lberki2018-01-26
| | | | | | | | | | | do the same with the target one. I'm not exactly happy at this development, but we already have a host of --host_* options so it's only incremental badness. Fixes #4484. RELNOTES: None. PiperOrigin-RevId: 183375817
* Remove package restriction on android_tools_defaults_jar rule.Gravatar ajmichael2018-01-25
| | | | | RELNOTES: None PiperOrigin-RevId: 183286641
* Step 1 of removing android_sdk.annotations_jar.Gravatar ajmichael2018-01-25
| | | | | | | | | | It seems we don't actually need it for anything. The next step will be to remove the flag from the tools, to stop passing it in all android_sdk rules and then to delete it from android_sdk. RELNOTES: None PiperOrigin-RevId: 183285855
generated by cgit v1.2.3 (git 2.39.1) at 2024-07-04 06:15:54 +0000