aboutsummaryrefslogtreecommitdiffhomepage
path: root/src/main/java/com/google/devtools/build/lib/rules
Commit message (Collapse)AuthorAge
* Remove support for ProtocolBuffers2.Gravatar kaipi2017-10-06
| | | | PiperOrigin-RevId: 171013687
* Remove dead code.Gravatar ajmichael2017-10-06
| | | | | RELNOTES: None PiperOrigin-RevId: 170899433
* Documentation fix for 'minimum_os_version' attributeGravatar cparsons2017-10-06
| | | | | RELNOTES: None. PiperOrigin-RevId: 170722989
* Put feature arguments after command line options for LTOBackendActionGravatar Googler2017-10-06
| | | | | | | For consistency with compile actions. RELNOTES: None PiperOrigin-RevId: 170702636
* LocationExpander: always require options to be passed inGravatar ulfjack2017-10-02
| | | | | | | | | Also update CommandHelper to split the heuristic label expansion code path from the normal code path. Progress on #2475. PiperOrigin-RevId: 170675702
* Make JavaToolchain use the new Expander APIGravatar ulfjack2017-10-02
| | | | | | Progress on #2475. PiperOrigin-RevId: 170671644
* Prevent overriding android_library srcless deps configuration value from ↵Gravatar jingwen2017-10-02
| | | | | | | HostTransition. RELNOTES: None. PiperOrigin-RevId: 170549656
* PiperOrigin-RevId: 170539405Gravatar Googler2017-10-02
|
* Use intermediate compiled resource symbols for linking rather than static ↵Gravatar Googler2017-10-02
| | | | | | | resource library APK's. RELNOTES: none PiperOrigin-RevId: 170517806
* Switch ExtraActionFactory over to the new Expander interfaceGravatar ulfjack2017-09-29
| | | | PiperOrigin-RevId: 170494940
* Move expansion functionality to a new classGravatar ulfjack2017-09-29
| | | | | | Progress on #2475. PiperOrigin-RevId: 170473111
* Pass CppSemantics down to the CppLinkActionBuilderGravatar hlopko2017-09-29
| | | | | | | | | Currently CppLinkActionBuilder is not using CppSemantics, but it will when we use full CppCompileAction for linkstamp compiles. This cl is a preparation for that. RELNOTES: None. PiperOrigin-RevId: 170467826
* BEP: correctly report AliasConfiguredTargetsGravatar Klaus Aehlig2017-09-29
| | | | | | | | | When reporting the completion of a target specified by an alias, report the label of the alias, not that of the target being aliased to. Change-Id: If8416ceef73b01b7531ffa0012251f25a4e9f062 PiperOrigin-RevId: 170466076
* Fix various linter errorsGravatar ruperts2017-09-29
| | | | | RELNOTES: None. PiperOrigin-RevId: 170418147
* Add flag to Blaze to double-check correct desugaring in Android deploy.jarsGravatar kmb2017-09-29
| | | | | | (without affecting other uses of singlejar) PiperOrigin-RevId: 170411730
* Remove compilation and bundling attributes from apple_binary and ↵Gravatar cparsons2017-09-29
| | | | | | | apple_static_library. RELNOTES: apple_binary and apple_static_library no longer support compilation attributes such as 'srcs'. If this breaks any existing targets, you may migrate all such attributes to a new objc_library target and depend on that objc_library target via the 'deps' attribute of apple_binary or apple_static_library. PiperOrigin-RevId: 170373794
* Small fixes to android_instrumentation_test.Gravatar ajmichael2017-09-29
| | | | | | | | | BazelJUnitRunner swallows argv, so we cannot pass args to the android test runner as flags. Also, trim the trailing newline off of the name of the test suite class. RELNOTES: None PiperOrigin-RevId: 170354472
* Remove the null lipo context collector from the objc rules, which do not use ↵Gravatar cpeyser2017-09-29
| | | | | | LIPO. Remove the check that requires a context collector to be present. PiperOrigin-RevId: 170341259
* Make windows_export_all_symbols feature work for cc_binaryGravatar Yun Peng2017-09-28
| | | | | | | | | | | | | | | Now Bazel can also export symbols when building dynamic library from cc_binary. The interface library generated can be accessed by interface_library output group. The DEF file can still be accessed by def_file output group even when windows_export_all_symbols feature is not specified. This is useful when users want to filter symbols in DEF file before using it, for example, working around the 64K symbols number limit. Change-Id: I5b4dae0840e20037c00d500181c40b5faedfdcd8 PiperOrigin-RevId: 170330409
* Automated rollback of commit 411039319c1c67f2b9c8a7ada9e0a11d9bd4023f.Gravatar elenairina2017-09-28
| | | | | | | | | | | | | | | | | | | | *** Reason for rollback *** Breaks coverage for android_test (N/A). Can be reproduced with unknown commit. *** Original change description *** Rollforward change of Java coverage logic. RELNOTES: None. *** Original change description *** Automated rollback of commit 8d6fc64b18c7e35b93f5c43dae1dbd2f8cae2147. PiperOrigin-RevId: 170322801
* Polish the interface of CppCompileActionBuilderGravatar hlopko2017-09-28
| | | | | | | Some of the setters didn't return the builder, this cl fixes that. RELNOTES: None. PiperOrigin-RevId: 170313892
* Make the state in RuleContext explicitGravatar ulfjack2017-09-28
| | | | | | | | | | | | | | | | | | | | | This isn't ideal - RuleContext should not have state, but this ended up happening between adding a cache and refactoring how make variables are discovered. I have carefully traced back all callers that provide custom make variable suppliers and added an init call to their rule initialization. Note that the ConfigurationMakeVariableContext is _cached_, so callers that call in without any make variable suppliers and then call again with them would get the context from the previous call. We now enforce that the ConfigurationMakeVariableContext is only initialized once, and that this happens before any usage, which is slightly better than the previous state, where initialization was silently ignored on any subsequent call. Progress on #2475. PiperOrigin-RevId: 170312285
* Set --experimental_allow_android_library_deps_without_srcs to false by ↵Gravatar jingwen2017-09-28
| | | | | | | default in AndroidConfiguration. RELNOTES: android_library targets are no longer allowed to use deps to export targets implicitly; please use android_library.exports instead. PiperOrigin-RevId: 170243241
* Fix comment typo.Gravatar gregce2017-09-28
| | | | PiperOrigin-RevId: 170240695
* Expose JavaRuleOutputJarsProvider in java_common.provider, when available.Gravatar Googler2017-09-28
| | | | | | | | | | | | I'm not attempting to fix b/65618333 here, just handling one case currently breaking users (JavaInfo created via java_common.compile). My temporary workaround attempt to expose this information in the soy custom rule failed (unknown commit) -- to fix users we really need java_common changes. RELNOTES: Expose output jars and jdeps in java_common.provider, when available. PiperOrigin-RevId: 170236096
* Add 'compiler' and 'libc' attributes to cc_toolchain. If platform/toolchainGravatar cpeyser2017-09-28
| | | | | | | resolution is used, use these attribute values to choose a CToolchain from --crosstool_top instead of --compiler and --glibc. PiperOrigin-RevId: 170217186
* Limit the use of LinkCommandLine#getBuildInfoHeaderArtifacts and #getLinkstampsGravatar hlopko2017-09-28
| | | | | RELNOTES: None. PiperOrigin-RevId: 170203679
* Simplify make variable expansion call sitesGravatar ulfjack2017-09-28
| | | | | | Progress on #2475. PiperOrigin-RevId: 170197341
* Rewrite CppHelper linkopts expansion to take a list and return a listGravatar ulfjack2017-09-27
| | | | | | Progress on #2475. PiperOrigin-RevId: 170187908
* Simplify RuleContext APIGravatar ulfjack2017-09-27
| | | | | | | | | Use ConfigurationMakeVariableContext instead of passing in lists of MakeVariableSuppliers. Progress on #2475. PiperOrigin-RevId: 170182945
* Reorganise the srcs_version attribute of py_binary/py_test/py_library rules:Gravatar Googler2017-09-27
| | | | | | | | | | | make the default for srcs_version be "PY2AND3" (while leaving default_python_version at "PY2"), since that's a more likely situation and a more obvious error condition. (Having PY2 as the default caused 2to3 to be used a lot more than necessary, which in turn can both hide and cause bugs.) Also, use attribute value restrictions to limit srcs_version and default_python_version to sensible values for the specific rule type. PiperOrigin-RevId: 170174546
* remove unused code for unused aapt flagsGravatar ajmichael2017-09-27
| | | | | RELNOTES: None PiperOrigin-RevId: 170136145
* Ensure that SpawnResults make their way back to the SpawnActions that caused ↵Gravatar ruperts2017-09-27
| | | | | | | them to be created. RELNOTES: None. PiperOrigin-RevId: 170058295
* Move ExternalPackageUtil to a new lib.repository package and simplify itGravatar ulfjack2017-09-26
| | | | | | | Move the nested Exception classes to top-level classes, remove unused functionality and move functionality only used in one place to that place. PiperOrigin-RevId: 170041246
* Rollforward change of Java coverage logic.Gravatar elenairina2017-09-26
| | | | | | | | | | RELNOTES: None. *** Original change description *** Automated rollback of commit 8d6fc64b18c7e35b93f5c43dae1dbd2f8cae2147. PiperOrigin-RevId: 170038845
* Add an option to control building of transitive py_binary runfiles trees. ,Gravatar Benjamin Peterson2017-09-26
| | | | | | | | | | | | | | | | | ) Currently, "bazel build :script" constructs the bazel-bin/python-prog.runfiles tree. We'd like to do anyway with this behavior but doing so is a backwards-incompatible change. To facilitate a transition, this CL adds an option --experimental_build_transitive_python_runfiles to control the behavior. In the example above, "bazel build --noexperimental_build_transitive_python_runfiles :script" won't construct bazel-bin/python-prog.runfiles. Change-Id: If4d5a84c956a0bbac9067dcf38a00e5732c450f2 PiperOrigin-RevId: 170038245
* Windows: Add static_link_msvcrt and related featuresGravatar Yun Peng2017-09-26
| | | | | | | | | | | | | | | This is to replace using USE_DYNAMIC_CRT env variable to configure msvcrt linking in CROSSTOOL. If user applies static_link_msvcrt feature to a specific target, Bazel will choose the correct options for statically linking msvcrt. If static_link_msvcrt is not specified, Bazel uses options for dynamically linking msvcrt by default. https://msdn.microsoft.com/en-us/library/2kzt1wy3.aspx Change-Id: Ia078dfb528de9ffdd8a11d392db9eb3f34463b09 PiperOrigin-RevId: 170021927
* Denittize CppModel and CppLinkActionBuilderGravatar hlopko2017-09-26
| | | | | | | Fly by fixes. RELNOTES: None. PiperOrigin-RevId: 170014092
* Move --plugin to JavaOptions and remove --plugin_copt.Gravatar mstaib2017-09-26
| | | | | | | | | | | | | | | --plugin, though once used for C++, is currently a Java-specific flag. --plugin_copt is currently a total no-op, and has been for a long time. Moving these to the Java fragment is a little neater and helps get one step closer to enforcing LateBoundDefault fragment access. Additionally, since the "no plugins with duplicate names" restriction was added to work with plugin_copt, this restriction can be lifted. It no longer adds any value. RELNOTES: None. PiperOrigin-RevId: 169981221
* Simplify CcLinkParams#addCcLibraryGravatar hlopko2017-09-25
| | | | | | | Some of the arguments are not used or are always passed blank. RELNOTES: None. PiperOrigin-RevId: 169899161
* Remove unnecessary collectLinkstamp switchGravatar hlopko2017-09-25
| | | | | | | All callers passed true, so there's no need to have the boolean there. RELNOTES: None. PiperOrigin-RevId: 169888322
* Fix cc_fake_binaries with linkstampingGravatar hlopko2017-09-25
| | | | | | | | In https://github.com/bazelbuild/bazel/commit/b7a9ecd08c958e9513ea234b470a0f86e89e1acd I introduced a bug when linkstamping outputs would no longer be stored in $TEST_TMPDIR. This cl fixes that. RELNOTES: None. PiperOrigin-RevId: 169882831
* Remove product name precomputed value.Gravatar janakr2017-09-25
| | | | PiperOrigin-RevId: 169723724
* Inject BlazeDirectories into RepositoryDelegatorFunction directly, instead ↵Gravatar janakr2017-09-25
| | | | | | | | of depending on a precomputed value. BlazeDirectories don't change over the lifetime of the Blaze server, and certainly not over the lifetime of a SkyframeExecutor instance, which already had a reference to them. Step 2. PiperOrigin-RevId: 169711443
* Inject BlazeDirectories into FdoSupportFunction directly, instead of ↵Gravatar janakr2017-09-25
| | | | | | | | depending on a precomputed value. BlazeDirectories don't change over the lifetime of the Blaze server, and certainly not over the lifetime of a SkyframeExecutor instance, which already had a reference to them. The goal is getting rid of the precomputed value entirely, but since this change necessitated a fair number of testing changes, I'm mailing this out as a first step. PiperOrigin-RevId: 169705474
* Create AndroidManifest.xml symlink if user provides a manifest not namedGravatar jingwen2017-09-25
| | | | | | | | | | "AndroidManifest.xml". aapt requires the manifest to be named "AndroidManifest.xml" (unknown commit) This allows users to provide manifests with custom names. RELNOTES: Android rules no longer restrict the manifest file to be named "AndroidManifest.xml". PiperOrigin-RevId: 169695062
* Contingent on a flag --experimental_objc_provider_from_linked, apple_binary ↵Gravatar cparsons2017-09-25
| | | | | | | and apple_static_library no longer propagate ObjcProvider. RELNOTES: None. PiperOrigin-RevId: 169690934
* Use FeatureConfiguration to compute linkstamping compile command lineGravatar hlopko2017-09-25
| | | | | | | | | Before, linkstamping compile actions were hardcoded in bazel and assumed gcc/clang and bash. This cl removes gcc/clang assumptions by using feature configuration. RELNOTES: None. PiperOrigin-RevId: 169685949
* Add more thorough error reporting behavior for when config_setting has ↵Gravatar juliexxia2017-09-22
| | | | | | multiple constraint_values from the same constraint_setting (re: #350) PiperOrigin-RevId: 169577576
* Allow config_settings to match on constraint_values to a target_platform. ↵Gravatar juliexxia2017-09-21
| | | | | | This is on the way to making select() work with constraint_values re: #305. PiperOrigin-RevId: 169454982
generated by cgit v1.2.3 (git 2.39.1) at 2024-07-08 01:45:35 +0000