aboutsummaryrefslogtreecommitdiffhomepage
path: root/src/main/java/com/google/devtools/build/lib/packages/GlobCache.java
Commit message (Collapse)AuthorAge
* Replace all usages of Blaze's Preconditions class with guava.Gravatar tomlu2017-11-09
| | | | | | | | Blaze had its own class to avoid GC from varargs array creation for the precondition happy path. Guava now (mostly) implements these, making it unnecessary to maintain our own. This change was almost entirely automated by search-and-replace. A few BUILD files needed fixing up since I removed an export of preconditions from lib:util, which was all done by add_deps. There was one incorrect usage of Preconditions that was caught by error prone (which checks Guava's version of Preconditions) that I had to change manually. PiperOrigin-RevId: 175033526
* Automated conversion to Java 8Gravatar laurentlb2017-06-30
| | | | | | | With a few manual fixes for readability. RELNOTES: None. PiperOrigin-RevId: 160582556
* Global cleanup change.Gravatar Googler2017-01-25
| | | | | | -- PiperOrigin-RevId: 145473478 MOS_MIGRATED_REVID=145473478
* Replace UnixGlob's package glob prefetching functionality with anGravatar Googler2016-12-02
| | | | | | | | optional method in FileSystem. Custom FileSystem implementations can use this to provide their own implementation of glob prefetching. -- MOS_MIGRATED_REVID=140736304
* Add the option --experimental_max_directories_to_eagerly_visit_in_globbing. ↵Gravatar Janak Ramakrishnan2016-10-05
| | | | | | | The first legacy glob that a package requires will, if this option is enabled, cause up to that many directories to be eagerly visited by a glob(['**']). The results are thrown away for memory reasons. -- MOS_MIGRATED_REVID=135148361
* Automated [] rollback of commit 846a5ab98fc26d72024890fdb79a5d3bc6a5a1ba + ↵Gravatar Nathan Harmata2016-08-23
| | | | | | | | | | | | | | | manual rollback of [] *** Reason for rollback *** Depot has been fixed / is in the process of being fixed. See the work tracked on [] *** Original change description *** Automated [] rollback of commit bb5d5efb4b50710241b5b374eb67084f4bf08278. -- MOS_MIGRATED_REVID=131095905
* Rollback of commit bb5d5efb4b50710241b5b374eb67084f4bf08278.Gravatar Tobias Werth2016-08-23
| | | | | -- MOS_MIGRATED_REVID=130941264
* RELNOTES: The string list returned by the skylark 'glob' function is now ↵Gravatar Nathan Harmata2016-08-18
| | | | | | | | | | | | | | sorted. Previously, it would return a list formed by concatenating the sorted results of each pattern in the 'includes' list. A bunch of cleanups and one bug fix: -Remove the unused-except-for tests GlobCache#globsUpToDate. This code has been dead for a very very long time, ever since we switched to using Skyframe. -Change the semantics of the 'glob' function as described above. -Change UnixGlob to return unsorted results. Document this in UnixGlob and GlobCache. -Change LegacyGlobber to conditionally return sorted results. Have users other than PackageFunction get sorted results (as described above). Have PackageFunction's use case get completely unsorted results, and have PackageFunction do the sorting itself. -Have PackageFunction's HybridGlobber unconditionally sort the glob result list. This ensure deterministic glob results, fixing a bug where the order of the elements of the result depended on the contents of the Skyframe graph, which of course depends on the sequence of incremental Blaze commands. -- MOS_MIGRATED_REVID=130540152
* Fix glob performance regression introduced by commit ↵Gravatar Nathan Harmata2016-03-17
| | | | | | | | | | | | | 3a95f353704dc2f7061e2c0786c2459ac1db0fd1. AbstractSet#removeAll has unexpected, yet oddly intentional (and documented), performance characteristics. Suppose we are evaluating 'set.removeAll(collection)' and 'collection.contains(x)' is 'O(e)'. Then 'set.removeAll(collection)' is 'O(set.size())' when 'set.size() <= collection.size()' and 'O(set.size()) * e' otherwise. When 'collection' is e.g. an ArrayList, 'e' is 'collection.size()' and so 'set.removeAll(collection)' is 'O(set.size() * collection.size())', which is bad. This meant we had poor performance when the excludes patterns of a glob matched more files than the includes patterns. Note that, while GlobCache#glob() *did* also use removeAll (potentially inefficiently), it was doing so for each list of exclude glob matches individually. So legacy globbing would have suboptimal performance for 'glob(includes=[i_1, i_2, ...i_k], excludes = [e_1, e_2, ..., e_j])' whenever the result of any e_i was larger than the union of all the includes matches. (But skyframe hybrid globbing has the performance issue when the union of the excludes matches is larger than the union of the includes matches, which is more likely to happen in practice.) I fixed this hypothetical problem too. -- MOS_MIGRATED_REVID=117367755
* Move the Globber interface into its own file.Gravatar Nathan Harmata2016-02-05
| | | | | -- MOS_MIGRATED_REVID=113893917
* Use Bazel Preconditions variant which avoids varargs array creationGravatar Mark Schaller2015-12-10
| | | | | | | Reduces garbage. -- MOS_MIGRATED_REVID=109914243
* Remove useless explicit 'stat' call on the BUILD file in directories ↵Gravatar Nathan Harmata2015-10-09
| | | | | | | encountered during legacy globbing (note that we end up doing this stat anyway in the 'locator.getBuildFileForPackage' call). The old motivation for this is obsolete; we have incremental correctness from Skyframe. -- MOS_MIGRATED_REVID=104982764
* Intern PackageIdentifiers as a memory optimization.Gravatar Miguel Alcon Pinto2015-10-01
| | | | | -- MOS_MIGRATED_REVID=104403040
* Rationalize copyright headersGravatar Damien Martin-Guillerez2015-09-25
| | | | | | | | | | | The headers were modified with `find . -type f -exec 'sed' '-Ei' 's|Copyright 201([45]) Google|Copyright 201\1 The Bazel Authors|' '{}' ';'` And manual edit for not Google owned copyright. Because of the nature of ijar, I did not modified the header of file owned by Alan Donovan. The list of authors were extracted from the git log. It is missing older Google contributors that can be added on-demand. -- MOS_MIGRATED_REVID=103938715
* Move PackageIdentifier to cmdlineGravatar Kristina Chodorow2015-08-31
| | | | | | | | This is necessary to have TargetResolver depend on it without making it depend on the packages target. First step of #389. -- MOS_MIGRATED_REVID=101790345
* Presize the LinkedHashSet to which we add glob results.Gravatar Nathan Harmata2015-08-28
| | | | | -- MOS_MIGRATED_REVID=101743615
* Make globs work in remote repositories.Gravatar Lukacs Berki2015-07-08
| | | | | | | | | | | This involved quite a few changes, mainly changing a bunch of places where we refer to packages by a PathFragment to PackageIdentifier. The only wart is the code in PathPackageLocator: ideally, it would just call into PackageLookupFunction. Unfortunately, it is (through globbing and Parser.include) called from within a Skyframe function, and we don't want to have two eval() calls going on at the same time, so we cannot use that. There is a potential correctness issue there: PathPackageLocator now assumes where external repositories are put and assumes that they are there when it gets control, but my understanding is that the associated RepositoryValue is always evaluated before, so it works out okay. -- MOS_MIGRATED_REVID=97751539
* Rollback of accidentally submitted change.Gravatar Lukacs Berki2015-07-07
| | | | | -- MOS_MIGRATED_REVID=97648982
* Make globs work in remote repositories.Gravatar Lukacs Berki2015-07-07
| | | | | | | | | | | This involved quite a few changes, mainly changing a bunch of places where we refer to packages by a PathFragment to PackageIdentifier. The only wart is the code in PathPackageLocator: ideally, it would just call into PackageLookupFunction. Unfortunately, it is (through globbing and Parser.include) called from within a Skyframe function, and we don't want to have two eval() calls going on at the same time, so we cannot use that. There is a potential correctness issue there: PathPackageLocator now assumes where external repositories are put and assumes that they are there when it gets control, but my understanding is that the associated RepositoryValue is always evaluated before, so it works out okay. -- MOS_MIGRATED_REVID=97647787
* Refactor the legacy globbing thread pool to make use of more modern ↵Gravatar Eric Fellheimer2015-06-29
| | | | | | | | | concurrency abstractions. Care is taken to maintain the invariant that the glob result returns after all of the work is done, even if there was an exception that cuts the task short. Interruption is an exception to this: In this case, the GlobCache later cancels the task and ensures it is done. -- MOS_MIGRATED_REVID=97000506
* Description redacted.Gravatar Googler2015-03-30
| | | | | -- MOS_MIGRATED_REVID=89740455
* Update from Google.Gravatar Han-Wen Nienhuys2015-02-25
-- MOE_MIGRATED_REVID=85702957
generated by cgit v1.2.3 (git 2.39.1) at 2024-07-05 12:17:17 +0000