check-code: make casting style check more precise

log: make file log slow path usable on large repos Running "hg log <pattern or directory>" on large repos took a very, very long time because it first read ctx.files() for every commit before even starting to process the results. This change makes the ctx.files() check lazy, which makes the command start producing results immediately.

plan9: update util.py for cpython 2.7 build

parse_manifest: rewrite to use memchr memchr is usually smarter than a simple for loop. With gcc 4.4.6 and glibc 2.12 on x86-64, for a 20 MB, 200,000 file manifest, parse_manifest goes from 0.116 seconds to 0.095 seconds.

parsers: correctly handle a failed allocation

parsers: use Py_INCREF safely

parsers: state is a char, not an int

setup: check if mercurial/util.h has been modified Before, if mercurial/util.h was modified but *.c wasn't, no extension was rebuilt.

util.h: add stdint basic type definitions MS C compiler v15 doesn't have stdint.h. Add basic int types those are defined in stdint.h.

hgweb: add revset syntax support to search This mode is used when all the conditions are met: - 'reverse(%s)' % query string can be parsed to a revset tree - this tree has depth more than two, i.e. the query has some part of revset syntax used - the repo can be actually matched against this tree, i.e. it has only existent function/operators and revisions/tags/bookmarks specified are correct - no revset regexes are used in the query (strings which start with 're:') - only functions explicitly marked as safe in revset.py are used in the query Add several new tests for different parsing conditions and exception handling.

revset: add a whitelist of DoS-safe symbols 'Safe' here means that they can't be used for a DoS attack for any given input.

revset: add helper function to get functions used in a revset parse tree Will be used to determine whether all functions used in a hgweb search query are allowed there.

revset: add helper function to get revset parse tree depth Will be used to determine if a hgweb search query actually uses any revset syntax.

parsers: use a lookup table to convert hex to binary This is a hotspot for parse_manifest. With this patch, for a 20 MB, 200,000 file manifest, parse_manifest goes down from 0.153 seconds to 0.116.

revlog: remove _chunkbase since it is no longer used This was introduced in 2011 for the lwcopy feature but never actually got used. A similar hook can easily be reintroduced if needed in the future.

revlog: move chunk cache preload from revision to _chunks In case we don't have a cached text already, add the base rev to the list passed to _chunks. In the cached case this also avoids unnecessarily preloading the chunk for the cached rev.

revlog._chunks: inline getchunk We do this in a somewhat hacky way, relying on the fact that our sole caller preloads the cache right before calling us. An upcoming patch will make this more sensible. For a 20 MB manifest with a delta chain of > 40k, perfmanifest goes from 0.49 seconds to 0.46.

revlog.revision: fix cache preload for inline revlogs Previously the length of data preloaded did not account for the interleaved io contents. This meant that we'd sometimes have cache misses in _chunks despite the preloading. Having a correctly filled out cache will become essential in an upcoming patch.

revlog: add a fast method for getting a list of chunks This moves _chunkraw into the loop. Doing that improves revlog decompression -- in particular, manifest decompression -- significantly. For a 20 MB manifest which is the result of a > 40k delta chain, hg perfmanifest improves from 0.55 seconds to 0.49 seconds.

perfmanifest: allow and require passing in a rev Previously perfmanifest would only test tip.

perfmanifest: fix cache invalidation The manifest cache is now stored in _mancache. The revlog cache is still stored in _cache.

lrucachedict: implement clear()

merge with stable

bundle: fix performance regression when bundling file changes (issue4031) Somewhere before 2.7, a change [ec896f9e8894] was committed that entailed a large performance regression when bundling (and therefore remote cloning) repositories. For each file in the repository, it would recompute the set of needed changesets even though it is the same for all files. This computation would dominate bundle runtimes according to profiler output (by 10x or more).