tests/test-fastannotate-revmap.py

commit: improve the files field of changelog for merges

Currently, the files list of merge commits repeats all the deletions
(either actual deletions, or files that got renamed) that happened
between base and p2 of the merge. If p2 is the main branch, the list
can easily be much bigger than the change being merged.

This results in various problems worth improving:
- changelog is bigger than necessary
- `hg log directory` lists many unrelated merge commits, and `hg log
  -v -r commit` frequently fills multiple screens worth of files
- it possibly slows down adjustlinkrev, by forcing it to read more
  manifests, and that function can certainly be a bottleneck
- the server side of pulls can waste a lot of time simply opening the
  filelogs for pointless files (the constant factors for opening even
  a tiny filelog is apparently pretty bad)

So stop listing such files as described in the code.  Impacted merge
commits and their descendants get a different hash than they would
have without this. This doesn't seem problematic, except for
convert. The previous commit helped with that in the hg->hg case (but
if you do svn->hg twice from scratch, hashes can still change).

The rest of the description is numbers. I don't have much to report,
because recreating the files list of existing repositories is not
easy:
- debugupgradeformat and bundle/unbundle don't recreate the list
- export/import tends to choke quickly applying patches or on
  description that contain diffs,
- merge commits from the convert extension don't have the right files
  list for reasons orthogonal to the current commit
- replaying the merge with hg update/hg merge/hg revert --all/hg
  commit can end up failing in hg revert
- I wasn't sure that using debugsetparents + debugrebuilddirstate
  would really build the right thing

I measured commit time before and after this change, in a case with no
files filtered out, several files filtered out (no difference) and 5k
files filtered out (+1% time).

Recreating the 100 more recent merges in a private repo, the
concatenated uncompressed files lists goes from 1.12MB to
0.52MB. Excluding 3 merges that are not representative, then the size
goes from 570k to 15k.
I converted part of mozilla-central, and observed file list shrinking
quite a bit too, starting at the very first merge, 733641d9feaf, going
from 550 files to 10 files (although they have relatively few merges,
so they probably wouldn't care).

Differential Revision: https://phab.mercurial-scm.org/D6613

from __future__ import absolute_import, print_function

import os
import tempfile

from mercurial import (
    pycompat,
    util,
)

from hgext.fastannotate import error, revmap

if pycompat.ispy3:
    xrange = range

def genhsh(i):
    return pycompat.bytechr(i) + b'\0' * 19

def gettemppath():
    fd, path = tempfile.mkstemp()
    os.close(fd)
    os.unlink(path)
    return path

def ensure(condition):
    if not condition:
        raise RuntimeError('Unexpected')

def testbasicreadwrite():
    path = gettemppath()

    rm = revmap.revmap(path)
    ensure(rm.maxrev == 0)
    for i in xrange(5):
        ensure(rm.rev2hsh(i) is None)
    ensure(rm.hsh2rev(b'\0' * 20) is None)

    paths = [
        b'', b'a', None, b'b', b'b', b'c', b'c', None, b'a', b'b', b'a', b'a']
    for i in xrange(1, 5):
        ensure(rm.append(genhsh(i), sidebranch=(i & 1), path=paths[i]) == i)

    ensure(rm.maxrev == 4)
    for i in xrange(1, 5):
        ensure(rm.hsh2rev(genhsh(i)) == i)
        ensure(rm.rev2hsh(i) == genhsh(i))

    # re-load and verify
    rm.flush()
    rm = revmap.revmap(path)
    ensure(rm.maxrev == 4)
    for i in xrange(1, 5):
        ensure(rm.hsh2rev(genhsh(i)) == i)
        ensure(rm.rev2hsh(i) == genhsh(i))
        ensure(bool(rm.rev2flag(i) & revmap.sidebranchflag) == bool(i & 1))

    # append without calling save() explicitly
    for i in xrange(5, 12):
        ensure(rm.append(genhsh(i), sidebranch=(i & 1), path=paths[i],
                         flush=True) == i)

    # re-load and verify
    rm = revmap.revmap(path)
    ensure(rm.maxrev == 11)
    for i in xrange(1, 12):
        ensure(rm.hsh2rev(genhsh(i)) == i)
        ensure(rm.rev2hsh(i) == genhsh(i))
        ensure(rm.rev2path(i) == paths[i] or paths[i - 1])
        ensure(bool(rm.rev2flag(i) & revmap.sidebranchflag) == bool(i & 1))

    os.unlink(path)

    # missing keys
    ensure(rm.rev2hsh(12) is None)
    ensure(rm.rev2hsh(0) is None)
    ensure(rm.rev2hsh(-1) is None)
    ensure(rm.rev2flag(12) is None)
    ensure(rm.rev2path(12) is None)
    ensure(rm.hsh2rev(b'\1' * 20) is None)

    # illformed hash (not 20 bytes)
    try:
        rm.append(b'\0')
        ensure(False)
    except Exception:
        pass

def testcorruptformat():
    path = gettemppath()

    # incorrect header
    with open(path, 'wb') as f:
        f.write(b'NOT A VALID HEADER')
    try:
        revmap.revmap(path)
        ensure(False)
    except error.CorruptedFileError:
        pass

    # rewrite the file
    os.unlink(path)
    rm = revmap.revmap(path)
    rm.append(genhsh(0), flush=True)

    rm = revmap.revmap(path)
    ensure(rm.maxrev == 1)

    # corrupt the file by appending a byte
    size = os.stat(path).st_size
    with open(path, 'ab') as f:
        f.write(b'\xff')
    try:
        revmap.revmap(path)
        ensure(False)
    except error.CorruptedFileError:
        pass

    # corrupt the file by removing the last byte
    ensure(size > 0)
    with open(path, 'wb') as f:
        f.truncate(size - 1)
    try:
        revmap.revmap(path)
        ensure(False)
    except error.CorruptedFileError:
        pass

    os.unlink(path)

def testcopyfrom():
    path = gettemppath()
    rm = revmap.revmap(path)
    for i in xrange(1, 10):
        ensure(rm.append(genhsh(i),
                         sidebranch=(i & 1), path=(b'%d' % (i // 3))) == i)
    rm.flush()

    # copy rm to rm2
    rm2 = revmap.revmap()
    rm2.copyfrom(rm)
    path2 = gettemppath()
    rm2.path = path2
    rm2.flush()

    # two files should be the same
    ensure(len(set(util.readfile(p) for p in [path, path2])) == 1)

    os.unlink(path)
    os.unlink(path2)

class fakefctx(object):
    def __init__(self, node, path=None):
        self._node = node
        self._path = path

    def node(self):
        return self._node

    def path(self):
        return self._path

def testcontains():
    path = gettemppath()

    rm = revmap.revmap(path)
    for i in xrange(1, 5):
        ensure(rm.append(genhsh(i), sidebranch=(i & 1)) == i)

    for i in xrange(1, 5):
        ensure(((genhsh(i), None) in rm) == ((i & 1) == 0))
        ensure((fakefctx(genhsh(i)) in rm) == ((i & 1) == 0))
    for i in xrange(5, 10):
        ensure(fakefctx(genhsh(i)) not in rm)
        ensure((genhsh(i), None) not in rm)

    # "contains" checks paths
    rm = revmap.revmap()
    for i in xrange(1, 5):
        ensure(rm.append(genhsh(i), path=(b'%d' % (i // 2))) == i)
    for i in xrange(1, 5):
        ensure(fakefctx(genhsh(i), path=(b'%d' % (i // 2))) in rm)
        ensure(fakefctx(genhsh(i), path=b'a') not in rm)

def testlastnode():
    path = gettemppath()
    ensure(revmap.getlastnode(path) is None)
    rm = revmap.revmap(path)
    ensure(revmap.getlastnode(path) is None)
    for i in xrange(1, 10):
        hsh = genhsh(i)
        rm.append(hsh, path=(b'%d' % (i // 2)), flush=True)
        ensure(revmap.getlastnode(path) == hsh)
        rm2 = revmap.revmap(path)
        ensure(rm2.rev2hsh(rm2.maxrev) == hsh)

testbasicreadwrite()
testcorruptformat()
testcopyfrom()
testcontains()
testlastnode()