contrib/check-commit

perf: add command for measuring revlog chunk operations

Upcoming commits will teach revlogs to leverage the new compression
engine API so that new compression formats can more easily be
leveraged in revlogs. We want to be sure this refactoring doesn't
regress performance. So this commit introduces "perfrevchunks" to
explicitly test performance of reading, decompressing, and
recompressing revlog chunks.

Here is output when run on the mozilla-unified repo:

$ hg perfrevlogchunks -c
! read
! wall 0.346603 comb 0.350000 user 0.340000 sys 0.010000 (best of 28)
! read w/ reused fd
! wall 0.337707 comb 0.340000 user 0.320000 sys 0.020000 (best of 30)
! read batch
! wall 0.013206 comb 0.020000 user 0.000000 sys 0.020000 (best of 221)
! read batch w/ reused fd
! wall 0.013259 comb 0.030000 user 0.010000 sys 0.020000 (best of 222)
! chunk
! wall 1.909939 comb 1.910000 user 1.900000 sys 0.010000 (best of 6)
! chunk batch
! wall 1.750677 comb 1.760000 user 1.740000 sys 0.020000 (best of 6)
! compress
! wall 5.668004 comb 5.670000 user 5.670000 sys 0.000000 (best of 3)

$ hg perfrevlogchunks -m
! read
! wall 0.365834 comb 0.370000 user 0.350000 sys 0.020000 (best of 26)
! read w/ reused fd
! wall 0.350160 comb 0.350000 user 0.320000 sys 0.030000 (best of 28)
! read batch
! wall 0.024777 comb 0.020000 user 0.000000 sys 0.020000 (best of 119)
! read batch w/ reused fd
! wall 0.024895 comb 0.030000 user 0.000000 sys 0.030000 (best of 118)
! chunk
! wall 2.514061 comb 2.520000 user 2.480000 sys 0.040000 (best of 4)
! chunk batch
! wall 2.380788 comb 2.380000 user 2.360000 sys 0.020000 (best of 5)
! compress
! wall 9.815297 comb 9.820000 user 9.820000 sys 0.000000 (best of 3)

We already see some interesting data, such as how much slower
non-batched chunk reading is and that zlib compression appears to be
>2x slower than decompression.

I didn't have the data when I wrote this commit message, but I ran this
on Mozilla's NFS-based Mercurial server and the time for reading with a
reused file descriptor was faster. So I think it is worth testing both
with and without file descriptor reuse so we can make informed
decisions about recycling file descriptors.

#!/usr/bin/env python
#
# Copyright 2014 Matt Mackall <[e-mail]>
#
# A tool/hook to run basic sanity checks on commits/patches for
# submission to Mercurial. Install by adding the following to your
# .hg/hgrc:
#
# [hooks]
# pretxncommit = contrib/check-commit
#
# The hook can be temporarily bypassed with:
#
# $ BYPASS= hg commit
#
# See also: https://mercurial-scm.org/wiki/ContributingChanges

from __future__ import absolute_import, print_function

import os
import re
import sys

commitheader = r"^(?:# [^\n]*\n)*"
afterheader = commitheader + r"(?!#)"
beforepatch = afterheader + r"(?!\n(?!@@))"

errors = [
    (beforepatch + r".*[(]bc[)]", "(BC) needs to be uppercase"),
    (beforepatch + r".*[(]issue \d\d\d",
     "no space allowed between issue and number"),
    (beforepatch + r".*[(]bug(\d|\s)", "use (issueDDDD) instead of bug"),
    (commitheader + r"# User [^@\n]+\n", "username is not an email address"),
    (commitheader + r"(?!merge with )[^#]\S+[^:] ",
     "summary line doesn't start with 'topic: '"),
    (afterheader + r"[A-Z][a-z]\S+", "don't capitalize summary lines"),
    (afterheader + r"[^\n]*: *[A-Z][a-z]\S+", "don't capitalize summary lines"),
    (afterheader + r"\S*[^A-Za-z0-9-_]\S*: ",
     "summary keyword should be most user-relevant one-word command or topic"),
    (afterheader + r".*\.\s*\n", "don't add trailing period on summary line"),
    (afterheader + r".{79,}", "summary line too long (limit is 78)"),
    (r"\n\+\n( |\+)\n", "adds double empty line"),
    (r"\n \n\+\n", "adds double empty line"),
    # Forbid "_" in function name.
    #
    # We skip the check for cffi related functions. They use names mapping the
    # name of the C function. C function names may contain "_".
    (r"\n\+[ \t]+def (?!cffi)[a-z]+_[a-z]",
     "adds a function with foo_bar naming"),
]

word = re.compile('\S')
def nonempty(first, second):
    if word.search(first):
        return first
    return second

def checkcommit(commit, node=None):
    exitcode = 0
    printed = node is None
    hits = []
    for exp, msg in errors:
        for m in re.finditer(exp, commit):
            end = m.end()
            trailing = re.search(r'(\\n)+$', exp)
            if trailing:
                end -= len(trailing.group()) / 2
            hits.append((end, exp, msg))
    if hits:
        hits.sort()
        pos = 0
        last = ''
        for n, l in enumerate(commit.splitlines(True)):
            pos += len(l)
            while len(hits):
                end, exp, msg = hits[0]
                if pos < end:
                    break
                if not printed:
                    printed = True
                    print("node: %s" % node)
                print("%d: %s" % (n, msg))
                print(" %s" % nonempty(l, last)[:-1])
                if "BYPASS" not in os.environ:
                    exitcode = 1
                del hits[0]
            last = nonempty(l, last)

    return exitcode

def readcommit(node):
    return os.popen("hg export %s" % node).read()

if __name__ == "__main__":
    exitcode = 0
    node = os.environ.get("HG_NODE")

    if node:
        commit = readcommit(node)
        exitcode = checkcommit(commit)
    elif sys.argv[1:]:
        for node in sys.argv[1:]:
            exitcode |= checkcommit(readcommit(node), node)
    else:
        commit = sys.stdin.read()
        exitcode = checkcommit(commit)
    sys.exit(exitcode)