contrib/check-commit

posix: use getutf8char to handle OS X filename percent-escaping

This replaces an open-coded utf-8 parser that was ignoring subtle issues
like overlong encodings.

#!/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

import re, sys, os

errors = [
    (r"[(]bc[)]", "(BC) needs to be uppercase"),
    (r"[(]issue \d\d\d", "no space allowed between issue and number"),
    (r"[(]bug(\d|\s)", "use (issueDDDD) instead of bug"),
    (r"^# User [^@\n]+$", "username is not an email address"),
    (r"^# .*\n(?!merge with )[^#]\S+[^:] ",
     "summary line doesn't start with 'topic: '"),
    (r"^# .*\n[A-Z][a-z]\S+", "don't capitalize summary lines"),
    (r"^# .*\n[^\n]*: *[A-Z][a-z]\S+", "don't capitalize summary lines"),
    (r"^# .*\n.*\.\s+$", "don't add trailing period on summary line"),
    (r"^# .*\n.{78,}", "summary line too long (limit is 78)"),
    (r"^\+\n \n", "adds double empty line"),
    (r"^ \n\+\n", "adds double empty line"),
    (r"^\+[ \t]+def [a-z]+_[a-z]", "adds a function with foo_bar naming"),
]

node = os.environ.get("HG_NODE")

if node:
    commit = os.popen("hg export %s" % node).read()
else:
    commit = sys.stdin.read()

exitcode = 0
for exp, msg in errors:
    m = re.search(exp, commit, re.MULTILINE)
    if m:
        pos = 0
        for n, l in enumerate(commit.splitlines(True)):
            pos += len(l)
            if pos >= m.end():
                print "%d: %s" % (n, msg)
                print " %s" % l[:-1]
                if "BYPASS" not in os.environ:
                    exitcode = 1
                break

sys.exit(exitcode)