i18n/hggettext

tests: escape bytes setting MSB in input of grep for portability

GNU grep (2.21-2 or later) assumes that input is encoded in LC_CTYPE,
and input is binary if it contains byte sequence not valid for that
encoding.

For example, if locale is configured as C, a byte setting most
significant bit (MSB) makes such GNU grep show "Binary file <FILENAME>
matches" message instead of matched lines unintentionally.

This behavior is recognized as a bug, and fixed in GNU grep 2.25-1 or
later. But some distributions are shipped with such buggy version
(e.g. Ubuntu xenial, which is used by launchpad buildbot).

    http://debbugs.gnu.org/cgi/bugreport.cgi?bug=19230
    https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=800670
    http://packages.ubuntu.com/xenial/grep

This causes failure of test-commit-interactive.t, which applies grep
on CP932 byte sequence since 1111e84de635.

But, explicit setting LC_CTYPE for CP932 might cause another problem,
because it can't be assumed that all environment running Mercurial
tests allows arbitrary locale setting.

To resolve this issue, this patch escapes bytes setting MSB in input
of grep.

For this purpose:

  - str.encode('string-escape') isn't useful, because it escapes also
    control code (less than 0x20), and makes EOL handling complicated

  - "f --hexdump" isn't useful, because it isn't line-oriented

  - "sed -n" seems reasonable, but "sed" itself sometimes causes
    portability issue, too (e.g. 900767dfa80d or afb86ee925bf)

This patch is posted with "stable" flag, because 1111e84de635 is on
stable branch.

#!/usr/bin/env python
#
# hggettext - carefully extract docstrings for Mercurial
#
# Copyright 2009 Matt Mackall <[e-mail]> and others
#
# This software may be used and distributed according to the terms of the
# GNU General Public License version 2 or any later version.

# The normalize function is taken from pygettext which is distributed
# with Python under the Python License, which is GPL compatible.

"""Extract docstrings from Mercurial commands.

Compared to pygettext, this script knows about the cmdtable and table
dictionaries used by Mercurial, and will only extract docstrings from
functions mentioned therein.

Use xgettext like normal to extract strings marked as translatable and
join the message cataloges to get the final catalog.
"""

import os, sys, inspect


def escape(s):
    # The order is important, the backslash must be escaped first
    # since the other replacements introduce new backslashes
    # themselves.
    s = s.replace('\\', '\\\\')
    s = s.replace('\n', '\\n')
    s = s.replace('\r', '\\r')
    s = s.replace('\t', '\\t')
    s = s.replace('"', '\\"')
    return s


def normalize(s):
    # This converts the various Python string types into a format that
    # is appropriate for .po files, namely much closer to C style.
    lines = s.split('\n')
    if len(lines) == 1:
        s = '"' + escape(s) + '"'
    else:
        if not lines[-1]:
            del lines[-1]
            lines[-1] = lines[-1] + '\n'
        lines = map(escape, lines)
        lineterm = '\\n"\n"'
        s = '""\n"' + lineterm.join(lines) + '"'
    return s


def poentry(path, lineno, s):
    return ('#: %s:%d\n' % (path, lineno) +
            'msgid %s\n' % normalize(s) +
            'msgstr ""\n')


def offset(src, doc, name, default):
    """Compute offset or issue a warning on stdout."""
    # Backslashes in doc appear doubled in src.
    end = src.find(doc.replace('\\', '\\\\'))
    if end == -1:
        # This can happen if the docstring contains unnecessary escape
        # sequences such as \" in a triple-quoted string. The problem
        # is that \" is turned into " and so doc wont appear in src.
        sys.stderr.write("warning: unknown offset in %s, assuming %d lines\n"
                         % (name, default))
        return default
    else:
        return src.count('\n', 0, end)


def importpath(path):
    """Import a path like foo/bar/baz.py and return the baz module."""
    if path.endswith('.py'):
        path = path[:-3]
    if path.endswith('/__init__'):
        path = path[:-9]
    path = path.replace('/', '.')
    mod = __import__(path)
    for comp in path.split('.')[1:]:
        mod = getattr(mod, comp)
    return mod


def docstrings(path):
    """Extract docstrings from path.

    This respects the Mercurial cmdtable/table convention and will
    only extract docstrings from functions mentioned in these tables.
    """
    mod = importpath(path)
    if mod.__doc__:
        src = open(path).read()
        lineno = 1 + offset(src, mod.__doc__, path, 7)
        print poentry(path, lineno, mod.__doc__)

    functions = list(getattr(mod, 'i18nfunctions', []))
    functions = [(f, True) for f in functions]

    cmdtable = getattr(mod, 'cmdtable', {})
    if not cmdtable:
        # Maybe we are processing mercurial.commands?
        cmdtable = getattr(mod, 'table', {})
    functions.extend((c[0], False) for c in cmdtable.itervalues())

    for func, rstrip in functions:
        if func.__doc__:
            src = inspect.getsource(func)
            name = "%s.%s" % (path, func.__name__)
            lineno = func.func_code.co_firstlineno
            doc = func.__doc__
            if rstrip:
                doc = doc.rstrip()
            lineno += offset(src, doc, name, 1)
            print poentry(path, lineno, doc)


def rawtext(path):
    src = open(path).read()
    print poentry(path, 1, src)


if __name__ == "__main__":
    # It is very important that we import the Mercurial modules from
    # the source tree where hggettext is executed. Otherwise we might
    # accidentally import and extract strings from a Mercurial
    # installation mentioned in PYTHONPATH.
    sys.path.insert(0, os.getcwd())
    from mercurial import demandimport; demandimport.enable()
    for path in sys.argv[1:]:
        if path.endswith('.txt'):
            rawtext(path)
        else:
            docstrings(path)