i18n/hggettext
author Siddharth Agarwal <sid0@fb.com>
Tue, 27 May 2014 14:27:41 -0700
changeset 21809 e250b8300e6e
parent 12823 80deae3bc5ea
child 29170 2516bba643e7
permissions -rwxr-xr-x
parsers: inline fields of dirstate values in C version Previously, while unpacking the dirstate we'd create 3-4 new CPython objects for most dirstate values: - the state is a single character string, which is pooled by CPython - the mode is a new object if it isn't 0 due to being in the lookup set - the size is a new object if it is greater than 255 - the mtime is a new object if it isn't -1 due to being in the lookup set - the tuple to contain them all In some cases such as regular hg status, we actually look at all the objects. In other cases like hg add, hg status for a subdirectory, or hg status with the third-party hgwatchman enabled, we look at almost none of the objects. This patch eliminates most object creation in these cases by defining a custom C struct that is exposed to Python with an interface similar to a tuple. Only when tuple elements are actually requested are the respective objects created. The gains, where they're expected, are significant. The following tests are run against a working copy with over 270,000 files. parse_dirstate becomes significantly faster: $ hg perfdirstate before: wall 0.186437 comb 0.180000 user 0.160000 sys 0.020000 (best of 35) after: wall 0.093158 comb 0.100000 user 0.090000 sys 0.010000 (best of 95) and as a result, several commands benefit: $ time hg status # with hgwatchman enabled before: 0.42s user 0.14s system 99% cpu 0.563 total after: 0.34s user 0.12s system 99% cpu 0.471 total $ time hg add new-file before: 0.85s user 0.18s system 99% cpu 1.033 total after: 0.76s user 0.17s system 99% cpu 0.931 total There is a slight regression in regular status performance, but this is fixed in an upcoming patch.

#!/usr/bin/env python
#
# hggettext - carefully extract docstrings for Mercurial
#
# Copyright 2009 Matt Mackall <mpm@selenic.com> 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)