i18n/hggettext

perf: add command to benchmark bundle reading

Upcoming commits will be refactoring bundle2 I/O code.

This commit establishes a `hg perfbundleread` command that measures
how long it takes to read a bundle using various mechanisms.

As a baseline, here's output from an uncompressed bundle1
bundle of my Firefox repo (7,098,622,890 bytes):

! read(8k)
! wall 0.763481 comb 0.760000 user 0.160000 sys 0.600000 (best of 6)
! read(16k)
! wall 0.644512 comb 0.640000 user 0.110000 sys 0.530000 (best of 16)
! read(32k)
! wall 0.581172 comb 0.590000 user 0.060000 sys 0.530000 (best of 18)
! read(128k)
! wall 0.535183 comb 0.530000 user 0.010000 sys 0.520000 (best of 19)
! cg1 deltaiter()
! wall 0.873500 comb 0.880000 user 0.840000 sys 0.040000 (best of 12)
! cg1 getchunks()
! wall 6.283797 comb 6.270000 user 5.570000 sys 0.700000 (best of 3)
! cg1 read(8k)
! wall 1.097173 comb 1.100000 user 0.400000 sys 0.700000 (best of 10)
! cg1 read(16k)
! wall 0.810750 comb 0.800000 user 0.200000 sys 0.600000 (best of 13)
! cg1 read(32k)
! wall 0.671215 comb 0.670000 user 0.110000 sys 0.560000 (best of 15)
! cg1 read(128k)
! wall 0.597857 comb 0.600000 user 0.020000 sys 0.580000 (best of 15)

And from an uncompressed bundle2 bundle (6,070,036,163 bytes):

! read(8k)
! wall 0.676997 comb 0.680000 user 0.160000 sys 0.520000 (best of 15)
! read(16k)
! wall 0.592706 comb 0.590000 user 0.080000 sys 0.510000 (best of 17)
! read(32k)
! wall 0.529395 comb 0.530000 user 0.050000 sys 0.480000 (best of 16)
! read(128k)
! wall 0.491270 comb 0.490000 user 0.010000 sys 0.480000 (best of 19)
! bundle2 forwardchunks()
! wall 2.997131 comb 2.990000 user 2.270000 sys 0.720000 (best of 4)
! bundle2 iterparts()
! wall 12.247197 comb 10.670000 user 8.170000 sys 2.500000 (best of 3)
! bundle2 part seek()
! wall 11.761675 comb 10.500000 user 8.240000 sys 2.260000 (best of 3)
! bundle2 part read(8k)
! wall 9.116163 comb 9.110000 user 8.240000 sys 0.870000 (best of 3)
! bundle2 part read(16k)
! wall 8.984362 comb 8.970000 user 8.110000 sys 0.860000 (best of 3)
! bundle2 part read(32k)
! wall 8.758364 comb 8.740000 user 7.860000 sys 0.880000 (best of 3)
! bundle2 part read(128k)
! wall 8.749040 comb 8.730000 user 7.830000 sys 0.900000 (best of 3)

We already see some interesting data. Notably that bundle2 has
significant overhead compared to bundle1. This matters for e.g. stream
clone bundles, which can be applied at >1Gbps.

Differential Revision: https://phab.mercurial-scm.org/D1385

#!/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.
"""

from __future__ import absolute_import, print_function

import inspect
import os
import re
import sys


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')

doctestre = re.compile(r'^ +>>> ', re.MULTILINE)

def offset(src, doc, name, default):
    """Compute offset or issue a warning on stdout."""
    # remove doctest part, in order to avoid backslash mismatching
    m = doctestre.search(doc)
    if m:
        doc = doc[:m.start()]

    # 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 not path.startswith('mercurial/') and 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__:
            docobj = func # this might be a proxy to provide formatted doc
            func = getattr(func, '_origfunc', func)
            funcmod = inspect.getmodule(func)
            extra = ''
            if funcmod.__package__ == funcmod.__name__:
                extra = '/__init__'
            actualpath = '%s%s.py' % (funcmod.__name__.replace('.', '/'), extra)

            src = inspect.getsource(func)
            name = "%s.%s" % (actualpath, func.__name__)
            lineno = inspect.getsourcelines(func)[1]
            doc = docobj.__doc__
            origdoc = getattr(docobj, '_origdoc', '')
            if rstrip:
                doc = doc.rstrip()
                origdoc = origdoc.rstrip()
            if origdoc:
                lineno += offset(src, origdoc, name, 1)
            else:
                lineno += offset(src, doc, name, 1)
            print(poentry(actualpath, 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)