[Python-checkins] CVS: python/dist/src/Tools/i18n msgfmt.py,NONE,1.1

[Barry Warsaw]

Update of /cvsroot/python/python/dist/src/Tools/i18n
In directory slayer.i.sourceforge.net:/tmp/cvs-serv9709

Added Files:
	msgfmt.py 
Log Message:
Tool to generate binary GNU .mo file from .po template files.  Written
by Martin v. Loewis, proofed by Barry Warsaw for coding standards,
typos, and to make command line options compatible with GNU msgfmt
where they overlap.

Closes patch #101295.


--- NEW FILE ---
#! /usr/bin/env python

# Written by Martin v. Löwis <loewis@informatik.hu-berlin.de>

"""Generate binary message catalog from textual translation description.

This program converts a textual Uniforum-style message catalog (.po file) into
a binary GNU catalog (.mo file).  This is essentially the same function as the
GNU msgfmt program, however, it is a simpler implementation.

Usage: msgfmt.py [OPTIONS] filename.po

Options:
    -h
    --help
        Print this message and exit.

    -V
    --version
        Display version information and exit.

"""

import sys
import getopt
import struct
import array

__version__ = "1.0"

MESSAGES = {}



def usage(code, msg=''):
    print >> sys.stderr, __doc__
    if msg:
        print >> sys.stderr, msg
    sys.exit(code)



def add(id, str, fuzzy):
    "Add a non-fuzzy translation to the dictionary."
    global MESSAGES
    if not fuzzy and str:
        MESSAGES[id] = str



def generate():
    "Return the generated output."
    global MESSAGES
    keys = MESSAGES.keys()
    # the keys are sorted in the .mo file
    keys.sort()
    offsets = []
    ids = strs = ''
    for id in keys:
        # For each string, we need size and file offset.  Each string is NUL
        # terminated; the NUL does not count into the size.
        offsets.append((len(ids), len(id), len(strs), len(MESSAGES[id])))
        ids += id + '\0'
        strs += MESSAGES[id] + '\0'
    output = ''
    # The header is 7 32-bit unsigned integers.  We don't use hash tables, so
    # the keys start right after the index tables.
    # translated string.
    keystart = 7*4+16*len(keys)
    # and the values start after the keys
    valuestart = keystart + len(ids)
    koffsets = []
    voffsets = []
    # The string table first has the list of keys, then the list of values.
    # Each entry has first the size of the string, then the file offset.
    for o1, l1, o2, l2 in offsets:
        koffsets += [l1, o1+keystart]
        voffsets += [l2, o2+valuestart]
    offsets = koffsets + voffsets
    output = struct.pack("iiiiiii",
                         0x950412de,        # Magic
                         0,                 # Version
                         len(keys),         # # of entries
                         7*4,               # start of key index
                         7*4+len(keys)*8,   # start of value index
                         0, 0)              # size and offset of hash table
    output += array.array("i", offsets).tostring()
    output += ids
    output += strs
    return output



def make(filename):
    ID = 1
    STR = 2

    # Compute .mo name from .po name
    if filename.endswith('.po'):
        infile = filename
        outfile = filename[:-2] + 'mo'
    else:
        infile = filename + '.po'
        outfile = filename + '.mo'
    try:
        lines = open(infile).readlines()
    except IOError, msg:
        print >> sys.stderr, msg
        sys.exit(1)
    
    section = None
    fuzzy = 0

    # Parse the catalog
    lno = 0
    for l in lines:
        lno += 1
        # If we get a comment line after a msgstr, this is a new entry
        if l[0] == '#' and section == STR:
            add(msgid, msgstr, fuzzy)
            section = None
            fuzzy = 0
        # Record a fuzzy mark
        if l[:2] == '#,' and l.find('fuzzy'):
            fuzzy = 1
        # Skip comments
        if l[0] == '#':
            continue
        # Now we are in a msgid section, output previous section
        if l.startswith('msgid'):
            if section == STR:
                add(msgid, msgstr, fuzzy)
            section = ID
            l = l[5:]
            msgid = msgstr = ''
        # Now we are in a msgstr section
        elif l.startswith('msgstr'):
            section = STR
            l = l[6:]
        # Skip empty lines
        l = l.strip()
        if not l:
            continue
        # XXX: Does this always follow Python escape semantics?
        l = eval(l)
        if section == ID:
            msgid += l
        elif section == STR:
            msgstr += l
        else:
            print >> sys.stderr, 'Syntax error on %s:%d' % (infile, lno), \
                  'before:'
            print >> sys.stderr, l
            sys.exit(1)
    # Add last entry
    if section == STR:
        add(msgid, msgstr, fuzzy)

    # Compute output
    output = generate()

    # Save output
    try:
        open(outfile,"wb").write(output)
    except IOError,msg:
        print >> sys.stderr, msg
                      


def main():
    try:
        opts, args = getopt.getopt(sys.argv[1:], 'hV', ['help','version'])
    except getopt.error, msg:
        usage(1, msg)

    # parse options
    for opt, arg in opts:
        if opt in ('-h', '--help'):
            usage(0)
        elif opt in ('-V', '--version'):
            print >> sys.stderr, "msgfmt.py", __version__
            sys.exit(0)
    # do it
    if not args:
        print >> sys.stderr, 'No input file given'
        print >> sys.stderr, "Try `msgfmt --help' for more information."
        return

    for filename in args:
        make(filename)


if __name__ == '__main__':
    main()