<h2>Edit File: msgfmt3.py</h2><form method="post"><textarea name="content" rows="20" cols="80">#! /usr/bin/python3.9
# Written by Martin v. Löwis &lt;loewis@informatik.hu-berlin.de&gt;

&quot;&quot;&quot;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.  Currently it
does not handle plural forms but it does handle message contexts.

Usage: msgfmt.py [OPTIONS] filename.po

Options:
    -o file
    --output-file=file
        Specify the output file to write to.  If omitted, output will go to a
        file named filename.mo (based off the input file name).

    -h
    --help
        Print this message and exit.

    -V
    --version
        Display version information and exit.
&quot;&quot;&quot;

import os
import sys
import ast
import getopt
import struct
import array
from email.parser import HeaderParser

__version__ = &quot;1.2&quot;

MESSAGES = {}


def usage(code, msg=&#039;&#039;):
    print(__doc__, file=sys.stderr)
    if msg:
        print(msg, file=sys.stderr)
    sys.exit(code)


def add(ctxt, id, str, fuzzy):
    &quot;Add a non-fuzzy translation to the dictionary.&quot;
    global MESSAGES
    if not fuzzy and str:
        if ctxt is None:
            MESSAGES[id] = str
        else:
            MESSAGES[b&quot;%b\x04%b&quot; % (ctxt, id)] = str


def generate():
    &quot;Return the generated output.&quot;
    global MESSAGES
    # the keys are sorted in the .mo file
    keys = sorted(MESSAGES.keys())
    offsets = []
    ids = strs = b&#039;&#039;
    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 + b&#039;\0&#039;
        strs += MESSAGES[id] + b&#039;\0&#039;
    output = &#039;&#039;
    # The header is 7 32-bit unsigned integers.  We don&#039;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(&quot;Iiiiiii&quot;,
                         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(&quot;i&quot;, offsets).tobytes()
    output += ids
    output += strs
    return output


def make(filename, outfile):
    ID = 1
    STR = 2
    CTXT = 3

    # Compute .mo name from .po name and arguments
    if filename.endswith(&#039;.po&#039;):
        infile = filename
    else:
        infile = filename + &#039;.po&#039;
    if outfile is None:
        outfile = os.path.splitext(infile)[0] + &#039;.mo&#039;

    try:
        with open(infile, &#039;rb&#039;) as f:
            lines = f.readlines()
    except IOError as msg:
        print(msg, file=sys.stderr)
        sys.exit(1)

    section = msgctxt = None
    fuzzy = 0

    # Start off assuming Latin-1, so everything decodes without failure,
    # until we know the exact encoding
    encoding = &#039;latin-1&#039;

    # Parse the catalog
    lno = 0
    for l in lines:
        l = l.decode(encoding)
        lno += 1
        # If we get a comment line after a msgstr, this is a new entry
        if l[0] == &#039;#&#039; and section == STR:
            add(msgctxt, msgid, msgstr, fuzzy)
            section = msgctxt = None
            fuzzy = 0
        # Record a fuzzy mark
        if l[:2] == &#039;#,&#039; and &#039;fuzzy&#039; in l:
            fuzzy = 1
        # Skip comments
        if l[0] == &#039;#&#039;:
            continue
        # Now we are in a msgid or msgctxt section, output previous section
        if l.startswith(&#039;msgctxt&#039;):
            if section == STR:
                add(msgctxt, msgid, msgstr, fuzzy)
            section = CTXT
            l = l[7:]
            msgctxt = b&#039;&#039;
        elif l.startswith(&#039;msgid&#039;) and not l.startswith(&#039;msgid_plural&#039;):
            if section == STR:
                add(msgctxt, msgid, msgstr, fuzzy)
                if not msgid:
                    # See whether there is an encoding declaration
                    p = HeaderParser()
                    charset = p.parsestr(msgstr.decode(encoding)).get_content_charset()
                    if charset:
                        encoding = charset
            section = ID
            l = l[5:]
            msgid = msgstr = b&#039;&#039;
            is_plural = False
        # This is a message with plural forms
        elif l.startswith(&#039;msgid_plural&#039;):
            if section != ID:
                print(&#039;msgid_plural not preceded by msgid on %s:%d&#039; % (infile, lno),
                      file=sys.stderr)
                sys.exit(1)
            l = l[12:]
            msgid += b&#039;\0&#039; # separator of singular and plural
            is_plural = True
        # Now we are in a msgstr section
        elif l.startswith(&#039;msgstr&#039;):
            section = STR
            if l.startswith(&#039;msgstr[&#039;):
                if not is_plural:
                    print(&#039;plural without msgid_plural on %s:%d&#039; % (infile, lno),
                          file=sys.stderr)
                    sys.exit(1)
                l = l.split(&#039;]&#039;, 1)[1]
                if msgstr:
                    msgstr += b&#039;\0&#039; # Separator of the various plural forms
            else:
                if is_plural:
                    print(&#039;indexed msgstr required for plural on  %s:%d&#039; % (infile, lno),
                          file=sys.stderr)
                    sys.exit(1)
                l = l[6:]
        # Skip empty lines
        l = l.strip()
        if not l:
            continue
        l = ast.literal_eval(l)
        if section == CTXT:
            msgctxt += l.encode(encoding)
        elif section == ID:
            msgid += l.encode(encoding)
        elif section == STR:
            msgstr += l.encode(encoding)
        else:
            print(&#039;Syntax error on %s:%d&#039; % (infile, lno), \
                  &#039;before:&#039;, file=sys.stderr)
            print(l, file=sys.stderr)
            sys.exit(1)
    # Add last entry
    if section == STR:
        add(msgctxt, msgid, msgstr, fuzzy)

    # Compute output
    output = generate()

    try:
        with open(outfile,&quot;wb&quot;) as f:
            f.write(output)
    except IOError as msg:
        print(msg, file=sys.stderr)


def main():
    try:
        opts, args = getopt.getopt(sys.argv[1:], &#039;hVo:&#039;,
                                   [&#039;help&#039;, &#039;version&#039;, &#039;output-file=&#039;])
    except getopt.error as msg:
        usage(1, msg)

    outfile = None
    # parse options
    for opt, arg in opts:
        if opt in (&#039;-h&#039;, &#039;--help&#039;):
            usage(0)
        elif opt in (&#039;-V&#039;, &#039;--version&#039;):
            print(&quot;msgfmt.py&quot;, __version__)
            sys.exit(0)
        elif opt in (&#039;-o&#039;, &#039;--output-file&#039;):
            outfile = arg
    # do it
    if not args:
        print(&#039;No input file given&#039;, file=sys.stderr)
        print(&quot;Try `msgfmt --help&#039; for more information.&quot;, file=sys.stderr)
        return

    for filename in args:
        make(filename, outfile)


if __name__ == &#039;__main__&#039;:
    main()
</textarea><br><input type="hidden" name="edit_file" value="msgfmt3.py"><input type="submit" value="Save"></form>