This is an archive of the discontinued Mercurial Phabricator instance.

Differential D5001

polib: update to latest release 1.0.7 (upstream rev d75ce6dbbc2a)
ClosedPublic

Authored by durin42 on Oct 12 2018, 12:53 PM.

Details

Reviewers
None
Group Reviewers
hg-reviewers
Commits
rHG19fc5a986669: polib: update to latest release 1.0.7 (upstream rev d75ce6dbbc2a)
Summary
  1. no-check-commit third-party code that doesn't match our style

Diff Detail

Repository
rHG Mercurial
Lint
Automatic diff as part of commit; lint not applicable.
Unit
Automatic diff as part of commit; unit tests not applicable.

Revision Contents
Changeset List

PathPackages
Mi18n/check-translation.py (8 lines)
Mi18n/polib.py (818 lines)

Diff 11932

i18n/check-translation.py

if options.doctest: if options.doctest:
import os import os
if 'TERM' in os.environ: if 'TERM' in os.environ:
del os.environ['TERM'] del os.environ['TERM']
import doctest import doctest
failures, tests = doctest.testmod() failures, tests = doctest.testmod()
sys.exit(failures and 1 or 0) sys.exit(failures and 1 or 0)
# replace polib._POFileParser to show linenum of problematic msgstr
class ExtPOFileParser(polib._POFileParser):
def process(self, symbol, linenum):
super(ExtPOFileParser, self).process(symbol, linenum)
if symbol == 'MS': # msgstr
self.current_entry.linenum = linenum
polib._POFileParser = ExtPOFileParser
detected = [] detected = []
warning = options.warning warning = options.warning
for f in args: for f in args:
detected.extend((f, pe, errors) detected.extend((f, pe, errors)
for pe, errors in check(polib.pofile(f), for pe, errors in check(polib.pofile(f),
warning=warning)) warning=warning))
if detected: if detected:
for f, pe, errors in detected: for f, pe, errors in detected:
for level, checker, error in errors: for level, checker, error in errors:
sys.stderr.write('%s:%d:%s(%s): %s\n' sys.stderr.write('%s:%d:%s(%s): %s\n'
% (f, pe.linenum, level, checker, error)) % (f, pe.linenum, level, checker, error))
sys.exit(1) sys.exit(1)

i18n/polib.py

# -*- coding: utf-8 -*-
# no-check-code # no-check-code
# -* coding: utf-8 -*-
# #
# License: MIT (see LICENSE file provided) # License: MIT (see LICENSE file provided)
# vim: set expandtab tabstop=4 shiftwidth=4 softtabstop=4: # vim: set expandtab tabstop=4 shiftwidth=4 softtabstop=4:
""" """
**polib** allows you to manipulate, create, modify gettext files (pot, po and **polib** allows you to manipulate, create, modify gettext files (pot, po and
mo files). You can load existing files, iterate through it's entries, add, mo files). You can load existing files, iterate through it's entries, add,
modify entries, comments or metadata, etc. or create new po files from scratch. modify entries, comments or metadata, etc. or create new po files from scratch.
**polib** provides a simple and pythonic API via the :func:`~polib.pofile` and **polib** provides a simple and pythonic API via the :func:`~polib.pofile` and
:func:`~polib.mofile` convenience functions. :func:`~polib.mofile` convenience functions.
""" """
from __future__ import absolute_import from __future__ import absolute_import
__author__ = 'David Jean Louis <izimobil@gmail.com>' __author__ = 'David Jean Louis <izimobil@gmail.com>'
__version__ = '0.6.4' __version__ = '1.0.7'
__all__ = ['pofile', 'POFile', 'POEntry', 'mofile', 'MOFile', 'MOEntry', __all__ = ['pofile', 'POFile', 'POEntry', 'mofile', 'MOFile', 'MOEntry',
'detect_encoding', 'escape', 'unescape', 'detect_encoding',] 'default_encoding', 'escape', 'unescape', 'detect_encoding', ]
import array import array
import codecs import codecs
import os import os
import re import re
import struct import struct
import sys import sys
import textwrap import textwrap
import types
try:
import io
except ImportError:
# replacement of io.open() for python < 2.6
# we use codecs instead
class io(object):
@staticmethod
def open(fpath, mode='r', encoding=None):
return codecs.open(fpath, mode, encoding)
# the default encoding to use when encoding cannot be detected # the default encoding to use when encoding cannot be detected
default_encoding = 'utf-8' default_encoding = 'utf-8'
# python 2/3 compatibility helpers {{{
if sys.version_info[:2] < (3, 0):
PY3 = False
text_type = unicode
def b(s):
return s
def u(s):
return unicode(s, "unicode_escape")
else:
PY3 = True
text_type = str
def b(s):
return s.encode("latin-1")
def u(s):
return s
# }}}
# _pofile_or_mofile {{{ # _pofile_or_mofile {{{
def _pofile_or_mofile(f, type, **kwargs): def _pofile_or_mofile(f, type, **kwargs):
""" """
Internal function used by :func:`polib.pofile` and :func:`polib.mofile` to Internal function used by :func:`polib.pofile` and :func:`polib.mofile` to
honor the DRY concept. honor the DRY concept.
""" """
# get the file encoding # get the file encoding
enc = kwargs.get('encoding') enc = kwargs.get('encoding')
if enc is None: if enc is None:
enc = detect_encoding(f, type == 'mofile') enc = detect_encoding(f, type == 'mofile')
# parse the file # parse the file
kls = type == 'pofile' and _POFileParser or _MOFileParser kls = type == 'pofile' and _POFileParser or _MOFileParser
parser = kls( parser = kls(
f, f,
encoding=enc, encoding=enc,
check_for_duplicates=kwargs.get('check_for_duplicates', False) check_for_duplicates=kwargs.get('check_for_duplicates', False),
klass=kwargs.get('klass')
) )
instance = parser.parse() instance = parser.parse()
instance.wrapwidth = kwargs.get('wrapwidth', 78) instance.wrapwidth = kwargs.get('wrapwidth', 78)
return instance return instance
# }}}
# _is_file {{{
def _is_file(filename_or_contents):
"""
Safely returns the value of os.path.exists(filename_or_contents).
Arguments:
``filename_or_contents``
either a filename, or a string holding the contents of some file.
In the latter case, this function will always return False.
"""
try:
return os.path.exists(filename_or_contents)
except (ValueError, UnicodeEncodeError):
return False
# }}} # }}}
# function pofile() {{{ # function pofile() {{{
def pofile(pofile, **kwargs): def pofile(pofile, **kwargs):
""" """
Convenience function that parses the po or pot file ``pofile`` and returns Convenience function that parses the po or pot file ``pofile`` and returns
a :class:`~polib.POFile` instance. a :class:`~polib.POFile` instance.
Arguments: Arguments:
``pofile`` ``pofile``
string, full or relative path to the po/pot file or its content (data). string, full or relative path to the po/pot file or its content (data).
``wrapwidth`` ``wrapwidth``
integer, the wrap width, only useful when the ``-w`` option was passed integer, the wrap width, only useful when the ``-w`` option was passed
to xgettext (optional, default: ``78``). to xgettext (optional, default: ``78``).
``encoding`` ``encoding``
string, the encoding to use (e.g. "utf-8") (default: ``None``, the string, the encoding to use (e.g. "utf-8") (default: ``None``, the
encoding will be auto-detected). encoding will be auto-detected).
``check_for_duplicates`` ``check_for_duplicates``
whether to check for duplicate entries when adding entries to the whether to check for duplicate entries when adding entries to the
file (optional, default: ``False``). file (optional, default: ``False``).
``klass``
class which is used to instantiate the return value (optional,
default: ``None``, the return value with be a :class:`~polib.POFile`
instance).
""" """
return _pofile_or_mofile(pofile, 'pofile', **kwargs) return _pofile_or_mofile(pofile, 'pofile', **kwargs)
# }}} # }}}
# function mofile() {{{ # function mofile() {{{
def mofile(mofile, **kwargs): def mofile(mofile, **kwargs):
""" """
Convenience function that parses the mo file ``mofile`` and returns a Convenience function that parses the mo file ``mofile`` and returns a
:class:`~polib.MOFile` instance. :class:`~polib.MOFile` instance.
Arguments: Arguments:
``mofile`` ``mofile``
string, full or relative path to the mo file or its content (data). string, full or relative path to the mo file or its content (data).
``wrapwidth`` ``wrapwidth``
integer, the wrap width, only useful when the ``-w`` option was passed integer, the wrap width, only useful when the ``-w`` option was passed
to xgettext to generate the po file that was used to format the mo file to xgettext to generate the po file that was used to format the mo file
(optional, default: ``78``). (optional, default: ``78``).
``encoding`` ``encoding``
string, the encoding to use (e.g. "utf-8") (default: ``None``, the string, the encoding to use (e.g. "utf-8") (default: ``None``, the
encoding will be auto-detected). encoding will be auto-detected).
``check_for_duplicates`` ``check_for_duplicates``
whether to check for duplicate entries when adding entries to the whether to check for duplicate entries when adding entries to the
file (optional, default: ``False``). file (optional, default: ``False``).
``klass``
class which is used to instantiate the return value (optional,
default: ``None``, the return value with be a :class:`~polib.POFile`
instance).
""" """
return _pofile_or_mofile(mofile, 'mofile', **kwargs) return _pofile_or_mofile(mofile, 'mofile', **kwargs)
# }}} # }}}
# function detect_encoding() {{{ # function detect_encoding() {{{
def detect_encoding(file, binary_mode=False): def detect_encoding(file, binary_mode=False):
""" """
Try to detect the encoding used by the ``file``. The ``file`` argument can Try to detect the encoding used by the ``file``. The ``file`` argument can
be a PO or MO file path or a string containing the contents of the file. be a PO or MO file path or a string containing the contents of the file.
If the encoding cannot be detected, the function will return the value of If the encoding cannot be detected, the function will return the value of
``default_encoding``. ``default_encoding``.
Arguments: Arguments:
``file`` ``file``
string, full or relative path to the po/mo file or its content. string, full or relative path to the po/mo file or its content.
``binary_mode`` ``binary_mode``
boolean, set this to True if ``file`` is a mo file. boolean, set this to True if ``file`` is a mo file.
""" """
rx = re.compile(r'"?Content-Type:.+? charset=([\w_\-:\.]+)') PATTERN = r'"?Content-Type:.+? charset=([\w_\-:\.]+)'
rxt = re.compile(u(PATTERN))
rxb = re.compile(b(PATTERN))
def charset_exists(charset): def charset_exists(charset):
"""Check whether ``charset`` is valid or not.""" """Check whether ``charset`` is valid or not."""
try: try:
codecs.lookup(charset) codecs.lookup(charset)
except LookupError: except LookupError:
return False return False
return True return True
if not os.path.exists(file): if not _is_file(file):
match = rx.search(file) match = rxt.search(file)
if match: if match:
enc = match.group(1).strip() enc = match.group(1).strip()
if charset_exists(enc): if charset_exists(enc):
return enc return enc
else: else:
if binary_mode: # For PY3, always treat as binary
if binary_mode or PY3:
mode = 'rb' mode = 'rb'
rx = rxb
else: else:
mode = 'r' mode = 'r'
rx = rxt
f = open(file, mode) f = open(file, mode)
for l in f.readlines(): for l in f.readlines():
match = rx.search(l) match = rx.search(l)
if match: if match:
f.close() f.close()
enc = match.group(1).strip() enc = match.group(1).strip()
if not isinstance(enc, text_type):
enc = enc.decode('utf-8')
if charset_exists(enc): if charset_exists(enc):
return enc return enc
f.close() f.close()
return default_encoding return default_encoding
# }}} # }}}
# function escape() {{{ # function escape() {{{
def escape(st): def escape(st):
""" """
Escapes the characters ``\\\\``, ``\\t``, ``\\n``, ``\\r`` and ``"`` in Escapes the characters ``\\\\``, ``\\t``, ``\\n``, ``\\r`` and ``"`` in
the given string ``st`` and returns it. the given string ``st`` and returns it.
""" """
return st.replace('\\', r'\\')\ return st.replace('\\', r'\\')\
.replace('\t', r'\t')\ .replace('\t', r'\t')\
.replace('\r', r'\r')\ .replace('\r', r'\r')\
.replace('\n', r'\n')\ .replace('\n', r'\n')\
.replace('\"', r'\"') .replace('\"', r'\"')
# }}} # }}}
# function unescape() {{{ # function unescape() {{{
def unescape(st): def unescape(st):
""" """
Unescapes the characters ``\\\\``, ``\\t``, ``\\n``, ``\\r`` and ``"`` in Unescapes the characters ``\\\\``, ``\\t``, ``\\n``, ``\\r`` and ``"`` in
the given string ``st`` and returns it. the given string ``st`` and returns it.
""" """
def unescape_repl(m): def unescape_repl(m):
m = m.group(1) m = m.group(1)
if m == 'n': if m == 'n':
return '\n' return '\n'
if m == 't': if m == 't':
return '\t' return '\t'
if m == 'r': if m == 'r':
return '\r' return '\r'
if m == '\\': if m == '\\':
return '\\' return '\\'
return m # handles escaped double quote return m # handles escaped double quote
return re.sub(r'\\(\\|n|t|r|")', unescape_repl, st) return re.sub(r'\\(\\|n|t|r|")', unescape_repl, st)
# }}} # }}}
# class _BaseFile {{{ # class _BaseFile {{{
class _BaseFile(list): class _BaseFile(list):
""" """
Common base class for the :class:`~polib.POFile` and :class:`~polib.MOFile` Common base class for the :class:`~polib.POFile` and :class:`~polib.MOFile`
classes. This class should **not** be instanciated directly. classes. This class should **not** be instanciated directly.
""" """
def __init__(self, *args, **kwargs): def __init__(self, *args, **kwargs):
""" """
``check_for_duplicates`` ``check_for_duplicates``
whether to check for duplicate entries when adding entries to the whether to check for duplicate entries when adding entries to the
file, (optional, default: ``False``). file, (optional, default: ``False``).
""" """
list.__init__(self) list.__init__(self)
# the opened file handle # the opened file handle
pofile = kwargs.get('pofile', None) pofile = kwargs.get('pofile', None)
if pofile and os.path.exists(pofile): if pofile and _is_file(pofile):
self.fpath = pofile self.fpath = pofile
else: else:
self.fpath = kwargs.get('fpath') self.fpath = kwargs.get('fpath')
# the width at which lines should be wrapped # the width at which lines should be wrapped
self.wrapwidth = kwargs.get('wrapwidth', 78) self.wrapwidth = kwargs.get('wrapwidth', 78)
# the file encoding # the file encoding
self.encoding = kwargs.get('encoding', default_encoding) self.encoding = kwargs.get('encoding', default_encoding)
# whether to check for duplicate entries or not # whether to check for duplicate entries or not
""" """
ret = [] ret = []
entries = [self.metadata_as_entry()] + \ entries = [self.metadata_as_entry()] + \
[e for e in self if not e.obsolete] [e for e in self if not e.obsolete]
for entry in entries: for entry in entries:
ret.append(entry.__unicode__(self.wrapwidth)) ret.append(entry.__unicode__(self.wrapwidth))
for entry in self.obsolete_entries(): for entry in self.obsolete_entries():
ret.append(entry.__unicode__(self.wrapwidth)) ret.append(entry.__unicode__(self.wrapwidth))
ret = '\n'.join(ret) ret = u('\n').join(ret)
if type(ret) != types.UnicodeType: assert isinstance(ret, text_type)
return unicode(ret, self.encoding) #if type(ret) != text_type:
# return unicode(ret, self.encoding)
return ret return ret
if PY3:
def __str__(self):
return self.__unicode__()
else:
def __str__(self): def __str__(self):
""" """
Returns the string representation of the file. Returns the string representation of the file.
""" """
return unicode(self).encode(self.encoding) return unicode(self).encode(self.encoding)
def __contains__(self, entry): def __contains__(self, entry):
""" """
Overriden ``list`` method to implement the membership test (in and Overridden ``list`` method to implement the membership test (in and
not in). not in).
The method considers that an entry is in the file if it finds an entry The method considers that an entry is in the file if it finds an entry
that has the same msgid (the test is **case sensitive**). that has the same msgid (the test is **case sensitive**) and the same
msgctxt (or none for both entries).
Argument: Argument:
``entry`` ``entry``
an instance of :class:`~polib._BaseEntry`. an instance of :class:`~polib._BaseEntry`.
""" """
return self.find(entry.msgid, by='msgid') is not None return self.find(entry.msgid, by='msgid', msgctxt=entry.msgctxt) \
is not None
def __eq__(self, other): def __eq__(self, other):
return unicode(self) == unicode(other) return str(self) == str(other)
def append(self, entry): def append(self, entry):
""" """
Overriden method to check for duplicates entries, if a user tries to Overridden method to check for duplicates entries, if a user tries to
add an entry that is already in the file, the method will raise a add an entry that is already in the file, the method will raise a
``ValueError`` exception. ``ValueError`` exception.
Argument: Argument:
``entry`` ``entry``
an instance of :class:`~polib._BaseEntry`. an instance of :class:`~polib._BaseEntry`.
""" """
if self.check_for_duplicates and entry in self: if self.check_for_duplicates and entry in self:
raise ValueError('Entry "%s" already exists' % entry.msgid) raise ValueError('Entry "%s" already exists' % entry.msgid)
super(_BaseFile, self).append(entry) super(_BaseFile, self).append(entry)
def insert(self, index, entry): def insert(self, index, entry):
""" """
Overriden method to check for duplicates entries, if a user tries to Overridden method to check for duplicates entries, if a user tries to
add an entry that is already in the file, the method will raise a add an entry that is already in the file, the method will raise a
``ValueError`` exception. ``ValueError`` exception.
Arguments: Arguments:
``index`` ``index``
index at which the entry should be inserted. index at which the entry should be inserted.
for name, value in mdata: for name, value in mdata:
# Strip whitespace off each line in a multi-line entry # Strip whitespace off each line in a multi-line entry
strs.append('%s: %s' % (name, value)) strs.append('%s: %s' % (name, value))
e.msgstr = '\n'.join(strs) + '\n' e.msgstr = '\n'.join(strs) + '\n'
if self.metadata_is_fuzzy: if self.metadata_is_fuzzy:
e.flags.append('fuzzy') e.flags.append('fuzzy')
return e return e
def save(self, fpath=None, repr_method='__str__'): def save(self, fpath=None, repr_method='__unicode__'):
""" """
Saves the po file to ``fpath``. Saves the po file to ``fpath``.
If it is an existing file and no ``fpath`` is provided, then the If it is an existing file and no ``fpath`` is provided, then the
existing file is rewritten with the modified data. existing file is rewritten with the modified data.
Keyword arguments: Keyword arguments:
``fpath`` ``fpath``
string, full or relative path to the file. string, full or relative path to the file.
``repr_method`` ``repr_method``
string, the method to use for output. string, the method to use for output.
""" """
if self.fpath is None and fpath is None: if self.fpath is None and fpath is None:
raise IOError('You must provide a file path to save() method') raise IOError('You must provide a file path to save() method')
contents = getattr(self, repr_method)() contents = getattr(self, repr_method)()
if fpath is None: if fpath is None:
fpath = self.fpath fpath = self.fpath
if repr_method == 'to_binary': if repr_method == 'to_binary':
fhandle = open(fpath, 'wb') fhandle = open(fpath, 'wb')
else: else:
fhandle = codecs.open(fpath, 'w', self.encoding) fhandle = io.open(fpath, 'w', encoding=self.encoding)
if type(contents) != types.UnicodeType: if not isinstance(contents, text_type):
contents = contents.decode(self.encoding) contents = contents.decode(self.encoding)
fhandle.write(contents) fhandle.write(contents)
fhandle.close() fhandle.close()
# set the file path if not set # set the file path if not set
if self.fpath is None and fpath: if self.fpath is None and fpath:
self.fpath = fpath self.fpath = fpath
def find(self, st, by='msgid', include_obsolete_entries=False, def find(self, st, by='msgid', include_obsolete_entries=False,
``by`` ``by``
string, the property to use for comparison (default: ``msgid``). string, the property to use for comparison (default: ``msgid``).
``include_obsolete_entries`` ``include_obsolete_entries``
boolean, whether to also search in entries that are obsolete. boolean, whether to also search in entries that are obsolete.
``msgctxt`` ``msgctxt``
string, allows to specify a specific message context for the string, allows specifying a specific message context for the
search. search.
""" """
if include_obsolete_entries: if include_obsolete_entries:
entries = self[:] entries = self[:]
else: else:
entries = [e for e in self if not e.obsolete] entries = [e for e in self if not e.obsolete]
for e in entries: for e in entries:
if getattr(e, by) == st: if getattr(e, by) == st:
if msgctxt and e.msgctxt != msgctxt: if msgctxt is not False and e.msgctxt != msgctxt:
continue continue
return e return e
return None return None
def ordered_metadata(self): def ordered_metadata(self):
""" """
Convenience method that returns an ordered version of the metadata Convenience method that returns an ordered version of the metadata
dictionary. The return value is list of tuples (metadata name, dictionary. The return value is list of tuples (metadata name,
metadata_value). metadata_value).
""" """
# copy the dict first # copy the dict first
metadata = self.metadata.copy() metadata = self.metadata.copy()
data_order = [ data_order = [
'Project-Id-Version', 'Project-Id-Version',
'Report-Msgid-Bugs-To', 'Report-Msgid-Bugs-To',
'POT-Creation-Date', 'POT-Creation-Date',
'PO-Revision-Date', 'PO-Revision-Date',
'Last-Translator', 'Last-Translator',
'Language-Team', 'Language-Team',
'MIME-Version', 'MIME-Version',
'Content-Type', 'Content-Type',
'Content-Transfer-Encoding' 'Content-Transfer-Encoding',
'Language',
'Plural-Forms'
] ]
ordered_data = [] ordered_data = []
for data in data_order: for data in data_order:
try: try:
value = metadata.pop(data) value = metadata.pop(data)
ordered_data.append((data, value)) ordered_data.append((data, value))
except KeyError: except KeyError:
pass pass
# the rest of the metadata will be alphabetically ordered since there # the rest of the metadata will be alphabetically ordered since there
# are no specs for this AFAIK # are no specs for this AFAIK
keys = metadata.keys() for data in sorted(metadata.keys()):
keys.sort()
for data in keys:
value = metadata[data] value = metadata[data]
ordered_data.append((data, value)) ordered_data.append((data, value))
return ordered_data return ordered_data
def to_binary(self): def to_binary(self):
""" """
Return the binary representation of the file. Return the binary representation of the file.
""" """
offsets = [] offsets = []
entries = self.translated_entries() entries = self.translated_entries()
# the keys are sorted in the .mo file # the keys are sorted in the .mo file
def cmp(_self, other): def cmp(_self, other):
# msgfmt compares entries with msgctxt if it exists # msgfmt compares entries with msgctxt if it exists
if _self.msgctxt: self_msgid = _self.msgctxt and _self.msgctxt or _self.msgid
self_msgid = _self.msgctxt other_msgid = other.msgctxt and other.msgctxt or other.msgid
else:
self_msgid = _self.msgid
if other.msgctxt:
other_msgid = other.msgctxt
else:
other_msgid = other.msgid
if self_msgid > other_msgid: if self_msgid > other_msgid:
return 1 return 1
elif self_msgid < other_msgid: elif self_msgid < other_msgid:
return -1 return -1
else: else:
return 0 return 0
# add metadata entry # add metadata entry
entries.sort(cmp) entries.sort(key=lambda o: o.msgctxt or o.msgid)
mentry = self.metadata_as_entry() mentry = self.metadata_as_entry()
#mentry.msgstr = mentry.msgstr.replace('\\n', '').lstrip() #mentry.msgstr = mentry.msgstr.replace('\\n', '').lstrip()
entries = [mentry] + entries entries = [mentry] + entries
entries_len = len(entries) entries_len = len(entries)
ids, strs = '', '' ids, strs = b(''), b('')
for e in entries: for e in entries:
# For each string, we need size and file offset. Each string is # For each string, we need size and file offset. Each string is
# NUL terminated; the NUL does not count into the size. # NUL terminated; the NUL does not count into the size.
msgid = '' msgid = b('')
if e.msgctxt: if e.msgctxt:
# Contexts are stored by storing the concatenation of the # Contexts are stored by storing the concatenation of the
# context, a <EOT> byte, and the original string # context, a <EOT> byte, and the original string
msgid = self._encode(e.msgctxt + '\4') msgid = self._encode(e.msgctxt + '\4')
if e.msgid_plural: if e.msgid_plural:
indexes = e.msgstr_plural.keys()
indexes.sort()
msgstr = [] msgstr = []
for index in indexes: for index in sorted(e.msgstr_plural.keys()):
msgstr.append(e.msgstr_plural[index]) msgstr.append(e.msgstr_plural[index])
msgid += self._encode(e.msgid + '\0' + e.msgid_plural) msgid += self._encode(e.msgid + '\0' + e.msgid_plural)
msgstr = self._encode('\0'.join(msgstr)) msgstr = self._encode('\0'.join(msgstr))
else: else:
msgid += self._encode(e.msgid) msgid += self._encode(e.msgid)
msgstr = self._encode(e.msgstr) msgstr = self._encode(e.msgstr)
offsets.append((len(ids), len(msgid), len(strs), len(msgstr))) offsets.append((len(ids), len(msgid), len(strs), len(msgstr)))
ids += msgid + '\0' ids += msgid + b('\0')
strs += msgstr + '\0' strs += msgstr + b('\0')
# The header is 7 32-bit unsigned integers. # The header is 7 32-bit unsigned integers.
keystart = 7*4+16*entries_len keystart = 7 * 4 + 16 * entries_len
# and the values start after the keys # and the values start after the keys
valuestart = keystart + len(ids) valuestart = keystart + len(ids)
koffsets = [] koffsets = []
voffsets = [] voffsets = []
# The string table first has the list of keys, then the list of values. # 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. # Each entry has first the size of the string, then the file offset.
for o1, l1, o2, l2 in offsets: for o1, l1, o2, l2 in offsets:
koffsets += [l1, o1+keystart] koffsets += [l1, o1 + keystart]
voffsets += [l2, o2+valuestart] voffsets += [l2, o2 + valuestart]
offsets = koffsets + voffsets offsets = koffsets + voffsets
# check endianness for magic number
if struct.pack('@h', 1) == struct.pack('<h', 1):
magic_number = MOFile.LITTLE_ENDIAN
else:
magic_number = MOFile.BIG_ENDIAN
output = struct.pack( output = struct.pack(
"Iiiiiii", "Iiiiiii",
magic_number, # Magic number # Magic number
0, # Version MOFile.MAGIC,
entries_len, # # of entries # Version
7*4, # start of key index 0,
7*4+entries_len*8, # start of value index # number of entries
0, keystart # size and offset of hash table entries_len,
# Important: we don't use hash tables # start of key index
7 * 4,
# start of value index
7 * 4 + entries_len * 8,
# size and offset of hash table, we don't use hash tables
0, keystart
) )
if PY3 and sys.version_info.minor > 1: # python 3.2 or superior
output += array.array("i", offsets).tobytes()
else:
output += array.array("i", offsets).tostring() output += array.array("i", offsets).tostring()
output += ids output += ids
output += strs output += strs
return output return output
def _encode(self, mixed): def _encode(self, mixed):
""" """
Encodes the given ``mixed`` argument with the file encoding if and Encodes the given ``mixed`` argument with the file encoding if and
only if it's an unicode string and returns the encoded string. only if it's an unicode string and returns the encoded string.
""" """
if type(mixed) == types.UnicodeType: if isinstance(mixed, text_type):
return mixed.encode(self.encoding) mixed = mixed.encode(self.encoding)
return mixed return mixed
# }}} # }}}
# class POFile {{{ # class POFile {{{
class POFile(_BaseFile): class POFile(_BaseFile):
""" """
Po (or Pot) file reader/writer. Po (or Pot) file reader/writer.
This class inherits the :class:`~polib._BaseFile` class and, by extension, This class inherits the :class:`~polib._BaseFile` class and, by extension,
the python ``list`` type. the python ``list`` type.
""" """
def __unicode__(self): def __unicode__(self):
""" """
Returns the unicode representation of the po file. Returns the unicode representation of the po file.
""" """
ret, headers = '', self.header.split('\n') ret, headers = '', self.header.split('\n')
for header in headers: for header in headers:
if header[:1] in [',', ':']: if not len(header):
ret += "#\n"
elif header[:1] in [',', ':']:
ret += '#%s\n' % header ret += '#%s\n' % header
else: else:
ret += '# %s\n' % header ret += '# %s\n' % header
if type(ret) != types.UnicodeType: if not isinstance(ret, text_type):
ret = unicode(ret, self.encoding) ret = ret.decode(self.encoding)
return ret + _BaseFile.__unicode__(self) return ret + _BaseFile.__unicode__(self)
def save_as_mofile(self, fpath): def save_as_mofile(self, fpath):
""" """
Saves the binary representation of the file to given ``fpath``. Saves the binary representation of the file to given ``fpath``.
Keyword argument: Keyword argument:
``fpath`` ``fpath``
string, full or relative path to the mo file. string, full or relative path to the mo file.
""" """
_BaseFile.save(self, fpath, 'to_binary') _BaseFile.save(self, fpath, 'to_binary')
def percent_translated(self): def percent_translated(self):
""" """
Convenience method that returns the percentage of translated Convenience method that returns the percentage of translated
messages. messages.
""" """
total = len([e for e in self if not e.obsolete]) total = len([e for e in self if not e.obsolete])
if total == 0: if total == 0:
return 100 return 100
translated = len(self.translated_entries()) translated = len(self.translated_entries())
return int((100.00 / float(total)) * translated) return int(translated * 100 / float(total))
def translated_entries(self): def translated_entries(self):
""" """
Convenience method that returns the list of translated entries. Convenience method that returns the list of translated entries.
""" """
return [e for e in self if e.translated()] return [e for e in self if e.translated()]
def untranslated_entries(self): def untranslated_entries(self):
""" """
Convenience method that returns the list of untranslated entries. Convenience method that returns the list of untranslated entries.
""" """
return [e for e in self if not e.translated() and not e.obsolete \ return [e for e in self if not e.translated() and not e.obsolete
and not 'fuzzy' in e.flags] and not 'fuzzy' in e.flags]
def fuzzy_entries(self): def fuzzy_entries(self):
""" """
Convenience method that returns the list of fuzzy entries. Convenience method that returns the list of fuzzy entries.
""" """
return [e for e in self if 'fuzzy' in e.flags] return [e for e in self if 'fuzzy' in e.flags]
dot comments and file positions will be preserved; dot comments and file positions will be preserved;
* the fuzzy flags are preserved. * the fuzzy flags are preserved.
Keyword argument: Keyword argument:
``refpot`` ``refpot``
object POFile, the reference catalog. object POFile, the reference catalog.
""" """
# Store entries in dict/set for faster access
self_entries = dict((entry.msgid, entry) for entry in self)
refpot_msgids = set(entry.msgid for entry in refpot)
# Merge entries that are in the refpot
for entry in refpot: for entry in refpot:
e = self.find(entry.msgid, include_obsolete_entries=True) e = self_entries.get(entry.msgid)
if e is None: if e is None:
e = POEntry() e = POEntry()
self.append(e) self.append(e)
e.merge(entry) e.merge(entry)
# ok, now we must "obsolete" entries that are not in the refpot anymore # ok, now we must "obsolete" entries that are not in the refpot anymore
for entry in self: for entry in self:
if refpot.find(entry.msgid) is None: if entry.msgid not in refpot_msgids:
entry.obsolete = True entry.obsolete = True
# }}} # }}}
# class MOFile {{{ # class MOFile {{{
class MOFile(_BaseFile): class MOFile(_BaseFile):
""" """
Mo file reader/writer. Mo file reader/writer.
This class inherits the :class:`~polib._BaseFile` class and, by This class inherits the :class:`~polib._BaseFile` class and, by
extension, the python ``list`` type. extension, the python ``list`` type.
""" """
BIG_ENDIAN = 0xde120495 MAGIC = 0x950412de
LITTLE_ENDIAN = 0x950412de MAGIC_SWAPPED = 0xde120495
def __init__(self, *args, **kwargs): def __init__(self, *args, **kwargs):
""" """
Constructor, accepts all keywords arguments accepted by Constructor, accepts all keywords arguments accepted by
:class:`~polib._BaseFile` class. :class:`~polib._BaseFile` class.
""" """
_BaseFile.__init__(self, *args, **kwargs) _BaseFile.__init__(self, *args, **kwargs)
self.magic_number = None self.magic_number = None
""" """
return [] return []
def obsolete_entries(self): def obsolete_entries(self):
""" """
Convenience method to keep the same interface with POFile instances. Convenience method to keep the same interface with POFile instances.
""" """
return [] return []
# }}} # }}}
# class _BaseEntry {{{ # class _BaseEntry {{{
class _BaseEntry(object): class _BaseEntry(object):
""" """
Base class for :class:`~polib.POEntry` and :class:`~polib.MOEntry` classes. Base class for :class:`~polib.POEntry` and :class:`~polib.MOEntry` classes.
This class should **not** be instanciated directly. This class should **not** be instanciated directly.
""" """
def __init__(self, *args, **kwargs): def __init__(self, *args, **kwargs):
""" """
""" """
if self.obsolete: if self.obsolete:
delflag = '#~ ' delflag = '#~ '
else: else:
delflag = '' delflag = ''
ret = [] ret = []
# write the msgctxt if any # write the msgctxt if any
if self.msgctxt is not None: if self.msgctxt is not None:
ret += self._str_field("msgctxt", delflag, "", self.msgctxt, wrapwidth) ret += self._str_field("msgctxt", delflag, "", self.msgctxt,
wrapwidth)
# write the msgid # write the msgid
ret += self._str_field("msgid", delflag, "", self.msgid, wrapwidth) ret += self._str_field("msgid", delflag, "", self.msgid, wrapwidth)
# write the msgid_plural if any # write the msgid_plural if any
if self.msgid_plural: if self.msgid_plural:
ret += self._str_field("msgid_plural", delflag, "", self.msgid_plural, wrapwidth) ret += self._str_field("msgid_plural", delflag, "",
self.msgid_plural, wrapwidth)
if self.msgstr_plural: if self.msgstr_plural:
# write the msgstr_plural if any # write the msgstr_plural if any
msgstrs = self.msgstr_plural msgstrs = self.msgstr_plural
keys = list(msgstrs) keys = list(msgstrs)
keys.sort() keys.sort()
for index in keys: for index in keys:
msgstr = msgstrs[index] msgstr = msgstrs[index]
plural_index = '[%s]' % index plural_index = '[%s]' % index
ret += self._str_field("msgstr", delflag, plural_index, msgstr, wrapwidth) ret += self._str_field("msgstr", delflag, plural_index, msgstr,
wrapwidth)
else: else:
# otherwise write the msgstr # otherwise write the msgstr
ret += self._str_field("msgstr", delflag, "", self.msgstr, wrapwidth) ret += self._str_field("msgstr", delflag, "", self.msgstr,
wrapwidth)
ret.append('') ret.append('')
ret = '\n'.join(ret) ret = u('\n').join(ret)
if type(ret) != types.UnicodeType:
return unicode(ret, self.encoding)
return ret return ret
if PY3:
def __str__(self):
return self.__unicode__()
else:
def __str__(self): def __str__(self):
""" """
Returns the string representation of the entry. Returns the string representation of the entry.
""" """
return unicode(self).encode(self.encoding) return unicode(self).encode(self.encoding)
def __eq__(self, other): def __eq__(self, other):
return unicode(self) == unicode(other) return str(self) == str(other)
def _str_field(self, fieldname, delflag, plural_index, field, wrapwidth=78): def _str_field(self, fieldname, delflag, plural_index, field,
wrapwidth=78):
lines = field.splitlines(True) lines = field.splitlines(True)
if len(lines) > 1: if len(lines) > 1:
lines = [''] + lines # start with initial empty line lines = [''] + lines # start with initial empty line
else: else:
escaped_field = escape(field) escaped_field = escape(field)
specialchars_count = 0 specialchars_count = 0
for c in ['\\', '\n', '\r', '\t', '"']: for c in ['\\', '\n', '\r', '\t', '"']:
specialchars_count += field.count(c) specialchars_count += field.count(c)
# comparison must take into account fieldname length + one space # comparison must take into account fieldname length + one space
# + 2 quotes (eg. msgid "<string>") # + 2 quotes (eg. msgid "<string>")
flength = len(fieldname) + 3 flength = len(fieldname) + 3
if plural_index: if plural_index:
flength += len(plural_index) flength += len(plural_index)
real_wrapwidth = wrapwidth - flength + specialchars_count real_wrapwidth = wrapwidth - flength + specialchars_count
if wrapwidth > 0 and len(field) > real_wrapwidth: if wrapwidth > 0 and len(field) > real_wrapwidth:
# Wrap the line but take field name into account # Wrap the line but take field name into account
lines = [''] + [unescape(item) for item in textwrap.wrap( lines = [''] + [unescape(item) for item in wrap(
escaped_field, escaped_field,
wrapwidth - 2, # 2 for quotes "" wrapwidth - 2, # 2 for quotes ""
drop_whitespace=False, drop_whitespace=False,
break_long_words=False break_long_words=False
)] )]
else: else:
lines = [field] lines = [field]
if fieldname.startswith('previous_'): if fieldname.startswith('previous_'):
# quick and dirty trick to get the real field name # quick and dirty trick to get the real field name
fieldname = fieldname[9:] fieldname = fieldname[9:]
ret = ['%s%s%s "%s"' % (delflag, fieldname, plural_index, ret = ['%s%s%s "%s"' % (delflag, fieldname, plural_index,
escape(lines.pop(0)))] escape(lines.pop(0)))]
for mstr in lines: for line in lines:
ret.append('%s"%s"' % (delflag, escape(mstr))) ret.append('%s"%s"' % (delflag, escape(line)))
return ret return ret
# }}} # }}}
# class POEntry {{{ # class POEntry {{{
class POEntry(_BaseEntry): class POEntry(_BaseEntry):
""" """
Represents a po file entry. Represents a po file entry.
""" """
def __init__(self, *args, **kwargs): def __init__(self, *args, **kwargs):
""" """
Constructor, accepts the following keyword arguments: Constructor, accepts the following keyword arguments:
``previous_msgctxt`` ``previous_msgctxt``
string, the entry previous context. string, the entry previous context.
``previous_msgid`` ``previous_msgid``
string, the entry previous msgid. string, the entry previous msgid.
``previous_msgid_plural`` ``previous_msgid_plural``
string, the entry previous msgid_plural. string, the entry previous msgid_plural.
``linenum``
integer, the line number of the entry
""" """
_BaseEntry.__init__(self, *args, **kwargs) _BaseEntry.__init__(self, *args, **kwargs)
self.comment = kwargs.get('comment', '') self.comment = kwargs.get('comment', '')
self.tcomment = kwargs.get('tcomment', '') self.tcomment = kwargs.get('tcomment', '')
self.occurrences = kwargs.get('occurrences', []) self.occurrences = kwargs.get('occurrences', [])
self.flags = kwargs.get('flags', []) self.flags = kwargs.get('flags', [])
self.previous_msgctxt = kwargs.get('previous_msgctxt', None) self.previous_msgctxt = kwargs.get('previous_msgctxt', None)
self.previous_msgid = kwargs.get('previous_msgid', None) self.previous_msgid = kwargs.get('previous_msgid', None)
self.previous_msgid_plural = kwargs.get('previous_msgid_plural', None) self.previous_msgid_plural = kwargs.get('previous_msgid_plural', None)
self.linenum = kwargs.get('linenum', None)
def __unicode__(self, wrapwidth=78): def __unicode__(self, wrapwidth=78):
""" """
Returns the unicode representation of the entry. Returns the unicode representation of the entry.
""" """
if self.obsolete: if self.obsolete:
return _BaseEntry.__unicode__(self, wrapwidth) return _BaseEntry.__unicode__(self, wrapwidth)
ret = [] ret = []
# comments first, if any (with text wrapping as xgettext does) # comments first, if any (with text wrapping as xgettext does)
comments = [('comment', '#. '), ('tcomment', '# ')] comments = [('comment', '#. '), ('tcomment', '# ')]
for c in comments: for c in comments:
val = getattr(self, c[0]) val = getattr(self, c[0])
if val: if val:
for comment in val.split('\n'): for comment in val.split('\n'):
if wrapwidth > 0 and len(comment) + len(c[1]) > wrapwidth: if wrapwidth > 0 and len(comment) + len(c[1]) > wrapwidth:
ret += textwrap.wrap( ret += wrap(
comment, comment,
wrapwidth, wrapwidth,
initial_indent=c[1], initial_indent=c[1],
subsequent_indent=c[1], subsequent_indent=c[1],
break_long_words=False break_long_words=False
) )
else: else:
ret.append('%s%s' % (c[1], comment)) ret.append('%s%s' % (c[1], comment))
# occurrences (with text wrapping as xgettext does) # occurrences (with text wrapping as xgettext does)
if self.occurrences: if self.occurrences:
filelist = [] filelist = []
for fpath, lineno in self.occurrences: for fpath, lineno in self.occurrences:
if lineno: if lineno:
filelist.append('%s:%s' % (fpath, lineno)) filelist.append('%s:%s' % (fpath, lineno))
else: else:
filelist.append(fpath) filelist.append(fpath)
filestr = ' '.join(filelist) filestr = ' '.join(filelist)
if wrapwidth > 0 and len(filestr) + 3 > wrapwidth: if wrapwidth > 0 and len(filestr) + 3 > wrapwidth:
# textwrap split words that contain hyphen, this is not # textwrap split words that contain hyphen, this is not
# what we want for filenames, so the dirty hack is to # what we want for filenames, so the dirty hack is to
# temporally replace hyphens with a char that a file cannot # temporally replace hyphens with a char that a file cannot
# contain, like "*" # contain, like "*"
ret += [l.replace('*', '-') for l in textwrap.wrap( ret += [l.replace('*', '-') for l in wrap(
filestr.replace('-', '*'), filestr.replace('-', '*'),
wrapwidth, wrapwidth,
initial_indent='#: ', initial_indent='#: ',
subsequent_indent='#: ', subsequent_indent='#: ',
break_long_words=False break_long_words=False
)] )]
else: else:
ret.append('#: ' + filestr) ret.append('#: ' + filestr)
# flags (TODO: wrapping ?) # flags (TODO: wrapping ?)
if self.flags: if self.flags:
ret.append('#, %s' % ', '.join(self.flags)) ret.append('#, %s' % ', '.join(self.flags))
# previous context and previous msgid/msgid_plural # previous context and previous msgid/msgid_plural
fields = ['previous_msgctxt', 'previous_msgid', 'previous_msgid_plural'] fields = ['previous_msgctxt', 'previous_msgid',
'previous_msgid_plural']
for f in fields: for f in fields:
val = getattr(self, f) val = getattr(self, f)
if val: if val:
ret += self._str_field(f, "#| ", "", val, wrapwidth) ret += self._str_field(f, "#| ", "", val, wrapwidth)
ret.append(_BaseEntry.__unicode__(self, wrapwidth)) ret.append(_BaseEntry.__unicode__(self, wrapwidth))
ret = '\n'.join(ret) ret = u('\n').join(ret)
if type(ret) != types.UnicodeType: assert isinstance(ret, text_type)
return unicode(ret, self.encoding) #if type(ret) != types.UnicodeType:
# return unicode(ret, self.encoding)
return ret return ret
def __cmp__(self, other): def __cmp__(self, other):
""" """
Called by comparison operations if rich comparison is not defined. Called by comparison operations if rich comparison is not defined.
""" """
def compare_occurrences(a, b):
"""
Compare an entry occurrence with another one.
"""
if a[0] != b[0]:
return a[0] < b[0]
if a[1] != b[1]:
return a[1] < b[1]
return 0
# First: Obsolete test # First: Obsolete test
if self.obsolete != other.obsolete: if self.obsolete != other.obsolete:
if self.obsolete: if self.obsolete:
return -1 return -1
else: else:
return 1 return 1
# Work on a copy to protect original # Work on a copy to protect original
occ1 = self.occurrences[:] occ1 = sorted(self.occurrences[:])
occ2 = other.occurrences[:] occ2 = sorted(other.occurrences[:])
# Sorting using compare method
occ1.sort(compare_occurrences)
occ2.sort(compare_occurrences)
# Comparing sorted occurrences
pos = 0 pos = 0
for entry1 in occ1: for entry1 in occ1:
try: try:
entry2 = occ2[pos] entry2 = occ2[pos]
except IndexError: except IndexError:
return 1 return 1
pos = pos + 1 pos = pos + 1
if entry1[0] != entry2[0]: if entry1[0] != entry2[0]:
if entry1[0] > entry2[0]: if entry1[0] > entry2[0]:
return 1 return 1
else: else:
return -1 return -1
if entry1[1] != entry2[1]: if entry1[1] != entry2[1]:
if entry1[1] > entry2[1]: if entry1[1] > entry2[1]:
return 1 return 1
else: else:
return -1 return -1
# Compare msgid_plural if set
if self.msgid_plural:
if not other.msgid_plural:
return 1
for pos in self.msgid_plural:
if pos not in other.msgid_plural:
return 1
if self.msgid_plural[pos] > other.msgid_plural[pos]:
return 1
if self.msgid_plural[pos] < other.msgid_plural[pos]:
return -1
# Finally: Compare message ID # Finally: Compare message ID
if self.msgid > other.msgid: return 1 if self.msgid > other.msgid:
else: return -1 return 1
elif self.msgid < other.msgid:
return -1
return 0
def __gt__(self, other):
return self.__cmp__(other) > 0
def __lt__(self, other):
return self.__cmp__(other) < 0
def __ge__(self, other):
return self.__cmp__(other) >= 0
def __le__(self, other):
return self.__cmp__(other) <= 0
def __eq__(self, other):
return self.__cmp__(other) == 0
def __ne__(self, other):
return self.__cmp__(other) != 0
def translated(self): def translated(self):
""" """
Returns ``True`` if the entry has been translated or ``False`` Returns ``True`` if the entry has been translated or ``False``
otherwise. otherwise.
""" """
if self.obsolete or 'fuzzy' in self.flags: if self.obsolete or 'fuzzy' in self.flags:
return False return False
if other.msgstr_plural: if other.msgstr_plural:
for pos in other.msgstr_plural: for pos in other.msgstr_plural:
try: try:
# keep existing translation at pos if any # keep existing translation at pos if any
self.msgstr_plural[pos] self.msgstr_plural[pos]
except KeyError: except KeyError:
self.msgstr_plural[pos] = '' self.msgstr_plural[pos] = ''
def __hash__(self):
return hash((self.msgid, self.msgstr))
# }}} # }}}
# class MOEntry {{{ # class MOEntry {{{
class MOEntry(_BaseEntry): class MOEntry(_BaseEntry):
""" """
Represents a mo file entry. Represents a mo file entry.
""" """
pass def __init__(self, *args, **kwargs):
"""
Constructor, accepts the following keyword arguments,
for consistency with :class:`~polib.POEntry`:
``comment``
``tcomment``
``occurrences``
``flags``
``previous_msgctxt``
``previous_msgid``
``previous_msgid_plural``
Note: even though these keyword arguments are accepted,
they hold no real meaning in the context of MO files
and are simply ignored.
"""
_BaseEntry.__init__(self, *args, **kwargs)
self.comment = ''
self.tcomment = ''
self.occurrences = []
self.flags = []
self.previous_msgctxt = None
self.previous_msgid = None
self.previous_msgid_plural = None
def __hash__(self):
return hash((self.msgid, self.msgstr))
# }}} # }}}
# class _POFileParser {{{ # class _POFileParser {{{
class _POFileParser(object): class _POFileParser(object):
""" """
A finite state machine to parse efficiently and correctly po A finite state machine to parse efficiently and correctly po
file format. file format.
""" """
def __init__(self, pofile, *args, **kwargs): def __init__(self, pofile, *args, **kwargs):
""" """
Constructor. Constructor.
Keyword arguments: Keyword arguments:
``pofile`` ``pofile``
string, path to the po file or its content string, path to the po file or its content
``encoding`` ``encoding``
string, the encoding to use, defaults to ``default_encoding`` string, the encoding to use, defaults to ``default_encoding``
global variable (optional). global variable (optional).
``check_for_duplicates`` ``check_for_duplicates``
whether to check for duplicate entries when adding entries to the whether to check for duplicate entries when adding entries to the
file (optional, default: ``False``). file (optional, default: ``False``).
""" """
enc = kwargs.get('encoding', default_encoding) enc = kwargs.get('encoding', default_encoding)
if os.path.exists(pofile): if _is_file(pofile):
try: try:
self.fhandle = codecs.open(pofile, 'rU', enc) self.fhandle = io.open(pofile, 'rt', encoding=enc)
except LookupError: except LookupError:
enc = default_encoding enc = default_encoding
self.fhandle = codecs.open(pofile, 'rU', enc) self.fhandle = io.open(pofile, 'rt', encoding=enc)
else: else:
self.fhandle = pofile.splitlines() self.fhandle = pofile.splitlines()
self.instance = POFile( klass = kwargs.get('klass')
if klass is None:
klass = POFile
self.instance = klass(
pofile=pofile, pofile=pofile,
encoding=enc, encoding=enc,
check_for_duplicates=kwargs.get('check_for_duplicates', False) check_for_duplicates=kwargs.get('check_for_duplicates', False)
) )
self.transitions = {} self.transitions = {}
self.current_entry = POEntry() self.current_line = 0
self.current_state = 'ST' self.current_entry = POEntry(linenum=self.current_line)
self.current_state = 'st'
self.current_token = None self.current_token = None
# two memo flags used in handlers # two memo flags used in handlers
self.msgstr_index = 0 self.msgstr_index = 0
self.entry_obsolete = 0 self.entry_obsolete = 0
# Configure the state machine, by adding transitions. # Configure the state machine, by adding transitions.
# Signification of symbols: # Signification of symbols:
# * ST: Beginning of the file (start) # * ST: Beginning of the file (start)
# * HE: Header # * HE: Header
# * TC: a translation comment # * TC: a translation comment
# * GC: a generated comment # * GC: a generated comment
# * OC: a file/line occurence # * OC: a file/line occurrence
# * FL: a flags line # * FL: a flags line
# * CT: a message context # * CT: a message context
# * PC: a previous msgctxt # * PC: a previous msgctxt
# * PM: a previous msgid # * PM: a previous msgid
# * PP: a previous msgid_plural # * PP: a previous msgid_plural
# * MI: a msgid # * MI: a msgid
# * MP: a msgid plural # * MP: a msgid plural
# * MS: a msgstr # * MS: a msgstr
# * MX: a msgstr plural # * MX: a msgstr plural
# * MC: a msgid or msgstr continuation line # * MC: a msgid or msgstr continuation line
all = ['ST', 'HE', 'GC', 'OC', 'FL', 'CT', 'PC', 'PM', 'PP', 'TC', all = ['st', 'he', 'gc', 'oc', 'fl', 'ct', 'pc', 'pm', 'pp', 'tc',
'MS', 'MP', 'MX', 'MI'] 'ms', 'mp', 'mx', 'mi']
self.add('TC', ['ST', 'HE'], 'HE') self.add('tc', ['st', 'he'], 'he')
self.add('TC', ['GC', 'OC', 'FL', 'TC', 'PC', 'PM', 'PP', 'MS', self.add('tc', ['gc', 'oc', 'fl', 'tc', 'pc', 'pm', 'pp', 'ms',
'MP', 'MX', 'MI'], 'TC') 'mp', 'mx', 'mi'], 'tc')
self.add('GC', all, 'GC') self.add('gc', all, 'gc')
self.add('OC', all, 'OC') self.add('oc', all, 'oc')
self.add('FL', all, 'FL') self.add('fl', all, 'fl')
self.add('PC', all, 'PC') self.add('pc', all, 'pc')
self.add('PM', all, 'PM') self.add('pm', all, 'pm')
self.add('PP', all, 'PP') self.add('pp', all, 'pp')
self.add('CT', ['ST', 'HE', 'GC', 'OC', 'FL', 'TC', 'PC', 'PM', self.add('ct', ['st', 'he', 'gc', 'oc', 'fl', 'tc', 'pc', 'pm',
'PP', 'MS', 'MX'], 'CT') 'pp', 'ms', 'mx'], 'ct')
self.add('MI', ['ST', 'HE', 'GC', 'OC', 'FL', 'CT', 'TC', 'PC', self.add('mi', ['st', 'he', 'gc', 'oc', 'fl', 'ct', 'tc', 'pc',
'PM', 'PP', 'MS', 'MX'], 'MI') 'pm', 'pp', 'ms', 'mx'], 'mi')
self.add('MP', ['TC', 'GC', 'PC', 'PM', 'PP', 'MI'], 'MP') self.add('mp', ['tc', 'gc', 'pc', 'pm', 'pp', 'mi'], 'mp')
self.add('MS', ['MI', 'MP', 'TC'], 'MS') self.add('ms', ['mi', 'mp', 'tc'], 'ms')
self.add('MX', ['MI', 'MX', 'MP', 'TC'], 'MX') self.add('mx', ['mi', 'mx', 'mp', 'tc'], 'mx')
self.add('MC', ['CT', 'MI', 'MP', 'MS', 'MX', 'PM', 'PP', 'PC'], 'MC') self.add('mc', ['ct', 'mi', 'mp', 'ms', 'mx', 'pm', 'pp', 'pc'], 'mc')
def parse(self): def parse(self):
""" """
Run the state machine, parse the file line by line and call process() Run the state machine, parse the file line by line and call process()
with the current matched symbol. with the current matched symbol.
""" """
i = 0
keywords = { keywords = {
'msgctxt': 'CT', 'msgctxt': 'ct',
'msgid': 'MI', 'msgid': 'mi',
'msgstr': 'MS', 'msgstr': 'ms',
'msgid_plural': 'MP', 'msgid_plural': 'mp',
} }
prev_keywords = { prev_keywords = {
'msgid_plural': 'PP', 'msgid_plural': 'pp',
'msgid': 'PM', 'msgid': 'pm',
'msgctxt': 'PC', 'msgctxt': 'pc',
} }
tokens = []
for line in self.fhandle: for line in self.fhandle:
i += 1 self.current_line += 1
line = line.strip() line = line.strip()
if line == '': if line == '':
continue continue
tokens = line.split(None, 2) tokens = line.split(None, 2)
nb_tokens = len(tokens) nb_tokens = len(tokens)
if tokens[0] == '#~|':
continue
if tokens[0] == '#~' and nb_tokens > 1: if tokens[0] == '#~' and nb_tokens > 1:
line = line[3:].strip() line = line[3:].strip()
tokens = tokens[1:] tokens = tokens[1:]
nb_tokens -= 1 nb_tokens -= 1
self.entry_obsolete = 1 self.entry_obsolete = 1
else: else:
self.entry_obsolete = 0 self.entry_obsolete = 0
# Take care of keywords like # Take care of keywords like
# msgid, msgid_plural, msgctxt & msgstr. # msgid, msgid_plural, msgctxt & msgstr.
if tokens[0] in keywords and nb_tokens > 1: if tokens[0] in keywords and nb_tokens > 1:
line = line[len(tokens[0]):].lstrip() line = line[len(tokens[0]):].lstrip()
if re.search(r'([^\\]|^)"', line[1:-1]):
raise IOError('Syntax error in po file %s (line %s): '
'unescaped double quote found' %
(self.instance.fpath, self.current_line))
self.current_token = line self.current_token = line
self.process(keywords[tokens[0]], i) self.process(keywords[tokens[0]])
continue continue
self.current_token = line self.current_token = line
if tokens[0] == '#:' and nb_tokens > 1: if tokens[0] == '#:':
if nb_tokens <= 1:
continue
# we are on a occurrences line # we are on a occurrences line
self.process('OC', i) self.process('oc')
elif line[:1] == '"': elif line[:1] == '"':
# we are on a continuation line # we are on a continuation line
self.process('MC', i) if re.search(r'([^\\]|^)"', line[1:-1]):
raise IOError('Syntax error in po file %s (line %s): '
'unescaped double quote found' %
(self.instance.fpath, self.current_line))
self.process('mc')
elif line[:7] == 'msgstr[': elif line[:7] == 'msgstr[':
# we are on a msgstr plural # we are on a msgstr plural
self.process('MX', i) self.process('mx')
elif tokens[0] == '#,' and nb_tokens > 1: elif tokens[0] == '#,':
if nb_tokens <= 1:
continue
# we are on a flags line # we are on a flags line
self.process('FL', i) self.process('fl')
elif tokens[0] == '#': elif tokens[0] == '#' or tokens[0].startswith('##'):
if line == '#': line += ' ' if line == '#':
line += ' '
# we are on a translator comment line # we are on a translator comment line
self.process('TC', i) self.process('tc')
elif tokens[0] == '#.' and nb_tokens > 1: elif tokens[0] == '#.':
if nb_tokens <= 1:
continue
# we are on a generated comment line # we are on a generated comment line
self.process('GC', i) self.process('gc')
elif tokens[0] == '#|': elif tokens[0] == '#|':
if nb_tokens < 2: if nb_tokens <= 1:
self.process('??', i) raise IOError('Syntax error in po file %s (line %s)' %
continue (self.instance.fpath, self.current_line))
# Remove the marker and any whitespace right after that. # Remove the marker and any whitespace right after that.
line = line[2:].lstrip() line = line[2:].lstrip()
self.current_token = line self.current_token = line
if tokens[1].startswith('"'): if tokens[1].startswith('"'):
# Continuation of previous metadata. # Continuation of previous metadata.
self.process('MC', i) self.process('mc')
continue continue
if nb_tokens == 2: if nb_tokens == 2:
# Invalid continuation line. # Invalid continuation line.
self.process('??', i) raise IOError('Syntax error in po file %s (line %s): '
'invalid continuation line' %
(self.instance.fpath, self.current_line))
# we are on a "previous translation" comment line, # we are on a "previous translation" comment line,
if tokens[1] not in prev_keywords: if tokens[1] not in prev_keywords:
# Unknown keyword in previous translation comment. # Unknown keyword in previous translation comment.
self.process('??', i) raise IOError('Syntax error in po file %s (line %s): '
'unknown keyword %s' %
(self.instance.fpath, self.current_line,
tokens[1]))
# Remove the keyword and any whitespace # Remove the keyword and any whitespace
# between it and the starting quote. # between it and the starting quote.
line = line[len(tokens[1]):].lstrip() line = line[len(tokens[1]):].lstrip()
self.current_token = line self.current_token = line
self.process(prev_keywords[tokens[1]], i) self.process(prev_keywords[tokens[1]])
else: else:
self.process('??', i) raise IOError('Syntax error in po file %s (line %s)' %
(self.instance.fpath, self.current_line))
if self.current_entry: if self.current_entry and len(tokens) > 0 and \
not tokens[0].startswith('#'):
# since entries are added when another entry is found, we must add # since entries are added when another entry is found, we must add
# the last entry here (only if there are lines) # the last entry here (only if there are lines). Trailing comments
# are ignored
self.instance.append(self.current_entry) self.instance.append(self.current_entry)
# before returning the instance, check if there's metadata and if # before returning the instance, check if there's metadata and if
# so extract it in a dict # so extract it in a dict
firstentry = self.instance[0] metadataentry = self.instance.find('')
if firstentry.msgid == '': # metadata found if metadataentry: # metadata found
# remove the entry # remove the entry
firstentry = self.instance.pop(0) self.instance.remove(metadataentry)
self.instance.metadata_is_fuzzy = firstentry.flags self.instance.metadata_is_fuzzy = metadataentry.flags
key = None key = None
for msg in firstentry.msgstr.splitlines(): for msg in metadataentry.msgstr.splitlines():
try: try:
key, val = msg.split(':', 1) key, val = msg.split(':', 1)
self.instance.metadata[key] = val.strip() self.instance.metadata[key] = val.strip()
except: except (ValueError, KeyError):
if key is not None: if key is not None:
self.instance.metadata[key] += '\n'+ msg.strip() self.instance.metadata[key] += '\n' + msg.strip()
# close opened file # close opened file
if isinstance(self.fhandle, file): if not isinstance(self.fhandle, list): # must be file
self.fhandle.close() self.fhandle.close()
return self.instance return self.instance
def add(self, symbol, states, next_state): def add(self, symbol, states, next_state):
""" """
Add a transition to the state machine. Add a transition to the state machine.
Keywords arguments: Keywords arguments:
``symbol`` ``symbol``
string, the matched token (two chars symbol). string, the matched token (two chars symbol).
``states`` ``states``
list, a list of states (two chars symbols). list, a list of states (two chars symbols).
``next_state`` ``next_state``
the next state the fsm will have after the action. the next state the fsm will have after the action.
""" """
for state in states: for state in states:
action = getattr(self, 'handle_%s' % next_state.lower()) action = getattr(self, 'handle_%s' % next_state)
self.transitions[(symbol, state)] = (action, next_state) self.transitions[(symbol, state)] = (action, next_state)
def process(self, symbol, linenum): def process(self, symbol):
""" """
Process the transition corresponding to the current state and the Process the transition corresponding to the current state and the
symbol provided. symbol provided.
Keywords arguments: Keywords arguments:
``symbol`` ``symbol``
string, the matched token (two chars symbol). string, the matched token (two chars symbol).
``linenum`` ``linenum``
integer, the current line number of the parsed file. integer, the current line number of the parsed file.
""" """
try: try:
(action, state) = self.transitions[(symbol, self.current_state)] (action, state) = self.transitions[(symbol, self.current_state)]
if action(): if action():
self.current_state = state self.current_state = state
except Exception as exc: except Exception:
raise IOError('Syntax error in po file (line %s)' % linenum) raise IOError('Syntax error in po file (line %s)' %
self.current_line)
# state handlers # state handlers
def handle_he(self): def handle_he(self):
"""Handle a header comment.""" """Handle a header comment."""
if self.instance.header != '': if self.instance.header != '':
self.instance.header += '\n' self.instance.header += '\n'
self.instance.header += self.current_token[2:] self.instance.header += self.current_token[2:]
return 1 return 1
def handle_tc(self): def handle_tc(self):
"""Handle a translator comment.""" """Handle a translator comment."""
if self.current_state in ['MC', 'MS', 'MX']: if self.current_state in ['mc', 'ms', 'mx']:
self.instance.append(self.current_entry) self.instance.append(self.current_entry)
self.current_entry = POEntry() self.current_entry = POEntry(linenum=self.current_line)
if self.current_entry.tcomment != '': if self.current_entry.tcomment != '':
self.current_entry.tcomment += '\n' self.current_entry.tcomment += '\n'
self.current_entry.tcomment += self.current_token[2:] tcomment = self.current_token.lstrip('#')
if tcomment.startswith(' '):
tcomment = tcomment[1:]
self.current_entry.tcomment += tcomment
return True return True
def handle_gc(self): def handle_gc(self):
"""Handle a generated comment.""" """Handle a generated comment."""
if self.current_state in ['MC', 'MS', 'MX']: if self.current_state in ['mc', 'ms', 'mx']:
self.instance.append(self.current_entry) self.instance.append(self.current_entry)
self.current_entry = POEntry() self.current_entry = POEntry(linenum=self.current_line)
if self.current_entry.comment != '': if self.current_entry.comment != '':
self.current_entry.comment += '\n' self.current_entry.comment += '\n'
self.current_entry.comment += self.current_token[3:] self.current_entry.comment += self.current_token[3:]
return True return True
def handle_oc(self): def handle_oc(self):
"""Handle a file:num occurence.""" """Handle a file:num occurrence."""
if self.current_state in ['MC', 'MS', 'MX']: if self.current_state in ['mc', 'ms', 'mx']:
self.instance.append(self.current_entry) self.instance.append(self.current_entry)
self.current_entry = POEntry() self.current_entry = POEntry(linenum=self.current_line)
occurrences = self.current_token[3:].split() occurrences = self.current_token[3:].split()
for occurrence in occurrences: for occurrence in occurrences:
if occurrence != '': if occurrence != '':
try: try:
fil, line = occurrence.split(':') fil, line = occurrence.split(':')
if not line.isdigit(): if not line.isdigit():
fil = fil + line fil = fil + line
line = '' line = ''
self.current_entry.occurrences.append((fil, line)) self.current_entry.occurrences.append((fil, line))
except: except (ValueError, AttributeError):
self.current_entry.occurrences.append((occurrence, '')) self.current_entry.occurrences.append((occurrence, ''))
return True return True
def handle_fl(self): def handle_fl(self):
"""Handle a flags line.""" """Handle a flags line."""
if self.current_state in ['MC', 'MS', 'MX']: if self.current_state in ['mc', 'ms', 'mx']:
self.instance.append(self.current_entry) self.instance.append(self.current_entry)
self.current_entry = POEntry() self.current_entry = POEntry(linenum=self.current_line)
self.current_entry.flags += self.current_token[3:].split(', ') self.current_entry.flags += [c.strip() for c in
self.current_token[3:].split(',')]
return True return True
def handle_pp(self): def handle_pp(self):
"""Handle a previous msgid_plural line.""" """Handle a previous msgid_plural line."""
if self.current_state in ['MC', 'MS', 'MX']: if self.current_state in ['mc', 'ms', 'mx']:
self.instance.append(self.current_entry) self.instance.append(self.current_entry)
self.current_entry = POEntry() self.current_entry = POEntry(linenum=self.current_line)
self.current_entry.previous_msgid_plural = \ self.current_entry.previous_msgid_plural = \
unescape(self.current_token[1:-1]) unescape(self.current_token[1:-1])
return True return True
def handle_pm(self): def handle_pm(self):
"""Handle a previous msgid line.""" """Handle a previous msgid line."""
if self.current_state in ['MC', 'MS', 'MX']: if self.current_state in ['mc', 'ms', 'mx']:
self.instance.append(self.current_entry) self.instance.append(self.current_entry)
self.current_entry = POEntry() self.current_entry = POEntry(linenum=self.current_line)
self.current_entry.previous_msgid = \ self.current_entry.previous_msgid = \
unescape(self.current_token[1:-1]) unescape(self.current_token[1:-1])
return True return True
def handle_pc(self): def handle_pc(self):
"""Handle a previous msgctxt line.""" """Handle a previous msgctxt line."""
if self.current_state in ['MC', 'MS', 'MX']: if self.current_state in ['mc', 'ms', 'mx']:
self.instance.append(self.current_entry) self.instance.append(self.current_entry)
self.current_entry = POEntry() self.current_entry = POEntry(linenum=self.current_line)
self.current_entry.previous_msgctxt = \ self.current_entry.previous_msgctxt = \
unescape(self.current_token[1:-1]) unescape(self.current_token[1:-1])
return True return True
def handle_ct(self): def handle_ct(self):
"""Handle a msgctxt.""" """Handle a msgctxt."""
if self.current_state in ['MC', 'MS', 'MX']: if self.current_state in ['mc', 'ms', 'mx']:
self.instance.append(self.current_entry) self.instance.append(self.current_entry)
self.current_entry = POEntry() self.current_entry = POEntry(linenum=self.current_line)
self.current_entry.msgctxt = unescape(self.current_token[1:-1]) self.current_entry.msgctxt = unescape(self.current_token[1:-1])
return True return True
def handle_mi(self): def handle_mi(self):
"""Handle a msgid.""" """Handle a msgid."""
if self.current_state in ['MC', 'MS', 'MX']: if self.current_state in ['mc', 'ms', 'mx']:
self.instance.append(self.current_entry) self.instance.append(self.current_entry)
self.current_entry = POEntry() self.current_entry = POEntry(linenum=self.current_line)
self.current_entry.obsolete = self.entry_obsolete self.current_entry.obsolete = self.entry_obsolete
self.current_entry.msgid = unescape(self.current_token[1:-1]) self.current_entry.msgid = unescape(self.current_token[1:-1])
return True return True
def handle_mp(self): def handle_mp(self):
"""Handle a msgid plural.""" """Handle a msgid plural."""
self.current_entry.msgid_plural = unescape(self.current_token[1:-1]) self.current_entry.msgid_plural = unescape(self.current_token[1:-1])
return True return True
def handle_ms(self): def handle_ms(self):
"""Handle a msgstr.""" """Handle a msgstr."""
self.current_entry.msgstr = unescape(self.current_token[1:-1]) self.current_entry.msgstr = unescape(self.current_token[1:-1])
return True return True
def handle_mx(self): def handle_mx(self):
"""Handle a msgstr plural.""" """Handle a msgstr plural."""
index, value = self.current_token[7], self.current_token[11:-1] index = self.current_token[7]
self.current_entry.msgstr_plural[index] = unescape(value) value = self.current_token[self.current_token.find('"') + 1:-1]
self.msgstr_index = index self.current_entry.msgstr_plural[int(index)] = unescape(value)
self.msgstr_index = int(index)
return True return True
def handle_mc(self): def handle_mc(self):
"""Handle a msgid or msgstr continuation line.""" """Handle a msgid or msgstr continuation line."""
token = unescape(self.current_token[1:-1]) token = unescape(self.current_token[1:-1])
if self.current_state == 'CT': if self.current_state == 'ct':
typ = 'msgctxt'
self.current_entry.msgctxt += token self.current_entry.msgctxt += token
elif self.current_state == 'MI': elif self.current_state == 'mi':
typ = 'msgid'
self.current_entry.msgid += token self.current_entry.msgid += token
elif self.current_state == 'MP': elif self.current_state == 'mp':
typ = 'msgid_plural'
self.current_entry.msgid_plural += token self.current_entry.msgid_plural += token
elif self.current_state == 'MS': elif self.current_state == 'ms':
typ = 'msgstr'
self.current_entry.msgstr += token self.current_entry.msgstr += token
elif self.current_state == 'MX': elif self.current_state == 'mx':
typ = 'msgstr[%s]' % self.msgstr_index
self.current_entry.msgstr_plural[self.msgstr_index] += token self.current_entry.msgstr_plural[self.msgstr_index] += token
elif self.current_state == 'PP': elif self.current_state == 'pp':
typ = 'previous_msgid_plural'
token = token[3:]
self.current_entry.previous_msgid_plural += token self.current_entry.previous_msgid_plural += token
elif self.current_state == 'PM': elif self.current_state == 'pm':
typ = 'previous_msgid'
token = token[3:]
self.current_entry.previous_msgid += token self.current_entry.previous_msgid += token
elif self.current_state == 'PC': elif self.current_state == 'pc':
typ = 'previous_msgctxt'
token = token[3:]
self.current_entry.previous_msgctxt += token self.current_entry.previous_msgctxt += token
# don't change the current state # don't change the current state
return False return False
# }}} # }}}
# class _MOFileParser {{{ # class _MOFileParser {{{
class _MOFileParser(object): class _MOFileParser(object):
""" """
A class to parse binary mo files. A class to parse binary mo files.
""" """
def __init__(self, mofile, *args, **kwargs): def __init__(self, mofile, *args, **kwargs):
""" """
Constructor. Constructor.
Keyword arguments: Keyword arguments:
``mofile`` ``mofile``
string, path to the mo file or its content string, path to the mo file or its content
``encoding`` ``encoding``
string, the encoding to use, defaults to ``default_encoding`` string, the encoding to use, defaults to ``default_encoding``
global variable (optional). global variable (optional).
``check_for_duplicates`` ``check_for_duplicates``
whether to check for duplicate entries when adding entries to the whether to check for duplicate entries when adding entries to the
file (optional, default: ``False``). file (optional, default: ``False``).
""" """
self.fhandle = open(mofile, 'rb') self.fhandle = open(mofile, 'rb')
self.instance = MOFile(
klass = kwargs.get('klass')
if klass is None:
klass = MOFile
self.instance = klass(
fpath=mofile, fpath=mofile,
encoding=kwargs.get('encoding', default_encoding), encoding=kwargs.get('encoding', default_encoding),
check_for_duplicates=kwargs.get('check_for_duplicates', False) check_for_duplicates=kwargs.get('check_for_duplicates', False)
) )
def __del__(self):
"""
Make sure the file is closed, this prevents warnings on unclosed file
when running tests with python >= 3.2.
"""
if self.fhandle:
self.fhandle.close()
def parse(self): def parse(self):
""" """
Build the instance with the file handle provided in the Build the instance with the file handle provided in the
constructor. constructor.
""" """
# parse magic number # parse magic number
magic_number = self._readbinary('<I', 4) magic_number = self._readbinary('<I', 4)
if magic_number == MOFile.LITTLE_ENDIAN: if magic_number == MOFile.MAGIC:
ii = '<II' ii = '<II'
elif magic_number == MOFile.BIG_ENDIAN: elif magic_number == MOFile.MAGIC_SWAPPED:
ii = '>II' ii = '>II'
else: else:
raise IOError('Invalid mo file, magic number is incorrect !') raise IOError('Invalid mo file, magic number is incorrect !')
self.instance.magic_number = magic_number self.instance.magic_number = magic_number
# parse the version number and the number of strings # parse the version number and the number of strings
self.instance.version, numofstrings = self._readbinary(ii, 8) version, numofstrings = self._readbinary(ii, 8)
# from MO file format specs: "A program seeing an unexpected major
# revision number should stop reading the MO file entirely"
if version not in (0, 1):
raise IOError('Invalid mo file, unexpected major revision number')
self.instance.version = version
# original strings and translation strings hash table offset # original strings and translation strings hash table offset
msgids_hash_offset, msgstrs_hash_offset = self._readbinary(ii, 8) msgids_hash_offset, msgstrs_hash_offset = self._readbinary(ii, 8)
# move to msgid hash table and read length and offset of msgids # move to msgid hash table and read length and offset of msgids
self.fhandle.seek(msgids_hash_offset) self.fhandle.seek(msgids_hash_offset)
msgids_index = [] msgids_index = []
for i in range(numofstrings): for i in range(numofstrings):
msgids_index.append(self._readbinary(ii, 8)) msgids_index.append(self._readbinary(ii, 8))
# move to msgstr hash table and read length and offset of msgstrs # move to msgstr hash table and read length and offset of msgstrs
self.fhandle.seek(msgstrs_hash_offset) self.fhandle.seek(msgstrs_hash_offset)
msgstrs_index = [] msgstrs_index = []
for i in range(numofstrings): for i in range(numofstrings):
msgstrs_index.append(self._readbinary(ii, 8)) msgstrs_index.append(self._readbinary(ii, 8))
# build entries # build entries
encoding = self.instance.encoding
for i in range(numofstrings): for i in range(numofstrings):
self.fhandle.seek(msgids_index[i][1]) self.fhandle.seek(msgids_index[i][1])
msgid = self.fhandle.read(msgids_index[i][0]) msgid = self.fhandle.read(msgids_index[i][0])
self.fhandle.seek(msgstrs_index[i][1]) self.fhandle.seek(msgstrs_index[i][1])
msgstr = self.fhandle.read(msgstrs_index[i][0]) msgstr = self.fhandle.read(msgstrs_index[i][0])
if i == 0: # metadata if i == 0 and not msgid: # metadata
raw_metadata, metadata = msgstr.split('\n'), {} raw_metadata, metadata = msgstr.split(b('\n')), {}
for line in raw_metadata: for line in raw_metadata:
tokens = line.split(':', 1) tokens = line.split(b(':'), 1)
if tokens[0] != '': if tokens[0] != b(''):
try: try:
metadata[tokens[0]] = tokens[1].strip() k = tokens[0].decode(encoding)
v = tokens[1].decode(encoding)
metadata[k] = v.strip()
except IndexError: except IndexError:
metadata[tokens[0]] = '' metadata[k] = u('')
self.instance.metadata = metadata self.instance.metadata = metadata
continue continue
# test if we have a plural entry # test if we have a plural entry
msgid_tokens = msgid.split('\0') msgid_tokens = msgid.split(b('\0'))
if len(msgid_tokens) > 1: if len(msgid_tokens) > 1:
entry = self._build_entry( entry = self._build_entry(
msgid=msgid_tokens[0], msgid=msgid_tokens[0],
msgid_plural=msgid_tokens[1], msgid_plural=msgid_tokens[1],
msgstr_plural=dict((k,v) for k,v in enumerate(msgstr.split('\0'))) msgstr_plural=dict((k, v) for k, v in
enumerate(msgstr.split(b('\0'))))
) )
else: else:
entry = self._build_entry(msgid=msgid, msgstr=msgstr) entry = self._build_entry(msgid=msgid, msgstr=msgstr)
self.instance.append(entry) self.instance.append(entry)
# close opened file # close opened file
self.fhandle.close() self.fhandle.close()
return self.instance return self.instance
def _build_entry(self, msgid, msgstr=None, msgid_plural=None, def _build_entry(self, msgid, msgstr=None, msgid_plural=None,
msgstr_plural=None): msgstr_plural=None):
msgctxt_msgid = msgid.split('\x04') msgctxt_msgid = msgid.split(b('\x04'))
encoding = self.instance.encoding
if len(msgctxt_msgid) > 1: if len(msgctxt_msgid) > 1:
kwargs = { kwargs = {
'msgctxt': msgctxt_msgid[0], 'msgctxt': msgctxt_msgid[0].decode(encoding),
'msgid' : msgctxt_msgid[1], 'msgid': msgctxt_msgid[1].decode(encoding),
} }
else: else:
kwargs = {'msgid': msgid} kwargs = {'msgid': msgid.decode(encoding)}
if msgstr: if msgstr:
kwargs['msgstr'] = msgstr kwargs['msgstr'] = msgstr.decode(encoding)
if msgid_plural: if msgid_plural:
kwargs['msgid_plural'] = msgid_plural kwargs['msgid_plural'] = msgid_plural.decode(encoding)
if msgstr_plural: if msgstr_plural:
for k in msgstr_plural:
msgstr_plural[k] = msgstr_plural[k].decode(encoding)
kwargs['msgstr_plural'] = msgstr_plural kwargs['msgstr_plural'] = msgstr_plural
return MOEntry(**kwargs) return MOEntry(**kwargs)
def _readbinary(self, fmt, numbytes): def _readbinary(self, fmt, numbytes):
""" """
Private method that unpack n bytes of data using format <fmt>. Private method that unpack n bytes of data using format <fmt>.
It returns a tuple or a mixed value if the tuple length is 1. It returns a tuple or a mixed value if the tuple length is 1.
""" """
bytes = self.fhandle.read(numbytes) bytes = self.fhandle.read(numbytes)
tup = struct.unpack(fmt, bytes) tup = struct.unpack(fmt, bytes)
if len(tup) == 1: if len(tup) == 1:
return tup[0] return tup[0]
return tup return tup
# }}}
# class TextWrapper {{{
class TextWrapper(textwrap.TextWrapper):
"""
Subclass of textwrap.TextWrapper that backport the
drop_whitespace option.
"""
def __init__(self, *args, **kwargs):
drop_whitespace = kwargs.pop('drop_whitespace', True)
textwrap.TextWrapper.__init__(self, *args, **kwargs)
self.drop_whitespace = drop_whitespace
def _wrap_chunks(self, chunks):
"""_wrap_chunks(chunks : [string]) -> [string]
Wrap a sequence of text chunks and return a list of lines of
length 'self.width' or less. (If 'break_long_words' is false,
some lines may be longer than this.) Chunks correspond roughly
to words and the whitespace between them: each chunk is
indivisible (modulo 'break_long_words'), but a line break can
come between any two chunks. Chunks should not have internal
whitespace; ie. a chunk is either all whitespace or a "word".
Whitespace chunks will be removed from the beginning and end of
lines, but apart from that whitespace is preserved.
"""
lines = []
if self.width <= 0:
raise ValueError("invalid width %r (must be > 0)" % self.width)
# Arrange in reverse order so items can be efficiently popped
# from a stack of chucks.
chunks.reverse()
while chunks:
# Start the list of chunks that will make up the current line.
# cur_len is just the length of all the chunks in cur_line.
cur_line = []
cur_len = 0
# Figure out which static string will prefix this line.
if lines:
indent = self.subsequent_indent
else:
indent = self.initial_indent
# Maximum width for this line.
width = self.width - len(indent)
# First chunk on line is whitespace -- drop it, unless this
# is the very beginning of the text (ie. no lines started yet).
if self.drop_whitespace and chunks[-1].strip() == '' and lines:
del chunks[-1]
while chunks:
l = len(chunks[-1])
# Can at least squeeze this chunk onto the current line.
if cur_len + l <= width:
cur_line.append(chunks.pop())
cur_len += l
# Nope, this line is full.
else:
break
# The current line is full, and the next chunk is too big to
# fit on *any* line (not just this one).
if chunks and len(chunks[-1]) > width:
self._handle_long_word(chunks, cur_line, cur_len, width)
# If the last chunk on this line is all whitespace, drop it.
if self.drop_whitespace and cur_line and not cur_line[-1].strip():
del cur_line[-1]
# Convert current line back to a string and store it in list
# of all lines (return value).
if cur_line:
lines.append(indent + ''.join(cur_line))
return lines
# }}}
# function wrap() {{{
def wrap(text, width=70, **kwargs):
"""
Wrap a single paragraph of text, returning a list of wrapped lines.
"""
if sys.version_info < (2, 6):
return TextWrapper(width=width, **kwargs).wrap(text)
return textwrap.wrap(text, width=width, **kwargs)
# }}} # }}}