def check_mime(self, ctx):
ctx.encoding = None
# MIME-Version:
mime_versions = ctx.metadata['MIME-Version']
if len(mime_versions) > 1:
self.tag('duplicate-header-field-mime-version')
mime_versions = sorted(set(mime_versions))
for mime_version in mime_versions:
if mime_version != '1.0':
self.tag('invalid-mime-version', mime_version, '=>', '1.0')
if len(mime_versions) == 0:
self.tag('no-mime-version-header-field',
tags.safestr('MIME-Version: 1.0'))
# Content-Transfer-Encoding:
ctes = ctx.metadata['Content-Transfer-Encoding']
if len(ctes) > 1:
self.tag('duplicate-header-field-content-transfer-encoding')
ctes = sorted(set(ctes))
for cte in ctes:
if cte != '8bit':
self.tag('invalid-content-transfer-encoding', cte, '=>',
'8bit')
if len(ctes) == 0:
self.tag('no-content-transfer-encoding-header-field',
tags.safestr('Content-Transfer-Encoding: 8bit'))
# Content-Type:
cts = ctx.metadata['Content-Type']
if len(cts) > 1:
self.tag('duplicate-header-field-content-type')
cts = sorted(set(cts))
elif len(cts) == 0:
content_type_hint = 'text/plain; charset=<encoding>'
self.tag('no-content-type-header-field',
tags.safestr('Content-Type: ' + content_type_hint))
return
encodings = set()
for ct in cts:
content_type_hint = 'text/plain; charset=<encoding>'
match = re.search(r'(\Atext/plain; )?\bcharset=([^\s;]+)\Z', ct)
if match:
encoding = match.group(2)
try:
is_ascii_compatible = encinfo.is_ascii_compatible_encoding(
encoding, missing_ok=False)
except encinfo.EncodingLookupError:
if encoding == 'CHARSET':
if not ctx.is_template:
self.tag('boilerplate-in-content-type', ct)
else:
self.tag('unknown-encoding', encoding)
encoding = None
else:
if not is_ascii_compatible:
self.tag('non-ascii-compatible-encoding', encoding)
elif encinfo.is_portable_encoding(encoding):
pass
else:
new_encoding = encinfo.propose_portable_encoding(
encoding)
if new_encoding is not None:
self.tag('non-portable-encoding', encoding, '=>',
new_encoding)
encoding = new_encoding
else:
self.tag('non-portable-encoding', encoding)
if ctx.language is not None:
unrepresentable_characters = ctx.language.get_unrepresentable_characters(
encoding)
if unrepresentable_characters:
if len(unrepresentable_characters) > 5:
unrepresentable_characters[4:] = ['...']
self.tag('unrepresentable-characters', encoding,
*unrepresentable_characters)
if match.group(1) is None:
if encoding is not None:
content_type_hint = content_type_hint.replace(
'<encoding>', encoding)
self.tag('invalid-content-type', ct, '=>',
content_type_hint)
if encoding is not None:
encodings.add(encoding)
else:
self.tag('invalid-content-type', ct, '=>', content_type_hint)
if len(encodings) == 1:
[ctx.encoding] = encodings
def t(encoding):
assert_false(E.is_ascii_compatible_encoding(encoding))
assert_false(
E.is_ascii_compatible_encoding(encoding, missing_ok=False))
def _test_missing(self, encoding):
assert_false(E.is_ascii_compatible_encoding(encoding))
with assert_raises(E.EncodingLookupError):
E.is_ascii_compatible_encoding(encoding, missing_ok=False)
def _parse_entry(self, i, msgid_offset, msgstr_offset):
view = self._view
[length, offset] = self._read_ints(at=msgid_offset, n=2)
msgid = view[offset:offset + length].tobytes()
try:
if view[offset + length] != b'\0':
raise SyntaxError('msgid is not null-terminated')
except IndexError:
raise SyntaxError('truncated file')
msgids = msgid.split(b'\0', 2)
msgid = msgids[0]
if len(msgids) > 2:
raise SyntaxError('unexpected null byte in msgid')
[length, offset] = self._read_ints(at=msgstr_offset, n=2)
msgstr = view[offset:offset + length].tobytes()
try:
if view[offset + length] != b'\0':
raise SyntaxError('msgstr is not null-terminated')
except IndexError:
raise SyntaxError('truncated file')
msgstrs = msgstr.split(b'\0')
if len(msgids) == 1 and len(msgstrs) > 1:
raise SyntaxError('unexpected null byte in msgstr')
encoding = self._encoding
if i == 0:
if encoding is None and msgid == b'':
# http://git.savannah.gnu.org/cgit/gettext.git/tree/gettext-runtime/intl/dcigettext.c?id=v0.18.3#n1106
match = re.search(b'charset=([^ \t\n]+)', msgstr)
if match is not None:
try:
encoding = match.group(1).decode('ASCII')
except UnicodeError:
pass
if encoding is None:
encoding = 'ASCII'
else:
if not encodings.is_ascii_compatible_encoding(encoding):
encoding = 'ASCII'
self._encoding = encoding
else:
if msgids == self._last_msgid:
raise SyntaxError('duplicate message definition')
elif msgid < self._last_msgid:
raise SyntaxError('messages are not sorted')
self._last_msgid = msgid # pylint: disable=attribute-defined-outside-init
assert encoding is not None
msgid, *msgctxt = msgid.split(b'\x04', 1)
kwargs = dict(msgid=msgid.decode(encoding))
if msgctxt:
[msgctxt] = msgctxt
kwargs.update(msgctxt=msgctxt.decode(encoding))
if len(msgids) == 1:
assert [msgstr] == msgstrs
kwargs.update(msgstr=msgstr.decode(encoding))
else:
assert len(msgids) == 2
assert len(msgstrs) >= 1
kwargs.update(msgid_plural=msgids[1].decode(encoding))
kwargs.update(msgstr_plural={
i: s.decode(encoding)
for i, s in enumerate(msgstrs)
})
entry = polib.MOEntry(**kwargs)
entry.comment = None
entry.occurrences = ()
entry.flags = () # https://bitbucket.org/izi/polib/issues/47
entry.translated = lambda: True
entry.previous_msgctxt = None
entry.previous_msgid = None
entry.previous_msgid_plural = None
return entry