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 test_found_nonpython(self):
assert_false(E.is_portable_encoding('KOI8-T'))
assert_true(E.is_portable_encoding('KOI8-T', python=False))
def test_notfound(self):
assert_false(E.is_portable_encoding('ISO-8859-16'))
assert_false(E.is_portable_encoding('ISO-8859-16', python=False))
def test_found_(self):
assert_true(E.is_portable_encoding('ISO_8859-2'))