def slugify(text, delim=u"_", encoding="ascii"):
""" Slugify given unicode text. """
text = normalize("NFKD", text)
period = u"."
name_and_extension = text.rsplit(period, 1)
name = name_and_extension[0]
name = six.text_type(delim).join(filter(None, (word for word in _punct_re.split(name.lower()))))
if len(name_and_extension) == 2:
extension = name_and_extension[1]
return six.text_type(period).join([name, extension])
else:
return name
def slugify(text, delim=u'_', encoding="ascii"):
""" Slugify given unicode text. """
text = normalize('NFKD', text)
period = u'.'
name_and_extension = text.rsplit(period, 1)
name = name_and_extension[0]
name = six.text_type(delim).join(
filter(None, (word for word in _punct_re.split(name.lower()))))
if len(name_and_extension) == 2:
extension = name_and_extension[1]
return six.text_type(period).join([name, extension])
else:
return name
def get_system_filename(sock, url, default="file"):
""" Get filename from given urllib2.urlopen object and URL.
First, attempts to extract Content-Disposition, second, extract
from URL, eventually fall back to default. Returns bytestring
in file system encoding. """
name = get_filename(sock, url)
if not name:
name = six.text_type(default)
return name.encode(sys.getfilesystemencoding(), 'ignore')
def get_filename(sock, url):
""" Get filename from given urllib2.urlopen object and URL.
First, tries Content-Disposition, if unavailable, extracts
name from URL. """
name = None
# NOTE: This gives bytes on 2 and unicode on 3.
# How does 3.x know the encoding?
cd = sock.headers.get('Content-Disposition', None)
if cd is not None:
try:
name = get_content_disposition(cd)
except IndexError:
pass
if not name:
parsed = urlparse(url)
name = parsed.path.rstrip('/').rsplit('/', 1)[-1]
return six.text_type(name)
def get_filename(sock, url):
""" Get filename from given urllib2.urlopen object and URL.
First, tries Content-Disposition, if unavailable, extracts
name from URL. """
name = None
# NOTE: This gives bytes on 2 and unicode on 3.
# How does 3.x know the encoding?
cd = sock.headers.get('Content-Disposition', None)
if cd is not None:
try:
name = get_content_disposition(cd)
except IndexError:
pass
if not name:
parsed = urlparse.urlparse(url)
name = parsed.path.rstrip('/').rsplit('/', 1)[-1]
return six.text_type(name)
