def process(self, lst):
"""Main processing method"""
result = []
for s in lst: # pylint: disable=invalid-name
translated = translate_string(s, keep_chars="'-").replace("'", ' ')
result += [
token
for token in nltk.word_tokenize(translated, self.language)
if token and len(token) > 1
]
return result
def FileView(request): # pylint: disable=invalid-name
"""Default file view"""
context = request.context
# set content type
content_type = context.content_type
if isinstance(content_type, bytes):
content_type = content_type.decode('utf-8')
# check for last modification date
response = Response(content_type=content_type)
zdc = IZopeDublinCore(context, None)
if zdc is not None:
modified = zdc.modified
if modified is not None:
if_modified_since = request.if_modified_since
# pylint: disable=no-member
if if_modified_since and \
(int(modified.timestamp()) <= int(if_modified_since.timestamp())):
return Response(content_type=content_type, status=NOT_MODIFIED)
response.last_modified = modified
body_file = context.get_blob(mode='c')
if request.params.get('dl') is not None:
filename = context.filename or 'noname.txt'
response.content_disposition = 'attachment; filename="{0}"'.format(
translate_string(filename, force_lower=False))
# check for range request
if request.range is not None:
try:
body = body_file.read()
body_length = len(body)
range_start = request.range.start or 0
if 'Firefox' in request.user_agent: # avoid partial range for Firefox videos
range_end = body_length
else:
range_end = request.range.end or min(
body_length, range_start + MAX_RANGE_LENGTH)
ranged_body = body[range_start:range_end]
response.status = PARTIAL_CONTENT
response.headers[
'Content-Range'] = 'bytes {first}-{last}/{len}'.format(
first=range_start,
last=range_start + len(ranged_body) - 1,
len=body_length)
response.body = ranged_body
finally:
body_file.close()
else:
response.body_file = body_file
return response
def process(self, lst):
"""Main process method"""
result = []
for s in lst: # pylint: disable=invalid-name
translated = translate_string(s, keep_chars="'-").replace("'", ' ')
tokens = nltk.word_tokenize(translated, self.language)
result += [
stem for stem in [
self.stemmer.stem(token) for token in tokens
if token not in self.stemmer.stopwords
] if stem and (len(stem) > 1) and (
stem not in self.stemmer.stopwords)
]
return result
def generate_url(title, min_word_length=2):
"""Generate an SEO-friendly content URL from it's title
The original title is translated to remove accents, converted to lowercase, and words
shorter than three characters (by default) are removed; terms are joined by hyphens.
:param title: the input text
:param min_word_length: minimum length of words to keep
>>> from pyams_utils.url import generate_url
>>> generate_url('This is my test')
'this-is-my-test'
Single letters are removed from generated URLs:
>>> generate_url('This word has a single a')
'this-word-has-single'
But you can define the minimum length of word:
>>> generate_url('This word has a single a', min_word_length=4)
'this-word-single'
If input text contains slashes, they are replaced with hyphens:
>>> generate_url('This string contains/slash')
'this-string-contains-slash'
Punctation and special characters are completely removed:
>>> generate_url('This is a string with a point. And why not?')
'this-is-string-with-point-and-why-not'
"""
return '-'.join(
filter(
lambda x: len(x) >= min_word_length,
translate_string(title.replace('/', '-'),
escape_slashes=False,
force_lower=True,
spaces='-',
remove_punctuation=True,
keep_chars='-').split('-')))