def load(class_, url, **args):
"""Loads the sample from a data URI."""
info_re = '(?P<mediatype>[^/]+/[^;]+(;[^=]+=[^;]+)*)?(?P<base64>;base64)?'
info, data = url.path.split(',')
info = re.match(info_re, info).groupdict()
mediatype = info['mediatype'].setdefault('text/plain;charset=US-ASCII')
if ';' in mediatype:
mimetype, params = mediatype.split(';', 1)
params = [p.strip().split('=') for p in params.split(';')]
params = dict((k.strip(), v.strip()) for k, v in params)
else:
mimetype, params = mediatype, dict()
data = base64.b64decode(data) if info['base64'] else urllib.unquote(data)
return class_(**content_types.get(mimetype).parse(data, **params))
def load(class_, url, **args):
"""Loads the sample from a file URL."""
if os.path.isdir(url.path):
files = {
'text': None,
'lexicon': None,
'word_delimiters': None,
'sentence_delimiters': None}
for filename in files:
with open(os.path.join(url.path, filename + '.txt'), 'rb') as txt:
files[filename] = txt.read().decode('UTF-8')
return class_(**files)
else:
with open(url.path, 'rb') as file_:
content = file_.read()
mimetype, encoding = mimetypes.guess_type(url.path)
content_encoding = args.get('content_encoding', encoding)
content_type = args.get('content_type', mimetype)
if content_encoding:
content = content_encodings.get(content_encoding).decode(content)
frozen = content_types.get(content_type).parse(content)
return class_(**frozen)
def load(class_, url, **args):
"""Loads the sample from a file URL."""
if os.path.isdir(url.path):
files = {
'text': None,
'lexicon': None,
'word_delimiters': None,
'sentence_delimiters': None
}
for filename in files:
with open(os.path.join(url.path, filename + '.txt'), 'rb') as txt:
files[filename] = txt.read().decode('UTF-8')
return class_(**files)
else:
with open(url.path, 'rb') as file_:
content = file_.read()
mimetype, encoding = mimetypes.guess_type(url.path)
content_encoding = args.get('content_encoding', encoding)
content_type = args.get('content_type', mimetype)
if content_encoding:
content = content_encodings.get(content_encoding).decode(content)
frozen = content_types.get(content_type).parse(content)
return class_(**frozen)
def dump(sample, url, **args):
"""Dunps the sample into a file URL."""
prefix, extension = os.path.splitext(url.path)
if not extension:
if not os.path.exists(url.path):
os.mkdir(url.path)
files = {
'text': sample['text'],
'lexicon': sample['lexicon'],
'word_delimiters': sample['word_delimiters'],
'sentence_delimiters': sample['sentence_delimiters']}
for filename in files:
with open(os.path.join(prefix, filename + '.txt'), 'wb') as file_:
file_.write(files.get(filename).encode('UTF-8'))
else:
mimetype, encoding = mimetypes.guess_type(url.path)
content_encoding = args.get('content_encoding', encoding)
content_type = args.get('content_type', mimetype)
content = content_types.get(content_type).format(sample.frozen())
if content_encoding:
content = content_encodings.get(content_encoding).encode(content)
with open(url.path, 'wb') as file_:
file_.write(content)
def dump(sample, url, **args):
"""Dunps the sample into a file URL."""
prefix, extension = os.path.splitext(url.path)
if not extension:
if not os.path.exists(url.path):
os.mkdir(url.path)
files = {
'text': sample['text'],
'lexicon': sample['lexicon'],
'word_delimiters': sample['word_delimiters'],
'sentence_delimiters': sample['sentence_delimiters']
}
for filename in files:
with open(os.path.join(prefix, filename + '.txt'), 'wb') as file_:
file_.write(files.get(filename).encode('UTF-8'))
else:
mimetype, encoding = mimetypes.guess_type(url.path)
content_encoding = args.get('content_encoding', encoding)
content_type = args.get('content_type', mimetype)
content = content_types.get(content_type).format(sample.frozen())
if content_encoding:
content = content_encodings.get(content_encoding).encode(content)
with open(url.path, 'wb') as file_:
file_.write(content)
