def list_parser(value, propertyListName, discrete_separators_re=discrete_separators_re, range_separators_re=range_separators_re, allow_discrete=False, fill_gaps=False):
discrete_elements = filter(lambda x: x != '', discrete_separators_re.split(value))
discrete_elements = [x.strip() for x in discrete_elements]
proper_discrete_elements = []
i = 0
while i < len(discrete_elements):
if i < len(discrete_elements) - 2 and range_separators_re.match(discrete_elements[i+1]):
proper_discrete_elements.append(discrete_elements[i] + discrete_elements[i+1] + discrete_elements[i+2])
i += 3
else:
match = range_separators_re.search(discrete_elements[i])
if match and match.start() == 0:
proper_discrete_elements[i-1] = proper_discrete_elements[i-1] + discrete_elements[i]
elif match and match.end() == len(discrete_elements[i]):
proper_discrete_elements.append(discrete_elements[i] + discrete_elements[i + 1])
else:
proper_discrete_elements.append(discrete_elements[i])
i += 1
discrete_elements = proper_discrete_elements
ret = []
for discrete_element in discrete_elements:
range_values = filter(lambda x: x != '', range_separators_re.split(discrete_element))
range_values = [x.strip() for x in range_values]
if len(range_values) > 1:
for x in range(0, len(range_values) - 1):
start_range_ep = parse_numeral(range_values[x])
end_range_ep = parse_numeral(range_values[x+1])
for range_ep in range(start_range_ep, end_range_ep + 1):
if range_ep not in ret:
ret.append(range_ep)
else:
discrete_value = parse_numeral(discrete_element)
if discrete_value not in ret:
ret.append(discrete_value)
if len(ret) > 1:
if not allow_discrete:
valid_ret = []
# replace discrete elements by ranges
valid_ret.append(ret[0])
for i in range(0, len(ret) - 1):
previous = valid_ret[len(valid_ret) - 1]
if ret[i+1] < previous:
pass
else:
valid_ret.append(ret[i+1])
ret = valid_ret
if fill_gaps:
ret = list(range(min(ret), max(ret) + 1))
if len(ret) > 1:
return {None: ret[0], propertyListName: ret}
if len(ret) > 0:
return ret[0]
return None
def list_parser(value, property_list_name, discrete_separators_re=discrete_separators_re, range_separators_re=range_separators_re, allow_discrete=False, fill_gaps=False):
discrete_elements = filter(lambda x: x != '', discrete_separators_re.split(value))
discrete_elements = [x.strip() for x in discrete_elements]
proper_discrete_elements = []
i = 0
while i < len(discrete_elements):
if i < len(discrete_elements) - 2 and range_separators_re.match(discrete_elements[i+1]):
proper_discrete_elements.append(discrete_elements[i] + discrete_elements[i+1] + discrete_elements[i+2])
i += 3
else:
match = range_separators_re.search(discrete_elements[i])
if match and match.start() == 0:
proper_discrete_elements[i-1] = proper_discrete_elements[i-1] + discrete_elements[i]
elif match and match.end() == len(discrete_elements[i]):
proper_discrete_elements.append(discrete_elements[i] + discrete_elements[i + 1])
else:
proper_discrete_elements.append(discrete_elements[i])
i += 1
discrete_elements = proper_discrete_elements
ret = []
for discrete_element in discrete_elements:
range_values = filter(lambda x: x != '', range_separators_re.split(discrete_element))
range_values = [x.strip() for x in range_values]
if len(range_values) > 1:
for x in range(0, len(range_values) - 1):
start_range_ep = parse_numeral(range_values[x])
end_range_ep = parse_numeral(range_values[x+1])
for range_ep in range(start_range_ep, end_range_ep + 1):
if range_ep not in ret:
ret.append(range_ep)
else:
discrete_value = parse_numeral(discrete_element)
if discrete_value not in ret:
ret.append(discrete_value)
if len(ret) > 1:
if not allow_discrete:
valid_ret = list()
# replace discrete elements by ranges
valid_ret.append(ret[0])
for i in range(0, len(ret) - 1):
previous = valid_ret[len(valid_ret) - 1]
if ret[i+1] < previous:
pass
else:
valid_ret.append(ret[i+1])
ret = valid_ret
if fill_gaps:
ret = list(range(min(ret), max(ret) + 1))
if len(ret) > 1:
return {None: ret[0], property_list_name: ret}
if len(ret) > 0:
return ret[0]
return None
def episode_parser(value):
values = re_split('[a-zA-Z]', value)
values = [x for x in values if x]
ret = []
for letters_elt in values:
dashed_values = letters_elt.split('-')
dashed_values = [x for x in dashed_values if x]
if len(dashed_values) > 1:
for _ in range(0, len(dashed_values) - 1):
start_dash_ep = parse_numeral(dashed_values[0])
end_dash_ep = parse_numeral(dashed_values[1])
for dash_ep in range(start_dash_ep, end_dash_ep + 1):
ret.append(dash_ep)
else:
ret.append(parse_numeral(letters_elt))
if len(ret) > 1:
return {None: ret[0], 'episodeList': ret} # TODO: Should support seasonList also
elif len(ret) > 0:
return ret[0]
else:
return None
def _formater(episodeNumber):
epnum = parse_numeral(episodeNumber)
if not valid_year(epnum):
if epnum > 100:
season, epnum = epnum // 100, epnum % 100
# episodes which have a season > 50 are most likely errors
# (Simpson is at 25!)
if season > 50:
return None
return {'season': season, 'episodeNumber': epnum}
else:
return epnum
def episode_parser(value):
values = re_split('[a-zA-Z]', value)
values = [x for x in values if x]
ret = []
for letters_elt in values:
dashed_values = letters_elt.split('-')
dashed_values = [x for x in dashed_values if x]
if len(dashed_values) > 1:
for _ in range(0, len(dashed_values) - 1):
start_dash_ep = parse_numeral(dashed_values[0])
end_dash_ep = parse_numeral(dashed_values[1])
for dash_ep in range(start_dash_ep, end_dash_ep + 1):
ret.append(dash_ep)
else:
ret.append(parse_numeral(letters_elt))
if len(ret) > 1:
return {None: ret[0], 'episodeList': ret} # TODO: Should support seasonList also
elif len(ret) > 0:
return ret[0]
else:
return None
def season_episode_parser(episode_number):
epnum = parse_numeral(episode_number)
if not valid_year(epnum):
if epnum > 100:
season, epnum = epnum // 100, epnum % 100
# episodes which have a season > 50 are most likely errors
# (Simpson is at 25!)
if season > 50:
return None
return {"season": season, "episodeNumber": epnum}
else:
return epnum
def format_guess(guess):
"""Format all the found values to their natural type.
For instance, a year would be stored as an int value, etc...
Note that this modifies the dictionary given as input.
"""
for prop, value in guess.items():
if prop in ('season', 'episodeNumber', 'year', 'cdNumber',
'cdNumberTotal', 'bonusNumber', 'filmNumber'):
guess[prop] = parse_numeral(guess[prop])
elif isinstance(value, base_text_type):
if prop in ('edition',):
value = clean_string(value)
guess[prop] = get_synonym(value).replace('\\', '')
return guess