dinislage/CouchPotatoServer
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Update libs

Browse Source
This commit is contained in:
Ruud
2013-11-22 16:09:15 +01:00
parent f53364eb6c
commit 88d6148500
119 changed files with 18448 additions and 8923 deletions
Download Patch File Download Diff File Expand all files Collapse all files
libs/guessit/transfo/__init__.py
+15 -7
View File
@@ -28,7 +28,7 @@ log = logging.getLogger(__name__)
def found_property(node, name, confidence):
node.guess = Guess({name: node.clean_value}, confidence=confidence)
node.guess = Guess({name: node.clean_value}, confidence=confidence, raw=node.value)
log.debug('Found with confidence %.2f: %s' % (confidence, node.guess))
@@ -52,11 +52,17 @@ def format_guess(guess):
def find_and_split_node(node, strategy, logger):
string = ' %s ' % node.value # add sentinels
for matcher, confidence in strategy:
for matcher, confidence, args, kwargs in strategy:
all_args = [string]
if getattr(matcher, 'use_node', False):
result, span = matcher(string, node)
all_args.append(node)
if args:
all_args.append(args)
if kwargs:
result, span = matcher(*all_args, **kwargs)
else:
result, span = matcher(string)
result, span = matcher(*all_args)
if result:
# readjust span to compensate for sentinels
@@ -69,7 +75,7 @@ def find_and_split_node(node, strategy, logger):
if confidence is None:
confidence = 1.0
guess = format_guess(Guess(result, confidence=confidence))
guess = format_guess(Guess(result, confidence=confidence, raw=string[span[0] + 1:span[1] + 1]))
msg = 'Found with confidence %.2f: %s' % (confidence, guess)
(logger or log).debug(msg)
@@ -84,10 +90,12 @@ def find_and_split_node(node, strategy, logger):
class SingleNodeGuesser(object):
def __init__(self, guess_func, confidence, logger=None):
def __init__(self, guess_func, confidence, logger, *args, **kwargs):
self.guess_func = guess_func
self.confidence = confidence
self.logger = logger
self.args = args
self.kwargs = kwargs
def process(self, mtree):
# strategy is a list of pairs (guesser, confidence)
@@ -95,7 +103,7 @@ class SingleNodeGuesser(object):
# it will override it, otherwise it will leave the guess confidence
# - if the guesser returns a simple dict as a guess and confidence is
# specified, it will use it, or 1.0 otherwise
strategy = [ (self.guess_func, self.confidence) ]
strategy = [ (self.guess_func, self.confidence, self.args, self.kwargs) ]
for node in mtree.unidentified_leaves():
find_and_split_node(node, strategy, self.logger)
libs/guessit/transfo/guess_country.py
+1 -1
View File
@@ -45,4 +45,4 @@ def process(mtree):
except ValueError:
continue
node.guess = Guess(country=country, confidence=1.0)
node.guess = Guess(country=country, confidence=1.0, raw=c)
libs/guessit/transfo/guess_episodes_rexps.py
+5 -10
View File
@@ -40,27 +40,22 @@ def guess_episodes_rexps(string):
for rexp, confidence, span_adjust in episode_rexps:
match = re.search(rexp, string, re.IGNORECASE)
if match:
guess = Guess(match.groupdict(), confidence=confidence)
span = (match.start() + span_adjust[0],
span = (match.start() + span_adjust[0],
match.end() + span_adjust[1])
# episodes which have a season > 30 are most likely errors
# (Simpsons is at 24!)
if int(guess.get('season', 0)) > 30:
continue
guess = Guess(match.groupdict(), confidence=confidence, raw=string[span[0]:span[1]])
# decide whether we have only a single episode number or an
# episode list
if guess.get('episodeNumber'):
eplist = number_list(guess['episodeNumber'])
guess.set('episodeNumber', eplist[0], confidence=confidence)
guess.set('episodeNumber', eplist[0], confidence=confidence, raw=string[span[0]:span[1]])
if len(eplist) > 1:
guess.set('episodeList', eplist, confidence=confidence)
guess.set('episodeList', eplist, confidence=confidence, raw=string[span[0]:span[1]])
if guess.get('bonusNumber'):
eplist = number_list(guess['bonusNumber'])
guess.set('bonusNumber', eplist[0], confidence=confidence)
guess.set('bonusNumber', eplist[0], confidence=confidence, raw=string[span[0]:span[1]])
return guess, span
libs/guessit/transfo/guess_filetype.py
+25 -2
View File
@@ -20,7 +20,7 @@
from __future__ import unicode_literals
from guessit import Guess
from guessit.patterns import (subtitle_exts, video_exts, episode_rexps,
from guessit.patterns import (subtitle_exts, info_exts, video_exts, episode_rexps,
find_properties, compute_canonical_form)
from guessit.date import valid_year
from guessit.textutils import clean_string
@@ -53,12 +53,16 @@ def guess_filetype(mtree, filetype):
filetype_container[0] = 'episode'
elif filetype_container[0] == 'subtitle':
filetype_container[0] = 'episodesubtitle'
elif filetype_container[0] == 'info':
filetype_container[0] = 'episodeinfo'
def upgrade_movie():
if filetype_container[0] == 'video':
filetype_container[0] = 'movie'
elif filetype_container[0] == 'subtitle':
filetype_container[0] = 'moviesubtitle'
elif filetype_container[0] == 'info':
filetype_container[0] = 'movieinfo'
def upgrade_subtitle():
if 'movie' in filetype_container[0]:
@@ -68,6 +72,14 @@ def guess_filetype(mtree, filetype):
else:
filetype_container[0] = 'subtitle'
def upgrade_info():
if 'movie' in filetype_container[0]:
filetype_container[0] = 'movieinfo'
elif 'episode' in filetype_container[0]:
filetype_container[0] = 'episodeinfo'
else:
filetype_container[0] = 'info'
def upgrade(type='unknown'):
if filetype_container[0] == 'autodetect':
filetype_container[0] = type
@@ -78,6 +90,9 @@ def guess_filetype(mtree, filetype):
if fileext in subtitle_exts:
upgrade_subtitle()
other = { 'container': fileext }
elif fileext in info_exts:
upgrade_info()
other = { 'container': fileext }
elif fileext in video_exts:
upgrade(type='video')
other = { 'container': fileext }
@@ -104,17 +119,20 @@ def guess_filetype(mtree, filetype):
fname = clean_string(filename).lower()
for m in MOVIES:
if m in fname:
log.debug('Found in exception list of movies -> type = movie')
upgrade_movie()
for s in SERIES:
if s in fname:
log.debug('Found in exception list of series -> type = episode')
upgrade_episode()
# now look whether there are some specific hints for episode vs movie
if filetype_container[0] in ('video', 'subtitle'):
if filetype_container[0] in ('video', 'subtitle', 'info'):
# if we have an episode_rexp (eg: s02e13), it is an episode
for rexp, _, _ in episode_rexps:
match = re.search(rexp, filename, re.IGNORECASE)
if match:
log.debug('Found matching regexp: "%s" (string = "%s") -> type = episode', rexp, match.group())
upgrade_episode()
break
@@ -133,24 +151,29 @@ def guess_filetype(mtree, filetype):
possible = False
if possible:
log.debug('Found possible episode number: %s (from string "%s") -> type = episode', epnumber, match.group())
upgrade_episode()
# if we have certain properties characteristic of episodes, it is an ep
for prop, value, _, _ in find_properties(filename):
log.debug('prop: %s = %s' % (prop, value))
if prop == 'episodeFormat':
log.debug('Found characteristic property of episodes: %s = "%s"', prop, value)
upgrade_episode()
break
elif compute_canonical_form('format', value) == 'DVB':
log.debug('Found characteristic property of episodes: %s = "%s"', prop, value)
upgrade_episode()
break
# origin-specific type
if 'tvu.org.ru' in filename:
log.debug('Found characteristic property of episodes: %s = "%s"', prop, value)
upgrade_episode()
# if no episode info found, assume it's a movie
log.debug('Nothing characteristic found, assuming type = movie')
upgrade_movie()
filetype = filetype_container[0]
libs/guessit/transfo/guess_language.py
+18 -6
View File
@@ -22,22 +22,34 @@ from __future__ import unicode_literals
from guessit import Guess
from guessit.transfo import SingleNodeGuesser
from guessit.language import search_language
from guessit.textutils import clean_string, find_words
import logging
log = logging.getLogger(__name__)
def guess_language(string):
language, span, confidence = search_language(string)
def guess_language(string, node, skip=None):
if skip:
relative_skip = []
for entry in skip:
node_idx = entry['node_idx']
span = entry['span']
if node_idx == node.node_idx[:len(node_idx)]:
relative_span = (span[0] - node.offset + 1, span[1] - node.offset + 1)
relative_skip.append(relative_span)
skip = relative_skip
language, span, confidence = search_language(string, skip=skip)
if language:
return (Guess({'language': language},
confidence=confidence),
confidence=confidence,
raw= string[span[0]:span[1]]),
span)
return None, None
guess_language.use_node = True
def process(mtree):
SingleNodeGuesser(guess_language, None, log).process(mtree)
def process(mtree, *args, **kwargs):
SingleNodeGuesser(guess_language, None, log, *args, **kwargs).process(mtree)
# Note: 'language' is promoted to 'subtitleLanguage' in the post_process transfo
libs/guessit/transfo/guess_movie_title_from_position.py
+2 -1
View File
@@ -29,7 +29,8 @@ log = logging.getLogger(__name__)
def process(mtree):
def found_property(node, name, value, confidence):
node.guess = Guess({ name: value },
confidence=confidence)
confidence=confidence,
raw=value)
log.debug('Found with confidence %.2f: %s' % (confidence, node.guess))
def found_title(node, confidence):
libs/guessit/transfo/guess_video_rexps.py
+4 -3
View File
@@ -38,9 +38,10 @@ def guess_video_rexps(string):
# the soonest that we can catch it)
if metadata.get('cdNumberTotal', -1) is None:
del metadata['cdNumberTotal']
return (Guess(metadata, confidence=confidence),
(match.start() + span_adjust[0],
match.end() + span_adjust[1] - 2))
span = (match.start() + span_adjust[0],
match.end() + span_adjust[1] - 2)
return (Guess(metadata, confidence=confidence, raw=string[span[0]:span[1]]),
span)
return None, None
libs/guessit/transfo/guess_weak_episodes_rexps.py
+2 -2
View File
@@ -48,9 +48,9 @@ def guess_weak_episodes_rexps(string, node):
continue
return Guess({ 'season': season,
'episodeNumber': epnum },
confidence=0.6), span
confidence=0.6, raw=string[span[0]:span[1]]), span
else:
return Guess(metadata, confidence=0.3), span
return Guess(metadata, confidence=0.3, raw=string[span[0]:span[1]]), span
return None, None