How to use guessit - 10 common examples

def __init__(self):
        Transformer.__init__(self, 15)

        self.properties = PropertiesContainer(enhance=False, canonical_from_pattern=False)

        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

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

        self.container.register_property(['episodeNumber', 'season'], '[0-9]{2,4}', confidence=0.6, formatter=_formater, disabler=lambda options: options.get('episode_prefer_number') if options else False)
        self.container.register_property(['episodeNumber', 'season'], '[0-9]{4}', confidence=0.6, formatter=_formater)
        self.container.register_property('episodeNumber', '[^0-9](\d{1,3})', confidence=0.6, formatter=parse_numeral, disabler=lambda options: not options.get('episode_prefer_number') if options else True)
        self.container.register_property(None, '(' + build_or_pattern(episode_words) + sep + '?(?P' + numeral + '))[^0-9]', confidence=0.4, formatter=parse_numeral)
        self.container.register_property(None, r'(?P' + numeral + ')' + sep + '?' + of_separators_re.pattern + sep + '?(?P' + numeral +')', confidence=0.6, formatter=parse_numeral)
        self.container.register_property('episodeNumber', r'^' + sep + '?(\d{1,3})' + sep, confidence=0.4, formatter=parse_numeral, disabler=lambda options: not options.get('episode_prefer_number') if options else True)
        self.container.register_property('episodeNumber', sep + r'(\d{1,3})' + sep + '?$', confidence=0.4, formatter=parse_numeral, disabler=lambda options: not options.get('episode_prefer_number') if options else True)

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

        self.container.register_property(['episodeNumber', 'season'], '[0-9]{2,4}', confidence=0.6, formatter=_formater, disabler=lambda options: options.get('episode_prefer_number') if options else False)
        self.container.register_property(['episodeNumber', 'season'], '[0-9]{4}', confidence=0.6, formatter=_formater)
        self.container.register_property('episodeNumber', '[^0-9](\d{1,3})', confidence=0.6, formatter=parse_numeral, disabler=lambda options: not options.get('episode_prefer_number') if options else True)
        self.container.register_property(None, '(' + build_or_pattern(episode_words) + sep + '?(?P' + numeral + '))[^0-9]', confidence=0.4, formatter=parse_numeral)
        self.container.register_property(None, r'(?P' + numeral + ')' + sep + '?' + of_separators_re.pattern + sep + '?(?P' + numeral +')', confidence=0.6, formatter=parse_numeral)
        self.container.register_property('episodeNumber', r'^' + sep + '?(\d{1,3})' + sep, confidence=0.4, formatter=parse_numeral, disabler=lambda options: not options.get('episode_prefer_number') if options else True)
        self.container.register_property('episodeNumber', sep + r'(\d{1,3})' + sep + '?$', confidence=0.4, formatter=parse_numeral, disabler=lambda options: not options.get('episode_prefer_number') if options else True)

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

# following heuristics confused (eg: OSS 117 will look like an episode,
        # season 1, epnum 17, when it is in fact a movie)
        fname = clean_string(filename).lower()
        for m in self.MOVIES:
            if m in fname:
                self.log.debug('Found in exception list of movies -> type = movie')
                upgrade_movie()
        for s in self.SERIES:
            if s in fname:
                self.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', 'info'):
            # if we have an episode_rexp (eg: s02e13), it is an episode
            episode_transformer = get_transformer('guess_episodes_rexps')
            if episode_transformer:
                for rexp, _, _ in episode_transformer.episode_rexps:
                    match = re.search(rexp, filename, re.IGNORECASE)
                    if match:
                        self.log.debug('Found matching regexp: "%s" (string = "%s") -> type = episode', rexp, match.group())
                        upgrade_episode()
                        break

            # if we have a 3-4 digit number that's not a year, maybe an episode
            match = re.search(r'[^0-9]([0-9]{3,4})[^0-9]', filename)
            if match:
                fullnumber = int(match.group()[1:-1])
                # season = fullnumber // 100
                epnumber = fullnumber % 100
                possible = True

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

        self.container.register_property(['episodeNumber', 'season'], '[0-9]{2,4}', confidence=0.6, formatter=_formater, disabler=lambda options: options.get('episode_prefer_number') if options else False)
        self.container.register_property(['episodeNumber', 'season'], '[0-9]{4}', confidence=0.6, formatter=_formater)
        self.container.register_property('episodeNumber', '[^0-9](\d{1,3})', confidence=0.6, formatter=parse_numeral, disabler=lambda options: not options.get('episode_prefer_number') if options else True)
        self.container.register_property(None, '(' + build_or_pattern(episode_words) + sep + '?(?P' + numeral + '))[^0-9]', confidence=0.4, formatter=parse_numeral)
        self.container.register_property(None, r'(?P' + numeral + ')' + sep + '?' + of_separators_re.pattern + sep + '?(?P' + numeral +')', confidence=0.6, formatter=parse_numeral)
        self.container.register_property('episodeNumber', r'^' + sep + '?(\d{1,3})' + sep, confidence=0.4, formatter=parse_numeral, disabler=lambda options: not options.get('episode_prefer_number') if options else True)
        self.container.register_property('episodeNumber', sep + r'(\d{1,3})' + sep + '?$', confidence=0.4, formatter=parse_numeral, disabler=lambda options: not options.get('episode_prefer_number') if options else True)

def __init__(self):
        Transformer.__init__(self, 15)

        of_separators = ['of', 'sur', '/', '\\']
        of_separators_re = re.compile(build_or_pattern(of_separators, escape=True), re.IGNORECASE)

        self.container = PropertiesContainer(enhance=False, canonical_from_pattern=False)

        episode_words = ['episodes?']

        def _formater(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

for rexp, span_adjust in weak_episode_rexps:
        match = re.search(rexp, string, re.IGNORECASE)
        if match:
            metadata = match.groupdict()
            span = (match.start() + span_adjust[0],
                    match.end() + span_adjust[1])

            epnum = int(metadata['episodeNumber'])
            if epnum > 100:
                season, epnum = epnum // 100, epnum % 100
                # episodes which have a season > 25 are most likely errors
                # (Simpsons is at 23!)
                if season > 25:
                    continue
                return Guess({ 'season': season,
                               'episodeNumber': epnum },
                             confidence=0.6, raw=string[span[0]:span[1]]), span
            else:
                return Guess(metadata, confidence=0.3, raw=string[span[0]:span[1]]), span

    return None, None

# search for titles
        titles = self.search_titles(title, season, year)

        subtitles = []
        # iterate over titles
        for title_id, t in titles.items():

            # iterate over title's archives
            for a in self.get_archives(title_id, language.legendastv):
                # clean name of path separators and pack flags
                clean_name = a.name.replace('/', '-')
                if a.pack and clean_name.startswith('(p)'):
                    clean_name = clean_name[3:]

                # guess from name
                guess = guessit(clean_name, {'type': t['type']})

                # episode
                if season and episode:
                    # discard mismatches on episode in non-pack archives
                    if not a.pack and 'episode' in guess and guess['episode'] != episode:
                        logger.debug('Mismatched episode %s, discarding archive: %s',
                                     guess['episode'], a.name)
                        continue

                # compute an expiration time based on the archive timestamp
                expiration_time = (datetime.utcnow().replace(tzinfo=pytz.utc) - a.timestamp).total_seconds()

                # attempt to get the releases from the cache
                cache_item = releases_key.format(archive_id=a.id, archive_name=a.name)
                releases = region.get(cache_item, expiration_time=expiration_time)

if video.episode and self.episode == video.episode:
                matches.add('episode')
            # guess
            matches |= guess_matches(video, guessit(self.version, {'type': 'episode', "single_value": True}))
            pass
        # movie
        elif isinstance(video, Movie):
            # title
            if video.title and (sanitize(self.title) in (
                    sanitize(name) for name in [video.title] + video.alternative_titles)):
                matches.add('title')
            # year
            if video.year and self.year == video.year:
                matches.add('year')
            # guess
            matches |= guess_matches(video, guessit(self.version, {'type': 'movie', "single_value": True}))

        self.matches = matches

        return matches