1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
|
# -*- coding: utf-8 -*-
from __future__ import absolute_import
import codecs
import logging
import os
import chardet
import pysrt
from .score import get_equivalent_release_groups
from .video import Episode, Movie
from .utils import sanitize, sanitize_release_group
from six import text_type
logger = logging.getLogger(__name__)
#: Subtitle extensions
SUBTITLE_EXTENSIONS = ('.srt', '.sub', '.smi', '.txt', '.ssa', '.ass', '.mpl')
class Subtitle(object):
"""Base class for subtitle.
:param language: language of the subtitle.
:type language: :class:`~babelfish.language.Language`
:param bool hearing_impaired: whether or not the subtitle is hearing impaired.
:param page_link: URL of the web page from which the subtitle can be downloaded.
:type page_link: str
:param encoding: Text encoding of the subtitle.
:type encoding: str
"""
#: Name of the provider that returns that class of subtitle
provider_name = ''
def __init__(self, language, hearing_impaired=False, page_link=None, encoding=None):
#: Language of the subtitle
self.language = language
#: Whether or not the subtitle is hearing impaired
self.hearing_impaired = hearing_impaired
#: URL of the web page from which the subtitle can be downloaded
self.page_link = page_link
#: Content as bytes
self.content = None
#: Encoding to decode with when accessing :attr:`text`
self.encoding = None
self.release_info = None
# validate the encoding
if encoding:
try:
self.encoding = codecs.lookup(encoding).name
except (TypeError, LookupError):
logger.debug('Unsupported encoding %s', encoding)
@property
def id(self):
"""Unique identifier of the subtitle"""
raise NotImplementedError
@property
def text(self):
"""Content as string
If :attr:`encoding` is None, the encoding is guessed with :meth:`guess_encoding`
"""
if not self.content:
return
if not isinstance(self.content, text_type):
if self.encoding:
return self.content.decode(self.encoding, errors='replace')
return self.content.decode(self.guess_encoding(), errors='replace')
return self.content
def is_valid(self):
"""Check if a :attr:`text` is a valid SubRip format.
:return: whether or not the subtitle is valid.
:rtype: bool
"""
if not self.text:
return False
try:
pysrt.from_string(self.text, error_handling=pysrt.ERROR_RAISE)
except pysrt.Error as e:
if e.args[0] < 80:
return False
return True
def guess_encoding(self):
"""Guess encoding using the language, falling back on chardet.
:return: the guessed encoding.
:rtype: str
"""
logger.info('Guessing encoding for language %s', self.language)
# always try utf-8 first
encodings = ['utf-8']
# add language-specific encodings
if self.language.alpha3 == 'zho':
encodings.extend(['gb18030', 'big5'])
elif self.language.alpha3 == 'jpn':
encodings.append('shift-jis')
elif self.language.alpha3 == 'ara':
encodings.append('windows-1256')
elif self.language.alpha3 == 'heb':
encodings.append('windows-1255')
elif self.language.alpha3 == 'tur':
encodings.extend(['iso-8859-9', 'windows-1254'])
elif self.language.alpha3 == 'pol':
# Eastern European Group 1
encodings.extend(['windows-1250'])
elif self.language.alpha3 == 'bul':
# Eastern European Group 2
encodings.extend(['windows-1251'])
else:
# Western European (windows-1252)
encodings.append('latin-1')
# try to decode
logger.debug('Trying encodings %r', encodings)
for encoding in encodings:
try:
self.content.decode(encoding)
except UnicodeDecodeError:
pass
else:
logger.info('Guessed encoding %s', encoding)
return encoding
logger.warning('Could not guess encoding from language')
# fallback on chardet
encoding = chardet.detect(self.content)['encoding']
logger.info('Chardet found encoding %s', encoding)
return encoding
def get_matches(self, video):
"""Get the matches against the `video`.
:param video: the video to get the matches with.
:type video: :class:`~subliminal.video.Video`
:return: matches of the subtitle.
:rtype: set
"""
raise NotImplementedError
def __hash__(self):
return hash(self.provider_name + '-' + self.id)
def __repr__(self):
return '<%s %r [%s]>' % (self.__class__.__name__, self.id, self.language)
def get_subtitle_path(video_path, language=None, extension='.srt'):
"""Get the subtitle path using the `video_path` and `language`.
:param str video_path: path to the video.
:param language: language of the subtitle to put in the path.
:type language: :class:`~babelfish.language.Language`
:param str extension: extension of the subtitle.
:return: path of the subtitle.
:rtype: str
"""
subtitle_root = os.path.splitext(video_path)[0]
if language:
subtitle_root += '.' + str(language)
return subtitle_root + extension
def guess_matches(video, guess, partial=False):
"""Get matches between a `video` and a `guess`.
If a guess is `partial`, the absence information won't be counted as a match.
:param video: the video.
:type video: :class:`~subliminal.video.Video`
:param guess: the guess.
:type guess: dict
:param bool partial: whether or not the guess is partial.
:return: matches between the `video` and the `guess`.
:rtype: set
"""
matches = set()
if isinstance(video, Episode):
# series
if video.series and 'title' in guess and sanitize(guess['title']) == sanitize(video.series):
matches.add('series')
# title
if video.title and 'episode_title' in guess and sanitize(guess['episode_title']) == sanitize(video.title):
matches.add('title')
# season
if video.season and 'season' in guess and guess['season'] == video.season:
matches.add('season')
# episode
# Currently we only have single-ep support (guessit returns a multi-ep as a list with int values)
# Most providers only support single-ep, so make sure it contains only 1 episode
# In case of multi-ep, take the lowest episode (subtitles will normally be available on lowest episode number)
if video.episode and 'episode' in guess:
episode_guess = guess['episode']
episode = min(episode_guess) if episode_guess and isinstance(episode_guess, list) else episode_guess
if episode == video.episode:
matches.add('episode')
# year
if video.year and 'year' in guess and guess['year'] == video.year:
matches.add('year')
# count "no year" as an information
if not partial and video.original_series and 'year' not in guess:
matches.add('year')
elif isinstance(video, Movie):
# year
if video.year and 'year' in guess and guess['year'] == video.year:
matches.add('year')
# title
if video.title and 'title' in guess and sanitize(guess['title']) == sanitize(video.title):
matches.add('title')
# release_group
if (video.release_group and 'release_group' in guess and
sanitize_release_group(guess['release_group']) in
get_equivalent_release_groups(sanitize_release_group(video.release_group))):
matches.add('release_group')
# resolution
if video.resolution and 'screen_size' in guess and guess['screen_size'] == video.resolution:
matches.add('resolution')
# source
if video.source and 'source' in guess and guess['source'].lower() == video.source.lower():
matches.add('source')
# video_codec
if video.video_codec and 'video_codec' in guess and guess['video_codec'] == video.video_codec:
matches.add('video_codec')
# audio_codec
if video.audio_codec and 'audio_codec' in guess and guess['audio_codec'] == video.audio_codec:
matches.add('audio_codec')
return matches
def fix_line_ending(content):
"""Fix line ending of `content` by changing it to \n.
:param bytes content: content of the subtitle.
:return: the content with fixed line endings.
:rtype: bytes
"""
return content.replace(b'\r\n', b'\n')
|
