X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fsubtitles.py;h=41d740e871883c66097fe15ea6fd52326964dc90;hb=5f6a1245ffa9276c1af59b0835afeef67e2fb5b1;hp=8843e02209214fa351cde40f6cbfdebc0d063e91;hpb=8377574c9cb8740e24d45e9b3d30921fd6ec846c;p=youtube-dl diff --git a/youtube_dl/extractor/subtitles.py b/youtube_dl/extractor/subtitles.py index 8843e0220..41d740e87 100644 --- a/youtube_dl/extractor/subtitles.py +++ b/youtube_dl/extractor/subtitles.py @@ -1,59 +1,75 @@ -import socket - from .common import InfoExtractor from ..utils import ( - compat_http_client, - compat_urllib_error, - compat_urllib_request, compat_str, + ExtractorError, ) -class SubtitlesIE(InfoExtractor): +class SubtitlesInfoExtractor(InfoExtractor): + @property + def _have_to_download_any_subtitles(self): + return any([self._downloader.params.get('writesubtitles', False), + self._downloader.params.get('writeautomaticsub')]) - def report_video_subtitles_available(self, video_id, sub_lang_list): - """Report available subtitles.""" + def _list_available_subtitles(self, video_id, webpage): + """ outputs the available subtitles for the video """ + sub_lang_list = self._get_available_subtitles(video_id, webpage) + auto_captions_list = self._get_available_automatic_caption(video_id, webpage) sub_lang = ",".join(list(sub_lang_list.keys())) self.to_screen(u'%s: Available subtitles for video: %s' % (video_id, sub_lang)) + auto_lang = ",".join(auto_captions_list.keys()) + self.to_screen(u'%s: Available automatic captions for video: %s' % + (video_id, auto_lang)) - def _list_available_subtitles(self, video_id): - sub_lang_list = self._get_available_subtitles(video_id) - self.report_video_subtitles_available(video_id, sub_lang_list) - - def _extract_subtitles(self, video_id): + def extract_subtitles(self, video_id, webpage): """ - Return a dictionary: {language: subtitles} or {} if the subtitles - couldn't be found + returns {sub_lang: sub} ,{} if subtitles not found or None if the + subtitles aren't requested. """ - sub_lang_list = self._get_available_subtitles(video_id) - if not sub_lang_list: # error, it didn't get the available subtitles - return {} - + if not self._have_to_download_any_subtitles: + return None + available_subs_list = {} + if self._downloader.params.get('writeautomaticsub', False): + available_subs_list.update(self._get_available_automatic_caption(video_id, webpage)) if self._downloader.params.get('writesubtitles', False): - if self._downloader.params.get('subtitleslang', False): - sub_lang = self._downloader.params.get('subtitleslang') - elif 'en' in sub_lang_list: - sub_lang = 'en' + available_subs_list.update(self._get_available_subtitles(video_id, webpage)) + + if not available_subs_list: # error, it didn't get the available subtitles + return {} + if self._downloader.params.get('allsubtitles', False): + sub_lang_list = available_subs_list + else: + if self._downloader.params.get('subtitleslangs', False): + requested_langs = self._downloader.params.get('subtitleslangs') + elif 'en' in available_subs_list: + requested_langs = ['en'] else: - sub_lang = list(sub_lang_list.keys())[0] - if not sub_lang in sub_lang_list: - self._downloader.report_warning(u'no closed captions found in the specified language "%s"' % sub_lang) - return {} - sub_lang_list = {sub_lang: sub_lang_list[sub_lang]} + requested_langs = [list(available_subs_list.keys())[0]] + + sub_lang_list = {} + for sub_lang in requested_langs: + if sub_lang not in available_subs_list: + self._downloader.report_warning(u'no closed captions found in the specified language "%s"' % sub_lang) + continue + sub_lang_list[sub_lang] = available_subs_list[sub_lang] subtitles = {} - for sub_lang, url in sub_lang_list.iteritems(): + for sub_lang, url in sub_lang_list.items(): subtitle = self._request_subtitle_url(sub_lang, url) if subtitle: subtitles[sub_lang] = subtitle return subtitles + def _download_subtitle_url(self, sub_lang, url): + return self._download_webpage(url, None, note=False) + def _request_subtitle_url(self, sub_lang, url): + """ makes the http request for the subtitle """ try: - sub = compat_urllib_request.urlopen(url).read().decode('utf-8') - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: + sub = self._download_subtitle_url(sub_lang, url) + except ExtractorError as err: self._downloader.report_warning(u'unable to download video subtitles for %s: %s' % (sub_lang, compat_str(err))) return if not sub: @@ -61,13 +77,22 @@ class SubtitlesIE(InfoExtractor): return return sub - def _get_available_subtitles(self, video_id): - """returns the list of available subtitles like this {lang: url} """ - """or {} if not available. Must be redefined by the subclasses.""" - pass + def _get_available_subtitles(self, video_id, webpage): + """ + returns {sub_lang: url} or {} if not available + Must be redefined by the subclasses + """ + + # By default, allow implementations to simply pass in the result + assert isinstance(webpage, dict), \ + '_get_available_subtitles not implemented' + return webpage - def _request_automatic_caption(self, video_id, webpage): - """Request automatic caption. Redefine in subclasses.""" - """returns a tuple of ... """ - # return [(err_msg, None, None)] - pass + def _get_available_automatic_caption(self, video_id, webpage): + """ + returns {sub_lang: url} or {} if not available + Must be redefined by the subclasses that support automatic captions, + otherwise it will return {} + """ + self._downloader.report_warning(u'Automatic Captions not supported by this server') + return {}