X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fcommon.py;h=8da70ae14409e4868c9ab72766c49c2b10454b9d;hb=78be2eca7cb2806c3a51547da14968336febb57c;hp=6ab2d68d6f3137ff7a9b4b201102959a7e732e9a;hpb=9c117d345ff05da2d4b107881cc387657208b76e;p=youtube-dl diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py index 6ab2d68d6..8da70ae14 100644 --- a/youtube_dl/extractor/common.py +++ b/youtube_dl/extractor/common.py @@ -18,7 +18,6 @@ from ..compat import ( compat_http_client, compat_urllib_error, compat_urllib_parse, - compat_urllib_parse_urlparse, compat_urlparse, compat_str, compat_etree_fromstring, @@ -30,10 +29,12 @@ from ..utils import ( clean_html, compiled_regex_type, determine_ext, + error_to_compat_str, ExtractorError, fix_xml_ampersands, float_or_none, int_or_none, + parse_iso8601, RegexNotFoundError, sanitize_filename, sanitized_Request, @@ -42,6 +43,7 @@ from ..utils import ( url_basename, xpath_text, xpath_with_ns, + determine_protocol, ) @@ -107,8 +109,9 @@ class InfoExtractor(object): -2 or smaller for less than default. < -1000 to hide the format (if there is another one which is strictly better) - * language_preference Is this in the correct requested - language? + * language Language code, e.g. "de" or "en-US". + * language_preference Is this in the language mentioned in + the URL? 10 if it's what the URL is about, -1 for default (don't know), -10 otherwise, other values reserved for now. @@ -199,6 +202,26 @@ class InfoExtractor(object): end_time: Time in seconds where the reproduction should end, as specified in the URL. + The following fields should only be used when the video belongs to some logical + chapter or section: + + chapter: Name or title of the chapter the video belongs to. + chapter_number: Number of the chapter the video belongs to, as an integer. + chapter_id: Id of the chapter the video belongs to, as a unicode string. + + The following fields should only be used when the video is an episode of some + series or programme: + + series: Title of the series or programme the video episode belongs to. + season: Title of the season the video episode belongs to. + season_number: Number of the season the video episode belongs to, as an integer. + season_id: Id of the season the video episode belongs to, as a unicode string. + episode: Title of the video episode. Unlike mandatory video title field, + this field should denote the exact title of the video episode + without any kind of decoration. + episode_number: Number of the video episode within a season, as an integer. + episode_id: Id of the video episode, as a unicode string. + Unless mentioned otherwise, the fields should be Unicode strings. Unless mentioned otherwise, None is equivalent to absence of information. @@ -291,9 +314,9 @@ class InfoExtractor(object): except ExtractorError: raise except compat_http_client.IncompleteRead as e: - raise ExtractorError('A network error has occured.', cause=e, expected=True) + raise ExtractorError('A network error has occurred.', cause=e, expected=True) except (KeyError, StopIteration) as e: - raise ExtractorError('An extractor error has occured.', cause=e) + raise ExtractorError('An extractor error has occurred.', cause=e) def set_downloader(self, downloader): """Sets the downloader for this IE.""" @@ -332,7 +355,8 @@ class InfoExtractor(object): return False if errnote is None: errnote = 'Unable to download webpage' - errmsg = '%s: %s' % (errnote, compat_str(err)) + + errmsg = '%s: %s' % (errnote, error_to_compat_str(err)) if fatal: raise ExtractorError(errmsg, sys.exc_info()[2], cause=err) else: @@ -622,7 +646,7 @@ class InfoExtractor(object): else: raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE) except (IOError, netrc.NetrcParseError) as err: - self._downloader.report_warning('parsing .netrc: %s' % compat_str(err)) + self._downloader.report_warning('parsing .netrc: %s' % error_to_compat_str(err)) return (username, password) @@ -739,6 +763,42 @@ class InfoExtractor(object): return self._html_search_meta('twitter:player', html, 'twitter card player') + def _search_json_ld(self, html, video_id, **kwargs): + json_ld = self._search_regex( + r'(?s)]+type=(["\'])application/ld\+json\1[^>]*>(?P.+?)', + html, 'JSON-LD', group='json_ld', **kwargs) + if not json_ld: + return {} + return self._json_ld(json_ld, video_id, fatal=kwargs.get('fatal', True)) + + def _json_ld(self, json_ld, video_id, fatal=True): + if isinstance(json_ld, compat_str): + json_ld = self._parse_json(json_ld, video_id, fatal=fatal) + if not json_ld: + return {} + info = {} + if json_ld.get('@context') == 'http://schema.org': + item_type = json_ld.get('@type') + if item_type == 'TVEpisode': + info.update({ + 'episode': unescapeHTML(json_ld.get('name')), + 'episode_number': int_or_none(json_ld.get('episodeNumber')), + 'description': unescapeHTML(json_ld.get('description')), + }) + part_of_season = json_ld.get('partOfSeason') + if isinstance(part_of_season, dict) and part_of_season.get('@type') == 'TVSeason': + info['season_number'] = int_or_none(part_of_season.get('seasonNumber')) + part_of_series = json_ld.get('partOfSeries') + if isinstance(part_of_series, dict) and part_of_series.get('@type') == 'TVSeries': + info['series'] = unescapeHTML(part_of_series.get('name')) + elif item_type == 'Article': + info.update({ + 'timestamp': parse_iso8601(json_ld.get('datePublished')), + 'title': unescapeHTML(json_ld.get('headline')), + 'description': unescapeHTML(json_ld.get('articleBody')), + }) + return dict((k, v) for k, v in info.items() if v is not None) + @staticmethod def _hidden_inputs(html): html = re.sub(r'', '', html) @@ -776,14 +836,12 @@ class InfoExtractor(object): preference = f.get('preference') if preference is None: - proto = f.get('protocol') - if proto is None: - proto = compat_urllib_parse_urlparse(f.get('url', '')).scheme - - preference = 0 if proto in ['http', 'https'] else -0.1 + preference = 0 if f.get('ext') in ['f4f', 'f4m']: # Not yet supported preference -= 0.5 + proto_preference = 0 if determine_protocol(f) in ['http', 'https'] else -0.1 + if f.get('vcodec') == 'none': # audio only if self._downloader.params.get('prefer_free_formats'): ORDER = ['aac', 'mp3', 'm4a', 'webm', 'ogg', 'opus'] @@ -814,6 +872,7 @@ class InfoExtractor(object): f.get('vbr') if f.get('vbr') is not None else -1, f.get('height') if f.get('height') is not None else -1, f.get('width') if f.get('width') is not None else -1, + proto_preference, ext_preference, f.get('abr') if f.get('abr') is not None else -1, audio_ext_preference, @@ -883,7 +942,7 @@ class InfoExtractor(object): fatal=fatal) if manifest is False: - return manifest + return [] formats = [] manifest_version = '1.0' @@ -909,10 +968,8 @@ class InfoExtractor(object): # may differ leading to inability to resolve the format by requested # bitrate in f4m downloader if determine_ext(manifest_url) == 'f4m': - f4m_formats = self._extract_f4m_formats( - manifest_url, video_id, preference, f4m_id, fatal=fatal) - if f4m_formats: - formats.extend(f4m_formats) + formats.extend(self._extract_f4m_formats( + manifest_url, video_id, preference, f4m_id, fatal=fatal)) continue tbr = int_or_none(media_el.attrib.get('bitrate')) formats.append({ @@ -954,7 +1011,7 @@ class InfoExtractor(object): errnote=errnote or 'Failed to download m3u8 information', fatal=fatal) if res is False: - return res + return [] m3u8_doc, urlh = res m3u8_url = urlh.geturl() last_info = None @@ -1146,10 +1203,8 @@ class InfoExtractor(object): src_url = src if src.startswith('http') else compat_urlparse.urljoin(base, src) if proto == 'm3u8' or src_ext == 'm3u8': - m3u8_formats = self._extract_m3u8_formats( - src_url, video_id, ext or 'mp4', m3u8_id='hls', fatal=False) - if m3u8_formats: - formats.extend(m3u8_formats) + formats.extend(self._extract_m3u8_formats( + src_url, video_id, ext or 'mp4', m3u8_id='hls', fatal=False)) continue if src_ext == 'f4m': @@ -1161,9 +1216,7 @@ class InfoExtractor(object): } f4m_url += '&' if '?' in f4m_url else '?' f4m_url += compat_urllib_parse.urlencode(f4m_params) - f4m_formats = self._extract_f4m_formats(f4m_url, video_id, f4m_id='hds', fatal=False) - if f4m_formats: - formats.extend(f4m_formats) + formats.extend(self._extract_f4m_formats(f4m_url, video_id, f4m_id='hds', fatal=False)) continue if src_url.startswith('http') and self._is_valid_url(src, video_id):