X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fcrunchyroll.py;h=109d1c5a864f283a01b2b2baaed784384776a5c1;hb=b7a8c1bcfaccb797609ea763a1aa3140148bcddc;hp=c38fd095a3ea95facf214e6eb350fc71a20b39a6;hpb=dcdb292fddc82ae11f4c0b647815a45c88a6b6d5;p=youtube-dl diff --git a/youtube_dl/extractor/crunchyroll.py b/youtube_dl/extractor/crunchyroll.py index c38fd095a..109d1c5a8 100644 --- a/youtube_dl/extractor/crunchyroll.py +++ b/youtube_dl/extractor/crunchyroll.py @@ -142,7 +142,7 @@ class CrunchyrollIE(CrunchyrollBaseIE): 'ext': 'flv', 'title': 'Culture Japan Episode 1 – Rebuilding Japan after the 3.11', 'description': 'md5:2fbc01f90b87e8e9137296f37b461c12', - 'thumbnail': 're:^https?://.*\.jpg$', + 'thumbnail': r're:^https?://.*\.jpg$', 'uploader': 'Danny Choo Network', 'upload_date': '20120213', }, @@ -150,6 +150,7 @@ class CrunchyrollIE(CrunchyrollBaseIE): # rtmp 'skip_download': True, }, + 'skip': 'Video gone', }, { 'url': 'http://www.crunchyroll.com/rezero-starting-life-in-another-world-/episode-5-the-morning-of-our-promise-is-still-distant-702409', 'info_dict': { @@ -157,7 +158,7 @@ class CrunchyrollIE(CrunchyrollBaseIE): 'ext': 'mp4', 'title': 'Re:ZERO -Starting Life in Another World- Episode 5 – The Morning of Our Promise Is Still Distant', 'description': 'md5:97664de1ab24bbf77a9c01918cb7dca9', - 'thumbnail': 're:^https?://.*\.jpg$', + 'thumbnail': r're:^https?://.*\.jpg$', 'uploader': 'TV TOKYO', 'upload_date': '20160508', }, @@ -165,6 +166,25 @@ class CrunchyrollIE(CrunchyrollBaseIE): # m3u8 download 'skip_download': True, }, + }, { + 'url': 'http://www.crunchyroll.com/konosuba-gods-blessing-on-this-wonderful-world/episode-1-give-me-deliverance-from-this-judicial-injustice-727589', + 'info_dict': { + 'id': '727589', + 'ext': 'mp4', + 'title': "KONOSUBA -God's blessing on this wonderful world! 2 Episode 1 – Give Me Deliverance from this Judicial Injustice!", + 'description': 'md5:cbcf05e528124b0f3a0a419fc805ea7d', + 'thumbnail': r're:^https?://.*\.jpg$', + 'uploader': 'Kadokawa Pictures Inc.', + 'upload_date': '20170118', + 'series': "KONOSUBA -God's blessing on this wonderful world!", + 'season_number': 2, + 'episode': 'Give Me Deliverance from this Judicial Injustice!', + 'episode_number': 1, + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, }, { 'url': 'http://www.crunchyroll.fr/girl-friend-beta/episode-11-goodbye-la-mode-661697', 'only_matching': True, @@ -235,8 +255,7 @@ class CrunchyrollIE(CrunchyrollBaseIE): output += 'WrapStyle: %s\n' % sub_root.attrib['wrap_style'] output += 'PlayResX: %s\n' % sub_root.attrib['play_res_x'] output += 'PlayResY: %s\n' % sub_root.attrib['play_res_y'] - output += """ScaledBorderAndShadow: yes - + output += """ [V4+ Styles] Format: Name, Fontname, Fontsize, PrimaryColour, SecondaryColour, OutlineColour, BackColour, Bold, Italic, Underline, StrikeOut, ScaleX, ScaleY, Spacing, Angle, BorderStyle, Outline, Shadow, Alignment, MarginL, MarginR, MarginV, Encoding """ @@ -438,6 +457,18 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text subtitles = self.extract_subtitles(video_id, webpage) + # webpage provide more accurate data than series_title from XML + series = self._html_search_regex( + r'id=["\']showmedia_about_episode_num[^>]+>\s*]+>([^<]+)', + webpage, 'series', default=xpath_text(metadata, 'series_title')) + + episode = xpath_text(metadata, 'episode_title') + episode_number = int_or_none(xpath_text(metadata, 'episode_number')) + + season_number = int_or_none(self._search_regex( + r'(?s)]+id=["\']showmedia_about_episode_num[^>]+>.+?\s*

\s*Season (\d+)', + webpage, 'season number', default=None)) + return { 'id': video_id, 'title': video_title, @@ -445,9 +476,10 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text 'thumbnail': xpath_text(metadata, 'episode_image_url'), 'uploader': video_uploader, 'upload_date': video_upload_date, - 'series': xpath_text(metadata, 'series_title'), - 'episode': xpath_text(metadata, 'episode_title'), - 'episode_number': int_or_none(xpath_text(metadata, 'episode_number')), + 'series': series, + 'season_number': season_number, + 'episode': episode, + 'episode_number': episode_number, 'subtitles': subtitles, 'formats': formats, }