diff --git a/youtube_dlc/extractor/bandcamp.py b/youtube_dlc/extractor/bandcamp.py index 9dbafe86d..8a37e1b94 100644 --- a/youtube_dlc/extractor/bandcamp.py +++ b/youtube_dlc/extractor/bandcamp.py @@ -25,7 +25,45 @@ ) -class BandcampIE(InfoExtractor): +class BandcampBaseIE(InfoExtractor): + """Provide base functions for Bandcamp extractors""" + + def _extract_json_from_html_data_attribute(self, webpage, suffix, video_id): + json_string = self._html_search_regex( + r' data-%s="([^"]*)' % suffix, + webpage, '%s json' % suffix, default='{}') + + return self._parse_json(json_string, video_id) + + def _parse_json_track(self, json): + formats = [] + file_ = json.get('file') + if isinstance(file_, dict): + for format_id, format_url in file_.items(): + if not url_or_none(format_url): + continue + ext, abr_str = format_id.split('-', 1) + formats.append({ + 'format_id': format_id, + 'url': self._proto_relative_url(format_url, 'http:'), + 'ext': ext, + 'vcodec': 'none', + 'acodec': ext, + 'abr': int_or_none(abr_str), + }) + + return { + 'duration': float_or_none(json.get('duration')), + 'id': str_or_none(json.get('track_id') or json.get('id')), + 'title': json.get('title'), + 'title_link': json.get('title_link'), + 'number': int_or_none(json.get('track_num')), + 'formats': formats + } + + +class BandcampIE(BandcampBaseIE): + IE_NAME = "Bandcamp:track" _VALID_URL = r'https?://[^/]+\.bandcamp\.com/track/(?P[^/?#&]+)' _TESTS = [{ 'url': 'http://youtube-dlc.bandcamp.com/track/youtube-dlc-test-song', @@ -85,52 +123,32 @@ class BandcampIE(InfoExtractor): def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) title = mobj.group('title') + url_track_title = title webpage = self._download_webpage(url, title) thumbnail = self._html_search_meta('og:image', webpage, default=None) - track_id = None - track = None - track_number = None - duration = None + json_tralbum = self._extract_json_from_html_data_attribute(webpage, "tralbum", url_track_title) + json_embed = self._extract_json_from_html_data_attribute(webpage, "embed", url_track_title) - formats = [] - trackinfo_block = self._html_search_regex( - r'trackinfo(?:["\']|"):\[\s*({.+?})\s*\],(?:["\']|")', - webpage, 'track info', default='{}') + json_tracks = json_tralbum.get('trackinfo') + if not json_tracks: + raise ExtractorError('Could not extract track') - track_info = self._parse_json(trackinfo_block, title) - if track_info: - file_ = track_info.get('file') - if isinstance(file_, dict): - for format_id, format_url in file_.items(): - if not url_or_none(format_url): - continue - ext, abr_str = format_id.split('-', 1) - formats.append({ - 'format_id': format_id, - 'url': self._proto_relative_url(format_url, 'http:'), - 'ext': ext, - 'vcodec': 'none', - 'acodec': ext, - 'abr': int_or_none(abr_str), - }) + track = self._parse_json_track(json_tracks[0]) + artist = json_tralbum.get('artist') + album_title = json_embed.get('album_title') - track_id = str_or_none(track_info.get('track_id') or track_info.get('id')) - track_number = int_or_none(track_info.get('track_num')) - duration = float_or_none(track_info.get('duration')) + json_album = json_tralbum.get('packages') + if json_album: + json_album = json_album[0] + album_publish_date = json_album.get('album_publish_date') + album_release_date = json_album.get('album_release_date') + else: + album_publish_date = None + album_release_date = json_tralbum.get('album_release_date') - def extract(key): - data = self._html_search_regex( - r',(["\']|")%s\1:\1(?P<value>(?:\\\1|((?!\1).))+)\1' % key, - webpage, key, default=None, group='value') - return data.replace(r'\"', '"').replace('\\\\', '\\') if data else data - - track = extract('title') - artist = extract('artist') - album = extract('album_title') - timestamp = unified_timestamp( - extract('publish_date') or extract('album_publish_date')) - release_date = unified_strdate(extract('album_release_date')) + timestamp = unified_timestamp(json_tralbum.get('current', {}).get('publish_date') or album_publish_date) + release_date = unified_strdate(album_release_date) download_link = self._search_regex( r'freeDownloadPage(?:["\']|"):\s*(["\']|")(?P<url>(?:(?!\1).)+)\1', webpage, @@ -155,8 +173,6 @@ def extract(key): if info: downloads = info.get('downloads') if isinstance(downloads, dict): - if not track: - track = info.get('title') if not artist: artist = info.get('artist') if not thumbnail: @@ -190,7 +206,7 @@ def extract(key): retry_url = url_or_none(stat.get('retry_url')) if not retry_url: continue - formats.append({ + track['formats'].append({ 'url': self._proto_relative_url(retry_url, 'http:'), 'ext': download_formats.get(format_id), 'format_id': format_id, @@ -199,32 +215,37 @@ def extract(key): 'vcodec': 'none', }) - self._sort_formats(formats) + self._sort_formats(track['formats']) - title = '%s - %s' % (artist, track) if artist else track - - if not duration: - duration = float_or_none(self._html_search_meta( - 'duration', webpage, default=None)) + title = '%s - %s' % (artist, track.get('title')) if artist else track.get('title') return { - 'id': track_id, - 'title': title, - 'thumbnail': thumbnail, - 'uploader': artist, - 'timestamp': timestamp, - 'release_date': release_date, - 'duration': duration, - 'track': track, - 'track_number': track_number, - 'track_id': track_id, + 'album': album_title, 'artist': artist, - 'album': album, - 'formats': formats, + 'duration': track['duration'], + 'formats': track['formats'], + 'id': track['id'], + 'release_date': release_date, + 'thumbnail': thumbnail, + 'timestamp': timestamp, + 'title': title, + 'track': track['title'], + 'track_id': track['id'], + 'track_number': track['number'], + 'uploader': artist } -class BandcampAlbumIE(InfoExtractor): +class BandcampAlbumTrackIE(BandcampIE): + IE_NAME = "Bandcamp:albumtrack" + """Hack class to force album downloads to have prefixed track numbers by default""" + def _real_extract(self, url): + data = super()._real_extract(url) + data['title'] = '{:02d} - {} - {}'.format(data['track_number'], data['artist'], data['track']) + return data + + +class BandcampAlbumIE(BandcampBaseIE): IE_NAME = 'Bandcamp:album' _VALID_URL = r'https?://(?:(?P<subdomain>[^.]+)\.)?bandcamp\.com(?:/album/(?P<album_id>[^/?#&]+))?' @@ -305,34 +326,32 @@ def _real_extract(self, url): album_id = mobj.group('album_id') playlist_id = album_id or uploader_id webpage = self._download_webpage(url, playlist_id) - track_elements = re.findall( - r'(?s)<div[^>]*>(.*?<a[^>]+href="([^"]+?)"[^>]+itemprop="url"[^>]*>.*?)</div>', webpage) - if not track_elements: - raise ExtractorError('The page doesn\'t contain any tracks') + + json_tralbum = self._extract_json_from_html_data_attribute(webpage, "tralbum", playlist_id) + json_embed = self._extract_json_from_html_data_attribute(webpage, "embed", playlist_id) + + json_tracks = json_tralbum.get('trackinfo') + if not json_tracks: + raise ExtractorError('Could not extract album tracks') + + album_title = json_embed.get('album_title') + # Only tracks with duration info have songs + tracks = [self._parse_json_track(track) for track in json_tracks] entries = [ self.url_result( - compat_urlparse.urljoin(url, t_path), - ie=BandcampIE.ie_key(), - video_title=self._search_regex( - r'<span\b[^>]+\bitemprop=["\']name["\'][^>]*>([^<]+)', - elem_content, 'track title', fatal=False)) - for elem_content, t_path in track_elements - if self._html_search_meta('duration', elem_content, default=None)] - - title = self._html_search_regex( - r'album_title\s*(?:"|["\']):\s*("|["\'])(?P<album>(?:\\\1|((?!\1).))+)\1', - webpage, 'title', fatal=False, group='album') - - if title: - title = title.replace(r'\"', '"') + compat_urlparse.urljoin(url, track['title_link']), + ie=BandcampAlbumTrackIE.ie_key(), + video_title=track['title']) + for track in tracks + if track.get('duration')] return { '_type': 'playlist', 'uploader_id': uploader_id, 'id': playlist_id, - 'title': title, - 'entries': entries, + 'title': album_title, + 'entries': entries } diff --git a/youtube_dlc/extractor/extractors.py b/youtube_dlc/extractor/extractors.py index d31edd7c8..fbd4ed1e3 100644 --- a/youtube_dlc/extractor/extractors.py +++ b/youtube_dlc/extractor/extractors.py @@ -84,7 +84,7 @@ ) from .azmedien import AZMedienIE from .baidu import BaiduVideoIE -from .bandcamp import BandcampIE, BandcampAlbumIE, BandcampWeeklyIE +from .bandcamp import BandcampIE, BandcampAlbumTrackIE, BandcampAlbumIE, BandcampWeeklyIE from .bbc import ( BBCCoUkIE, BBCCoUkArticleIE,