From 3cbecdd11121b9c7ff0284e481992f7230806399 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Thu, 26 Jan 2017 23:25:40 +0700 Subject: [PATCH] [jamendo] Improve and extract more metadata (closes #11836) --- youtube_dl/extractor/jamendo.py | 65 ++++++++++++++++++++++++--------- 1 file changed, 48 insertions(+), 17 deletions(-) diff --git a/youtube_dl/extractor/jamendo.py b/youtube_dl/extractor/jamendo.py index 3db07e79f..595d7a5b7 100644 --- a/youtube_dl/extractor/jamendo.py +++ b/youtube_dl/extractor/jamendo.py @@ -5,9 +5,27 @@ import re from ..compat import compat_urlparse from .common import InfoExtractor +from ..utils import parse_duration -class JamendoIE(InfoExtractor): +class JamendoBaseIE(InfoExtractor): + def _extract_meta(self, webpage, fatal=True): + title = self._og_search_title( + webpage, default=None) or self._search_regex( + r'([^<]+)', webpage, + 'title', default=None) + if title: + title = self._search_regex( + r'(.+?)\s*\|\s*Jamendo Music', title, 'title', default=None) + if not title: + title = self._html_search_meta( + 'name', webpage, 'title', fatal=fatal) + mobj = re.search(r'(.+) - (.+)', title or '') + artist, second = mobj.groups() if mobj else [None] * 2 + return title, artist, second + + +class JamendoIE(JamendoBaseIE): _VALID_URL = r'https?://(?:www\.)?jamendo\.com/track/(?P<id>[0-9]+)/(?P<display_id>[^/?#&]+)' _TEST = { 'url': 'https://www.jamendo.com/track/196219/stories-from-emona-i', @@ -17,6 +35,9 @@ class JamendoIE(InfoExtractor): 'display_id': 'stories-from-emona-i', 'ext': 'flac', 'title': 'Maya Filipič - Stories from Emona I', + 'artist': 'Maya Filipič', + 'track': 'Stories from Emona I', + 'duration': 210, 'thumbnail': r're:^https?://.*\.jpg' } } @@ -28,7 +49,7 @@ class JamendoIE(InfoExtractor): webpage = self._download_webpage(url, display_id) - title = self._search_regex(r'<title>(.*?)\ \|\ Jamendo\ Music\ .*', webpage, 'title') + title, artist, track = self._extract_meta(webpage) formats = [{ 'url': 'https://%s.jamendo.com/?trackid=%s&format=%s&from=app-97dab294' @@ -46,17 +67,23 @@ class JamendoIE(InfoExtractor): thumbnail = self._html_search_meta( 'image', webpage, 'thumbnail', fatal=False) + duration = parse_duration(self._search_regex( + r']+itemprop=["\']duration["\'][^>]+content=["\'](.+?)["\']', + webpage, 'duration', fatal=False)) return { 'id': track_id, 'display_id': display_id, 'thumbnail': thumbnail, 'title': title, + 'duration': duration, + 'artist': artist, + 'track': track, 'formats': formats } -class JamendoAlbumIE(InfoExtractor): +class JamendoAlbumIE(JamendoBaseIE): _VALID_URL = r'https?://(?:www\.)?jamendo\.com/album/(?P[0-9]+)/(?P[\w-]+)' _TEST = { 'url': 'https://www.jamendo.com/album/121486/duck-on-cover', @@ -69,14 +96,18 @@ class JamendoAlbumIE(InfoExtractor): 'info_dict': { 'id': '1032333', 'ext': 'flac', - 'title': 'Shearer - Warmachine' + 'title': 'Shearer - Warmachine', + 'artist': 'Shearer', + 'track': 'Warmachine', } }, { 'md5': '1f358d7b2f98edfe90fd55dac0799d50', 'info_dict': { 'id': '1032330', 'ext': 'flac', - 'title': 'Shearer - Without Your Ghost' + 'title': 'Shearer - Without Your Ghost', + 'artist': 'Shearer', + 'track': 'Without Your Ghost', } }], 'params': { @@ -90,18 +121,18 @@ class JamendoAlbumIE(InfoExtractor): webpage = self._download_webpage(url, mobj.group('display_id')) - title = self._search_regex(r'(.*?)\ \|\ Jamendo\ Music\ .*', webpage, 'title') + title, artist, album = self._extract_meta(webpage, fatal=False) - entries = [ - self.url_result( - compat_urlparse.urljoin(url, m.group('path')), - ie=JamendoIE.ie_key(), - video_id=self._search_regex( - r'/track/(\d+)', m.group('path'), - 'track id', default=None)) - for m in re.finditer( - r']+href=(["\'])(?P(?:(?!\1).)+)\1[^>]+class=["\'][^>]*js-trackrow-albumpage-link', - webpage) - ] + entries = [{ + '_type': 'url_transparent', + 'url': compat_urlparse.urljoin(url, m.group('path')), + 'ie_key': JamendoIE.ie_key(), + 'id': self._search_regex( + r'/track/(\d+)', m.group('path'), 'track id', default=None), + 'artist': artist, + 'album': album, + } for m in re.finditer( + r']+href=(["\'])(?P(?:(?!\1).)+)\1[^>]+class=["\'][^>]*js-trackrow-albumpage-link', + webpage)] return self.playlist_result(entries, album_id, title)