[bilibili] Support new Bangumi URLs (closes #11845)

To reduce complexity, I don't support old Bangumi URLs directly via _VALID_URL. Instead, I choose to let it go to generic redirection. An example can be found in #10190: http://bangumi.bilibili.com/anime/v/40062
2017-02-02 21:51:31 +08:00 · 2017-02-02 21:51:31 +08:00 · bd8f48c78b
commit bd8f48c78b
parent 81aeafeb44
3 changed files with 134 additions and 11 deletions
--- a/5
+++ b/5
@ -1,3 +1,8 @@
+version <unreleased>
+
+Extractors
+ [bilibili] Support new Bangumi URLs (#11845)
+
 version 2017.02.01

 Extractors
--- a/youtube_dl/extractor/bilibili.py
+++ b/youtube_dl/extractor/bilibili.py
@ -5,19 +5,27 @@ import hashlib
 import re

 from .common import InfoExtractor
-from ..compat import compat_parse_qs
+from ..compat import (
+    compat_parse_qs,
+    compat_urlparse,
+)
 from ..utils import (
+    ExtractorError,
    int_or_none,
    float_or_none,
+    parse_iso8601,
+    smuggle_url,
+    strip_jsonp,
    unified_timestamp,
+    unsmuggle_url,
    urlencode_postdata,
 )


 class BiliBiliIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.|bangumi\.|)bilibili\.(?:tv|com)/(?:video/av|anime/v/)(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:www\.|bangumi\.|)bilibili\.(?:tv|com)/(?:video/av|anime/(?P<anime_id>\d+)/play#)(?P<id>\d+)'

-    _TEST = {
+    _TESTS = [{
        'url': 'http://www.bilibili.tv/video/av1074402/',
        'md5': '9fa226fe2b8a9a4d5a69b4c6a183417e',
        'info_dict': {
@ -32,25 +40,61 @@ class BiliBiliIE(InfoExtractor):
            'uploader': '菊子桑',
            'uploader_id': '156160',
        },
-    }
+    }, {
+        # Tested in BiliBiliBangumiIE
+        'url': 'http://bangumi.bilibili.com/anime/1869/play#40062',
+        'only_matching': True,
+    }, {
+        'url': 'http://bangumi.bilibili.com/anime/5802/play#100643',
+        'md5': '3f721ad1e75030cc06faf73587cfec57',
+        'info_dict': {
+            'id': '100643',
+            'ext': 'mp4',
+            'title': 'CHAOS;CHILD',
+            'description': '如果你是神明，并且能够让妄想成为现实。那你会进行怎么样的妄想？是淫靡的世界？独裁社会？毁灭性的制裁？还是……2015年，涩谷。从6年前发生的大灾害“涩谷地震”之后复兴了的这个街区里新设立的私立高中...',
+        },
+        'skip': 'Geo-restricted to China',
+    }]

    _APP_KEY = '84956560bc028eb7'
    _BILIBILI_KEY = '94aba54af9065f71de72f5508f1cd42e'

+    def _report_error(self, result):
+        if 'message' in result:
+            raise ExtractorError('%s said: %s' % (self.IE_NAME, result['message']), expected=True)
+        elif 'code' in result:
+            raise ExtractorError('%s returns error %d' % (self.IE_NAME, result['code']), expected=True)
+        else:
+            raise ExtractorError('Can\'t extract Bangumi episode ID')
+
    def _real_extract(self, url):
-        video_id = self._match_id(url)
+        url, smuggled_data = unsmuggle_url(url, {})
+
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        anime_id = mobj.group('anime_id')
        webpage = self._download_webpage(url, video_id)

-        if 'anime/v' not in url:
+        if 'anime/' not in url:
            cid = compat_parse_qs(self._search_regex(
                [r'EmbedPlayer\([^)]+,\s*"([^"]+)"\)',
                 r'<iframe[^>]+src="https://secure\.bilibili\.com/secure,([^"]+)"'],
                webpage, 'player parameters'))['cid'][0]
        else:
+            if 'no_bangumi_tip' not in smuggled_data:
+                self.to_screen('Downloading episode %s. To download all videos in anime %s, re-run youtube-dl with %s' % (
+                    video_id, anime_id, compat_urlparse.urljoin(url, '//bangumi.bilibili.com/anime/%s' % anime_id)))
+            headers = {
+                'Content-Type': 'application/x-www-form-urlencoded; charset=UTF-8',
+            }
+            headers.update(self.geo_verification_headers())
+
            js = self._download_json(
                'http://bangumi.bilibili.com/web_api/get_source', video_id,
                data=urlencode_postdata({'episode_id': video_id}),
-                headers={'Content-Type': 'application/x-www-form-urlencoded; charset=UTF-8'})
+                headers=headers)
+            if 'result' not in js:
+                self._report_error(js)
            cid = js['result']['cid']

        payload = 'appkey=%s&cid=%s&otype=json&quality=2&type=mp4' % (self._APP_KEY, cid)
@ -58,7 +102,11 @@ class BiliBiliIE(InfoExtractor):

        video_info = self._download_json(
            'http://interface.bilibili.com/playurl?%s&sign=%s' % (payload, sign),
-            video_id, note='Downloading video info page')
+            video_id, note='Downloading video info page',
+            headers=self.geo_verification_headers())
+
+        if 'durl' not in video_info:
+            self._report_error(video_info)

        entries = []

@ -85,7 +133,7 @@ class BiliBiliIE(InfoExtractor):
        title = self._html_search_regex('<h1[^>]+title="([^"]+)">', webpage, 'title')
        description = self._html_search_meta('description', webpage)
        timestamp = unified_timestamp(self._html_search_regex(
-            r'<time[^>]+datetime="([^"]+)"', webpage, 'upload time', fatal=False))
+            r'<time[^>]+datetime="([^"]+)"', webpage, 'upload time', default=None))
        thumbnail = self._html_search_meta(['og:image', 'thumbnailUrl'], webpage)

        # TODO 'view_count' requires deobfuscating Javascript
@ -99,7 +147,7 @@ class BiliBiliIE(InfoExtractor):
        }

        uploader_mobj = re.search(
-            r'<a[^>]+href="https?://space\.bilibili\.com/(?P<id>\d+)"[^>]+title="(?P<name>[^"]+)"',
+            r'<a[^>]+href="(?:https?:)?//space\.bilibili\.com/(?P<id>\d+)"[^>]+title="(?P<name>[^"]+)"',
            webpage)
        if uploader_mobj:
            info.update({
@ -123,3 +171,70 @@ class BiliBiliIE(InfoExtractor):
                'description': description,
                'entries': entries,
            }
+
+
+class BiliBiliBangumiIE(InfoExtractor):
+    _VALID_URL = r'https?://bangumi\.bilibili\.com/anime/(?P<id>\d+)'
+
+    IE_NAME = 'bangumi.bilibili.com'
+    IE_DESC = 'BiliBili番剧'
+
+    _TESTS = [{
+        'url': 'http://bangumi.bilibili.com/anime/1869',
+        'info_dict': {
+            'id': '1869',
+            'title': '混沌武士',
+            'description': 'md5:6a9622b911565794c11f25f81d6a97d2',
+        },
+        'playlist_count': 26,
+    }, {
+        'url': 'http://bangumi.bilibili.com/anime/1869',
+        'info_dict': {
+            'id': '1869',
+            'title': '混沌武士',
+            'description': 'md5:6a9622b911565794c11f25f81d6a97d2',
+        },
+        'playlist': [{
+            'md5': '91da8621454dd58316851c27c68b0c13',
+            'info_dict': {
+                'id': '40062',
+                'ext': 'mp4',
+                'title': '混沌武士',
+                'description': '故事发生在日本的江户时代。风是一个小酒馆的打工女。一日，酒馆里来了一群恶霸，虽然他们的举动令风十分不满，但是毕竟风只是一届女流，无法对他们采取什么行动，只能在心里嘟哝。这时，酒家里又进来了个“不良份子...',
+                'timestamp': 1414538739,
+                'upload_date': '20141028',
+                'episode': '疾风怒涛 Tempestuous Temperaments',
+                'episode_number': 1,
+            },
+        }],
+        'params': {
+            'playlist_items': '1',
+        },
+    }]
+
+    @classmethod
+    def suitable(cls, url):
+        return False if BiliBiliIE.suitable(url) else super(BiliBiliBangumiIE, cls).suitable(url)
+
+    def _real_extract(self, url):
+        bangumi_id = self._match_id(url)
+
+        # Sometimes this API returns a JSONP response
+        season_info = self._download_json(
+            'http://bangumi.bilibili.com/jsonp/seasoninfo/%s.ver' % bangumi_id,
+            bangumi_id, transform_source=strip_jsonp)['result']
+
+        entries = [{
+            '_type': 'url_transparent',
+            'url': smuggle_url(episode['webplay_url'], {'no_bangumi_tip': 1}),
+            'ie_key': BiliBiliIE.ie_key(),
+            'timestamp': parse_iso8601(episode.get('update_time'), delimiter=' '),
+            'episode': episode.get('index_title'),
+            'episode_number': int_or_none(episode.get('index')),
+        } for episode in season_info['episodes']]
+
+        entries = sorted(entries, key=lambda entry: entry.get('episode_number'))
+
+        return self.playlist_result(
+            entries, bangumi_id,
+            season_info.get('bangumi_title'), season_info.get('evaluate'))
--- a/youtube_dl/extractor/extractors.py
+++ b/youtube_dl/extractor/extractors.py
@ -103,7 +103,10 @@ from .beatport import BeatportIE
 from .bet import BetIE
 from .bigflix import BigflixIE
 from .bild import BildIE
-from .bilibili import BiliBiliIE
+from .bilibili import (
+    BiliBiliIE,
+    BiliBiliBangumiIE,
+)
 from .biobiochiletv import BioBioChileTVIE
 from .biqle import BIQLEIE
 from .bleacherreport import (