Imported Upstream version 2015.06.04.1

[youtubedl] / youtube_dl / extractor / bilibili.py

diff --git a/youtube_dl/extractor/bilibili.py b/youtube_dl/extractor/bilibili.py
index 45067b944996f7d3f6ce1ced97b379cd6ce40ff9..2103ed73aad860738bba5108ee27a86fd921d29c 100644 (file)
--- a/youtube_dl/extractor/bilibili.py
+++ b/youtube_dl/extractor/bilibili.py
@@ -2,37 +2,47 @@
 from __future__ import unicode_literals
 
 import re
 from __future__ import unicode_literals
 
 import re

+import itertools
+import json
+import xml.etree.ElementTree as ET

 
 from .common import InfoExtractor
 from ..utils import (
 
 from .common import InfoExtractor
 from ..utils import (

-    compat_parse_qs,
-    ExtractorError,

     int_or_none,
     unified_strdate,
     int_or_none,
     unified_strdate,

+    ExtractorError,

 )
 
 
 class BiliBiliIE(InfoExtractor):
 )
 
 
 class BiliBiliIE(InfoExtractor):

-    _VALID_URL = r'http://www\.bilibili\.tv/video/av(?P<id>[0-9]+)/'
+    _VALID_URL = r'http://www\.bilibili\.(?:tv|com)/video/av(?P<id>[0-9]+)/'

 
 

-    _TEST = {
+    _TESTS = [{

         'url': 'http://www.bilibili.tv/video/av1074402/',
         'md5': '2c301e4dab317596e837c3e7633e7d86',
         'info_dict': {
         'url': 'http://www.bilibili.tv/video/av1074402/',
         'md5': '2c301e4dab317596e837c3e7633e7d86',
         'info_dict': {

-            'id': '1074402',
+            'id': '1074402_part1',

             'ext': 'flv',
             'title': '【金坷垃】金泡沫',
             'duration': 308,
             'upload_date': '20140420',
             'thumbnail': 're:^https?://.+\.jpg',
         },
             'ext': 'flv',
             'title': '【金坷垃】金泡沫',
             'duration': 308,
             'upload_date': '20140420',
             'thumbnail': 're:^https?://.+\.jpg',
         },

-    }
+    }, {
+        'url': 'http://www.bilibili.com/video/av1041170/',
+        'info_dict': {
+            'id': '1041170',
+            'title': '【BD1080P】刀语【诸神&异域】',
+        },
+        'playlist_count': 9,
+    }]

 
     def _real_extract(self, url):
 
     def _real_extract(self, url):

-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-
+        video_id = self._match_id(url)

         webpage = self._download_webpage(url, video_id)
         webpage = self._download_webpage(url, video_id)

+
+        if self._search_regex(r'(此视频不存在或被删除)', webpage, 'error message', default=None):
+            raise ExtractorError('The video does not exist or was deleted', expected=True)

         video_code = self._search_regex(
             r'(?s)<div itemprop="video".*?>(.*?)</div>', webpage, 'video code')
 
         video_code = self._search_regex(
             r'(?s)<div itemprop="video".*?>(.*?)</div>', webpage, 'video code')
 


@@ -55,19 +65,39 @@ class BiliBiliIE(InfoExtractor):
         thumbnail = self._html_search_meta(
             'thumbnailUrl', video_code, 'thumbnail', fatal=False)
 
         thumbnail = self._html_search_meta(
             'thumbnailUrl', video_code, 'thumbnail', fatal=False)
 

-        player_params = compat_parse_qs(self._html_search_regex(
-            r'<iframe .*?class="player" src="https://secure.bilibili.tv/secure,([^"]+)"',
-            webpage, 'player params'))
+        cid = self._search_regex(r'cid=(\d+)', webpage, 'cid')
+
+        entries = []

 
 

-        if 'cid' in player_params:
-            cid = player_params['cid'][0]
+        lq_page = self._download_webpage(
+            'http://interface.bilibili.com/v_cdn_play?appkey=1&cid=%s' % cid,
+            video_id,
+            note='Downloading LQ video info'
+        )
+        try:
+            err_info = json.loads(lq_page)
+            raise ExtractorError(
+                'BiliBili said: ' + err_info['error_text'], expected=True)
+        except ValueError:
+            pass

 
 

-            lq_doc = self._download_xml(
-                'http://interface.bilibili.cn/v_cdn_play?cid=%s' % cid,
-                video_id,
-                note='Downloading LQ video info'
-            )
-            lq_durl = lq_doc.find('.//durl')
+        lq_doc = ET.fromstring(lq_page)
+        lq_durls = lq_doc.findall('./durl')
+
+        hq_doc = self._download_xml(
+            'http://interface.bilibili.com/playurl?appkey=1&cid=%s' % cid,
+            video_id,
+            note='Downloading HQ video info',
+            fatal=False,
+        )
+        if hq_doc is not False:
+            hq_durls = hq_doc.findall('./durl')
+            assert len(lq_durls) == len(hq_durls)
+        else:
+            hq_durls = itertools.repeat(None)
+
+        i = 1
+        for lq_durl, hq_durl in zip(lq_durls, hq_durls):

             formats = [{
                 'format_id': 'lq',
                 'quality': 1,
             formats = [{
                 'format_id': 'lq',
                 'quality': 1,


@@ -75,15 +105,7 @@ class BiliBiliIE(InfoExtractor):
                 'filesize': int_or_none(
                     lq_durl.find('./size'), get_attr='text'),
             }]
                 'filesize': int_or_none(
                     lq_durl.find('./size'), get_attr='text'),
             }]

-
-            hq_doc = self._download_xml(
-                'http://interface.bilibili.cn/playurl?cid=%s' % cid,
-                video_id,
-                note='Downloading HQ video info',
-                fatal=False,
-            )
-            if hq_doc is not False:
-                hq_durl = hq_doc.find('.//durl')
+            if hq_durl:

                 formats.append({
                     'format_id': 'hq',
                     'quality': 2,
                 formats.append({
                     'format_id': 'hq',
                     'quality': 2,


@@ -92,15 +114,22 @@ class BiliBiliIE(InfoExtractor):
                     'filesize': int_or_none(
                         hq_durl.find('./size'), get_attr='text'),
                 })
                     'filesize': int_or_none(
                         hq_durl.find('./size'), get_attr='text'),
                 })

-        else:
-            raise ExtractorError('Unsupported player parameters: %r' % (player_params,))
+            self._sort_formats(formats)
+
+            entries.append({
+                'id': '%s_part%d' % (video_id, i),
+                'title': title,
+                'formats': formats,
+                'duration': duration,
+                'upload_date': upload_date,
+                'thumbnail': thumbnail,
+            })
+
+            i += 1

 
 

-        self._sort_formats(formats)

         return {
         return {

+            '_type': 'multi_video',
+            'entries': entries,

             'id': video_id,
             'id': video_id,

-            'title': title,
-            'formats': formats,
-            'duration': duration,
-            'upload_date': upload_date,
-            'thumbnail': thumbnail,
+            'title': title

         }
         }