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 75d744852edc382721cee8556067f89ccb0092df..2103ed73aad860738bba5108ee27a86fd921d29c 100644 (file)
--- a/youtube_dl/extractor/bilibili.py
+++ b/youtube_dl/extractor/bilibili.py
@@ -2,34 +2,47 @@
 from __future__ import unicode_literals
 
 import re
+import itertools
+import json
+import xml.etree.ElementTree as ET
 
 from .common import InfoExtractor
 from ..utils import (
     int_or_none,
     unified_strdate,
+    ExtractorError,
 )
 
 
 class BiliBiliIE(InfoExtractor):
     _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': {
-            'id': '1074402',
+            'id': '1074402_part1',
             '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):
         video_id = self._match_id(url)
         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')
 
@@ -54,19 +67,22 @@ class BiliBiliIE(InfoExtractor):
 
         cid = self._search_regex(r'cid=(\d+)', webpage, 'cid')
 
-        lq_doc = self._download_xml(
+        entries = []
+
+        lq_page = self._download_webpage(
             'http://interface.bilibili.com/v_cdn_play?appkey=1&cid=%s' % cid,
             video_id,
             note='Downloading LQ video info'
         )
-        lq_durl = lq_doc.find('./durl')
-        formats = [{
-            'format_id': 'lq',
-            'quality': 1,
-            'url': lq_durl.find('./url').text,
-            'filesize': int_or_none(
-                lq_durl.find('./size'), get_attr='text'),
-        }]
+        try:
+            err_info = json.loads(lq_page)
+            raise ExtractorError(
+                'BiliBili said: ' + err_info['error_text'], expected=True)
+        except ValueError:
+            pass
+
+        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,
@@ -75,22 +91,45 @@ class BiliBiliIE(InfoExtractor):
             fatal=False,
         )
         if hq_doc is not False:
-            hq_durl = hq_doc.find('./durl')
-            formats.append({
-                'format_id': 'hq',
-                'quality': 2,
-                'ext': 'flv',
-                'url': hq_durl.find('./url').text,
+            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,
+                'url': lq_durl.find('./url').text,
                 'filesize': int_or_none(
-                    hq_durl.find('./size'), get_attr='text'),
+                    lq_durl.find('./size'), get_attr='text'),
+            }]
+            if hq_durl:
+                formats.append({
+                    'format_id': 'hq',
+                    'quality': 2,
+                    'ext': 'flv',
+                    'url': hq_durl.find('./url').text,
+                    'filesize': int_or_none(
+                        hq_durl.find('./size'), get_attr='text'),
+                })
+            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,
             })
 
-        self._sort_formats(formats)
+            i += 1
+
         return {
+            '_type': 'multi_video',
+            'entries': entries,
             'id': video_id,
-            'title': title,
-            'formats': formats,
-            'duration': duration,
-            'upload_date': upload_date,
-            'thumbnail': thumbnail,
+            'title': title
         }