Merge remote-tracking branch 'upstream/master' into bliptv

[youtube-dl] / youtube_dl / extractor / iqiyi.py

diff --git a/youtube_dl/extractor/iqiyi.py b/youtube_dl/extractor/iqiyi.py
index c17e1fde4c3ff969c45e37acd0e3cbab723325f8..2df1da3f00a9d9655a0c9f6a447510dcf9e2caf0 100644 (file)
--- a/youtube_dl/extractor/iqiyi.py
+++ b/youtube_dl/extractor/iqiyi.py
@@ -4,10 +4,8 @@ from __future__ import unicode_literals
 import hashlib
 import math
 import random
-import re
 import time
 import uuid
-import zlib
 
 from .common import InfoExtractor
 from ..compat import compat_urllib_parse
@@ -16,10 +14,11 @@ from ..utils import ExtractorError
 
 class IqiyiIE(InfoExtractor):
     IE_NAME = 'iqiyi'
+    IE_DESC = '爱奇艺'
 
-    _VALID_URL = r'http://(?:www\.)iqiyi.com/.+?\.html'
+    _VALID_URL = r'http://(?:www\.)iqiyi.com/v_.+?\.html'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.iqiyi.com/v_19rrojlavg.html',
         'md5': '2cb594dc2781e6c941a110d8f358118b',
         'info_dict': {
@@ -27,7 +26,78 @@ class IqiyiIE(InfoExtractor):
             'title': '美国德州空中惊现奇异云团 酷似UFO',
             'ext': 'f4v',
         }
-    }
+    }, {
+        'url': 'http://www.iqiyi.com/v_19rrhnnclk.html',
+        'info_dict': {
+            'id': 'e3f585b550a280af23c98b6cb2be19fb',
+            'title': '名侦探柯南第752集',
+        },
+        'playlist': [{
+            'info_dict': {
+                'id': 'e3f585b550a280af23c98b6cb2be19fb_part1',
+                'ext': 'f4v',
+                'title': '名侦探柯南第752集',
+            },
+        }, {
+            'info_dict': {
+                'id': 'e3f585b550a280af23c98b6cb2be19fb_part2',
+                'ext': 'f4v',
+                'title': '名侦探柯南第752集',
+            },
+        }, {
+            'info_dict': {
+                'id': 'e3f585b550a280af23c98b6cb2be19fb_part3',
+                'ext': 'f4v',
+                'title': '名侦探柯南第752集',
+            },
+        }, {
+            'info_dict': {
+                'id': 'e3f585b550a280af23c98b6cb2be19fb_part4',
+                'ext': 'f4v',
+                'title': '名侦探柯南第752集',
+            },
+        }, {
+            'info_dict': {
+                'id': 'e3f585b550a280af23c98b6cb2be19fb_part5',
+                'ext': 'f4v',
+                'title': '名侦探柯南第752集',
+            },
+        }, {
+            'info_dict': {
+                'id': 'e3f585b550a280af23c98b6cb2be19fb_part6',
+                'ext': 'f4v',
+                'title': '名侦探柯南第752集',
+            },
+        }, {
+            'info_dict': {
+                'id': 'e3f585b550a280af23c98b6cb2be19fb_part7',
+                'ext': 'f4v',
+                'title': '名侦探柯南第752集',
+            },
+        }, {
+            'info_dict': {
+                'id': 'e3f585b550a280af23c98b6cb2be19fb_part8',
+                'ext': 'f4v',
+                'title': '名侦探柯南第752集',
+            },
+        }],
+        'params': {
+            'skip_download': True,
+        },
+    }]
+
+    _FORMATS_MAP = [
+        ('1', 'h6'),
+        ('2', 'h5'),
+        ('3', 'h4'),
+        ('4', 'h3'),
+        ('5', 'h2'),
+        ('10', 'h1'),
+    ]
+
+    @staticmethod
+    def md5_text(text):
+        return hashlib.md5(text.encode('utf-8')).hexdigest()
 
     def construct_video_urls(self, data, video_id, _uuid):
         def do_xor(x, y):
@@ -55,7 +125,7 @@ class IqiyiIE(InfoExtractor):
                 note='Download path key of segment %d for format %s' % (segment_index + 1, format_id)
             )['t']
             t = str(int(math.floor(int(tm) / (600.0))))
-            return hashlib.md5((t + mg + x).encode('utf8')).hexdigest()
+            return self.md5_text(t + mg + x)
 
         video_urls_dict = {}
         for format_item in data['vp']['tkl'][0]['vs']:
@@ -104,44 +174,28 @@ class IqiyiIE(InfoExtractor):
         return video_urls_dict
 
     def get_format(self, bid):
-        _dict = {
-            '1': 'h6',
-            '2': 'h5',
-            '3': 'h4',
-            '4': 'h3',
-            '5': 'h2',
-            '10': 'h1'
-        }
-        return _dict.get(str(bid), None)
+        matched_format_ids = [_format_id for _bid, _format_id in self._FORMATS_MAP if _bid == str(bid)]
+        return matched_format_ids[0] if len(matched_format_ids) else None
 
     def get_bid(self, format_id):
-        _dict = {
-            'h6': '1',
-            'h5': '2',
-            'h4': '3',
-            'h3': '4',
-            'h2': '5',
-            'h1': '10',
-            'best': 'best'
-        }
-        return _dict.get(format_id, None)
+        matched_bids = [_bid for _bid, _format_id in self._FORMATS_MAP if _format_id == format_id]
+        return matched_bids[0] if len(matched_bids) else None
 
     def get_raw_data(self, tvid, video_id, enc_key, _uuid):
         tm = str(int(time.time()))
+        tail = tm + tvid
         param = {
             'key': 'fvip',
-            'src': hashlib.md5(b'youtube-dl').hexdigest(),
+            'src': self.md5_text('youtube-dl'),
             'tvId': tvid,
             'vid': video_id,
             'vinfo': 1,
             'tm': tm,
-            'enc': hashlib.md5(
-                (enc_key + tm + tvid).encode('utf8')).hexdigest(),
+            'enc': self.md5_text(enc_key + tail),
             'qyid': _uuid,
             'tn': random.random(),
             'um': 0,
-            'authkey': hashlib.md5(
-                (tm + tvid).encode('utf8')).hexdigest()
+            'authkey': self.md5_text(self.md5_text('') + tail),
         }
 
         api_url = 'http://cache.video.qiyi.com/vms' + '?' + \
@@ -150,12 +204,10 @@ class IqiyiIE(InfoExtractor):
         return raw_data
 
     def get_enc_key(self, swf_url, video_id):
-        req = self._request_webpage(
-            swf_url, video_id, note='download swf content')
-        cn = req.read()
-        cn = zlib.decompress(cn[8:])
-        pt = re.compile(b'MixerRemote\x08(?P<enc_key>.+?)\$&vv')
-        enc_key = self._search_regex(pt, cn, 'enc_key').decode('utf8')
+        # TODO: automatic key extraction
+        # last update at 2015-10-22 for Zombie::bite
+        # '7223c67061dbea1259d0ceb44f44b6d62288f4f80c972170de5201d2321060270e05'[2:66][0::2]
+        enc_key = '2c76de15dcb44bd28ff0927d50d31620'
         return enc_key
 
     def _real_extract(self, url):
@@ -166,7 +218,7 @@ class IqiyiIE(InfoExtractor):
         video_id = self._search_regex(
             r'data-player-videoid\s*=\s*[\'"]([a-f\d]+)', webpage, 'video_id')
         swf_url = self._search_regex(
-            r'(http://.+?MainPlayer.+?\.swf)', webpage, 'swf player URL')
+            r'(http://[^\'"]+MainPlayer[^.]+\.swf)', webpage, 'swf player URL')
         _uuid = uuid.uuid4().hex
 
         enc_key = self.get_enc_key(swf_url, video_id)
@@ -207,7 +259,7 @@ class IqiyiIE(InfoExtractor):
             self._sort_formats(entries[i]['formats'])
             entries[i].update(
                 {
-                    'id': '_part%d' % (i + 1),
+                    'id': '%s_part%d' % (video_id, i + 1),
                     'title': title,
                 }
             )