[iqiyi] Simplify and improve regex patterns

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

diff --git a/youtube_dl/extractor/iqiyi.py b/youtube_dl/extractor/iqiyi.py
index d96d13225ee0d874e827b24403c55bf76d2e230f..18a7587a257ae1f4fb2dfcb48978afec06ad113a 100644 (file)
--- a/youtube_dl/extractor/iqiyi.py
+++ b/youtube_dl/extractor/iqiyi.py
@@ -4,39 +4,32 @@ from __future__ import unicode_literals
 
 from .common import InfoExtractor
 
-from ..compat import (
-    compat_chr,
-    compat_parse_qs,
-    compat_urllib_parse,
-    compat_urllib_request,
-    compat_urlparse,
-    compat_str,
-)
+from ..compat import compat_urllib_parse
 
 from ..utils import ExtractorError
 
 import re
 import time
-import json
 import uuid
 import math
 import random
 import zlib
 import hashlib
 
+
 class IqiyiIE(InfoExtractor):
     IE_NAME = 'iqiyi'
 
     _VALID_URL = r'http://(?:www\.)iqiyi.com/.+?\.html'
 
     _TEST = {
-            'url': 'http://www.iqiyi.com/v_19rrojlavg.html',
-            'md5': '260f0f59686e65e886995d0ba791ab83',
-            'info_dict': {
-                'id': '9c1fb1b99d192b21c559e5a1a2cb3c73',
-                'title': '美国德州空中惊现奇异云团 酷似UFO',
-                'ext': 'f4v'
-            }
+        'url': 'http://www.iqiyi.com/v_19rrojlavg.html',
+        'md5': '2cb594dc2781e6c941a110d8f358118b',
+        'info_dict': {
+            'id': '9c1fb1b99d192b21c559e5a1a2cb3c73',
+            'title': '美国德州空中惊现奇异云团 酷似UFO',
+            'ext': 'f4v',
+        }
     }
 
     def construct_video_urls(self, data, video_id, _uuid):
@@ -54,7 +47,7 @@ class IqiyiIE(InfoExtractor):
             c = len(b)
             s = ''
             for i in range(c - 1, -1, -1):
-                a = do_xor(int(b[c-i-1], 16), i)
+                a = do_xor(int(b[c - i - 1], 16), i)
                 s += chr(a)
             return s[::-1]
 
@@ -62,14 +55,17 @@ class IqiyiIE(InfoExtractor):
             mg = ')(*&^flash@#$%a'
             tm = self._download_json(
                 'http://data.video.qiyi.com/t?tn=' + str(random.random()), video_id)['t']
-            t = str(int(math.floor(int(tm)/(600.0))))
-            return hashlib.md5(
-                 (t+mg+x).encode('utf8')).hexdigest()
+            t = str(int(math.floor(int(tm) / (600.0))))
+            return hashlib.md5((t + mg + x).encode('utf8')).hexdigest()
 
         video_urls_dict = {}
         for i in data['vp']['tkl'][0]['vs']:
             if 0 < int(i['bid']) <= 10:
                 format_id = self.get_format(i['bid'])
+            else:
+                continue
+
+            video_urls = []
 
             video_urls_info = i['fs']
             if not i['fs'][0]['l'].startswith('/'):
@@ -77,7 +73,6 @@ class IqiyiIE(InfoExtractor):
                 if t.endswith('mp4'):
                     video_urls_info = i['flvs']
 
-            video_urls = []
             for ii in video_urls_info:
                 vl = ii['l']
                 if not vl.startswith('/'):
@@ -108,15 +103,27 @@ class IqiyiIE(InfoExtractor):
         return video_urls_dict
 
     def get_format(self, bid):
-        bid_dict = {
-            '1': 'standard',
-            '2': 'high',
-            '3': 'super',
-            '4': 'suprt-high',
-            '5': 'fullhd',
-            '10': '4k'
+        _dict = {
+            '1': 'h6',
+            '2': 'h5',
+            '3': 'h4',
+            '4': 'h3',
+            '5': 'h2',
+            '10': 'h1'
+        }
+        return _dict.get(str(bid), None)
+
+    def get_bid(self, format_id):
+        _dict = {
+            'h6': '1',
+            'h5': '2',
+            'h4': '3',
+            'h3': '4',
+            'h2': '5',
+            'h1': '10',
+            'best': 'best'
         }
-        return bid_dict[str(bid)]
+        return _dict.get(format_id, None)
 
     def get_raw_data(self, tvid, video_id, enc_key, _uuid):
         tm = str(int(time.time()))
@@ -154,12 +161,11 @@ class IqiyiIE(InfoExtractor):
         webpage = self._download_webpage(
             url, 'temp_id', note='download video page')
         tvid = self._search_regex(
-            r'tvId ?= ?(\'|\")(?P<tvid>\d+)', webpage, 'tvid', flags=re.I, group='tvid')
+            r'data-player-tvid\s*=\s*[\'"](\d+)', webpage, 'tvid')
         video_id = self._search_regex(
-            r'videoId ?= ?(\'|\")(?P<video_id>[a-z\d]+)',
-            webpage, 'video_id', flags=re.I, group='video_id')
+            r'data-player-videoid\s*=\s*[\'"]([a-f\d]+)', webpage, 'video_id')
         swf_url = self._search_regex(
-            r'(?P<swf>http://.+?MainPlayer.+?\.swf)', webpage, 'swf')
+            r'(http://.+?MainPlayer.+?\.swf)', webpage, 'swf player URL')
         _uuid = uuid.uuid4().hex
 
         enc_key = self.get_enc_key(swf_url, video_id)
@@ -174,27 +180,30 @@ class IqiyiIE(InfoExtractor):
         title = data['vi']['vn']
 
         # generate video_urls_dict
-        video_urls_dict = self.construct_video_urls(data, video_id, _uuid)
+        video_urls_dict = self.construct_video_urls(
+            data, video_id, _uuid)
 
         # construct info
         entries = []
         for format_id in video_urls_dict:
             video_urls = video_urls_dict[format_id]
             for i, video_url_info in enumerate(video_urls):
-                if len(entries) < i+1:
+                if len(entries) < i + 1:
                     entries.append({'formats': []})
                 entries[i]['formats'].append(
                     {
                         'url': video_url_info[0],
                         'filesize': video_url_info[-1],
                         'format_id': format_id,
+                        'preference': int(self.get_bid(format_id))
                     }
                 )
 
         for i in range(len(entries)):
+            self._sort_formats(entries[i]['formats'])
             entries[i].update(
                 {
-                    'id': '_part%d' % (i+1),
+                    'id': '_part%d' % (i + 1),
                     'title': title,
                 }
             )