Merge pull request #6463 from jaimeMF/format_spec_fix_dashes

[YoutubeDL] format spec: correctly handle dashes and other unused operators

diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py
index 8cef61c3c9a235a6d3f3230b8517222334dc0fbc..6df89f81403a34a5c12b3b37110e8258e3f4499d 100644 (file)
--- a/youtube_dl/extractor/generic.py
+++ b/youtube_dl/extractor/generic.py
@@ -1655,7 +1655,7 @@ class GenericIE(InfoExtractor):
         if not found:
             # Broaden the findall a little bit: JWPlayer JS loader
             found = filter_video(re.findall(
-                r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage))
+                r'[^A-Za-z0-9]?(?:file|video_url)["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage))
         if not found:
             # Flow player
             found = filter_video(re.findall(r'''(?xs)

diff --git a/youtube_dl/extractor/lynda.py b/youtube_dl/extractor/lynda.py
index a00f6e5e5eb1d398ef0776d8b20dcb1dd51ec082..deead220af21bd5e37deab93446dc9e9cac59ba8 100644 (file)
--- a/youtube_dl/extractor/lynda.py
+++ b/youtube_dl/extractor/lynda.py
@@ -17,7 +17,6 @@ from ..utils import (
 
 class LyndaBaseIE(InfoExtractor):
     _LOGIN_URL = 'https://www.lynda.com/login/login.aspx'
-    _SUCCESSFUL_LOGIN_REGEX = r'isLoggedIn: true'
     _ACCOUNT_CREDENTIALS_HINT = 'Use --username and --password options to provide lynda.com account credentials.'
     _NETRC_MACHINE = 'lynda'
 
@@ -41,7 +40,7 @@ class LyndaBaseIE(InfoExtractor):
             request, None, 'Logging in as %s' % username)
 
         # Not (yet) logged in
-        m = re.search(r'loginResultJson = \'(?P<json>[^\']+)\';', login_page)
+        m = re.search(r'loginResultJson\s*=\s*\'(?P<json>[^\']+)\';', login_page)
         if m is not None:
             response = m.group('json')
             response_json = json.loads(response)
@@ -70,7 +69,7 @@ class LyndaBaseIE(InfoExtractor):
                     request, None,
                     'Confirming log in and log out from another device')
 
-        if re.search(self._SUCCESSFUL_LOGIN_REGEX, login_page) is None:
+        if all(not re.search(p, login_page) for p in ('isLoggedIn\s*:\s*true', r'logout\.aspx', r'>Log out<')):
             raise ExtractorError('Unable to log in')
 
 

diff --git a/youtube_dl/extractor/tudou.py b/youtube_dl/extractor/tudou.py
index c89de5ba4a46bb261987d8dbee5f55b3d05492da..84fe71aef8501dcd9055dc517ae9097844d99a5d 100644 (file)
--- a/youtube_dl/extractor/tudou.py
+++ b/youtube_dl/extractor/tudou.py
@@ -29,6 +29,8 @@ class TudouIE(InfoExtractor):
         }
     }]
 
+    _PLAYER_URL = 'http://js.tudouui.com/bin/lingtong/PortalPlayer_177.swf'
+
     def _url_for_id(self, id, quality=None):
         info_url = "http://v2.tudou.com/f?id=" + str(id)
         if quality:
@@ -54,6 +56,10 @@ class TudouIE(InfoExtractor):
         thumbnail_url = self._search_regex(
             r",pic:\s*[\"'](.+?)[\"']", webpage, 'thumbnail URL', fatal=False)
 
+        player_url = self._search_regex(
+            r"playerUrl\s*:\s*['\"](.+?\.swf)[\"']",
+            webpage, 'player URL', default=self._PLAYER_URL)
+
         segs_json = self._search_regex(r'segs: \'(.*)\'', webpage, 'segments')
         segments = json.loads(segs_json)
         # It looks like the keys are the arguments that have to be passed as
@@ -76,6 +82,9 @@ class TudouIE(InfoExtractor):
                 'ext': ext,
                 'title': title,
                 'thumbnail': thumbnail_url,
+                'http_headers': {
+                    'Referer': player_url,
+                },
             }
             result.append(part_info)
 

diff --git a/youtube_dl/extractor/xhamster.py b/youtube_dl/extractor/xhamster.py
index b4ad513a0d18ecbae558deceb6234efaa5ce5415..97315750fd59198022a72eca53482efdd8d98e75 100644 (file)
--- a/youtube_dl/extractor/xhamster.py
+++ b/youtube_dl/extractor/xhamster.py
@@ -4,7 +4,6 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
-    ExtractorError,
     unified_strdate,
     str_to_int,
     int_or_none,
@@ -22,7 +21,7 @@ class XHamsterIE(InfoExtractor):
                 'ext': 'mp4',
                 'title': 'FemaleAgent Shy beauty takes the bait',
                 'upload_date': '20121014',
-                'uploader_id': 'Ruseful2011',
+                'uploader': 'Ruseful2011',
                 'duration': 893,
                 'age_limit': 18,
             }
@@ -34,7 +33,7 @@ class XHamsterIE(InfoExtractor):
                 'ext': 'mp4',
                 'title': 'Britney Spears  Sexy Booty',
                 'upload_date': '20130914',
-                'uploader_id': 'jojo747400',
+                'uploader': 'jojo747400',
                 'duration': 200,
                 'age_limit': 18,
             }
@@ -46,12 +45,12 @@ class XHamsterIE(InfoExtractor):
     ]
 
     def _real_extract(self, url):
-        def extract_video_url(webpage):
-            mp4 = re.search(r'<video\s+.*?file="([^"]+)".*?>', webpage)
-            if mp4 is None:
-                raise ExtractorError('Unable to extract media URL')
-            else:
-                return mp4.group(1)
+        def extract_video_url(webpage, name):
+            return self._search_regex(
+                [r'''file\s*:\s*(?P<q>["'])(?P<mp4>.+?)(?P=q)''',
+                 r'''<a\s+href=(?P<q>["'])(?P<mp4>.+?)(?P=q)\s+class=["']mp4Thumb''',
+                 r'''<video[^>]+file=(?P<q>["'])(?P<mp4>.+?)(?P=q)[^>]*>'''],
+                webpage, name, group='mp4')
 
         def is_hd(webpage):
             return '<div class=\'icon iconHD\'' in webpage
@@ -75,10 +74,14 @@ class XHamsterIE(InfoExtractor):
         if upload_date:
             upload_date = unified_strdate(upload_date)
 
-        uploader_id = self._html_search_regex(r'<a href=\'/user/[^>]+>(?P<uploader_id>[^<]+)',
-                                              webpage, 'uploader id', default='anonymous')
+        uploader = self._html_search_regex(
+            r"<a href='[^']+xhamster\.com/user/[^>]+>(?P<uploader>[^<]+)",
+            webpage, 'uploader', default='anonymous')
 
-        thumbnail = self._html_search_regex(r'<video\s+.*?poster="([^"]+)".*?>', webpage, 'thumbnail', fatal=False)
+        thumbnail = self._search_regex(
+            [r'''thumb\s*:\s*(?P<q>["'])(?P<thumbnail>.+?)(?P=q)''',
+             r'''<video[^>]+poster=(?P<q>["'])(?P<thumbnail>.+?)(?P=q)[^>]*>'''],
+            webpage, 'thumbnail', fatal=False, group='thumbnail')
 
         duration = parse_duration(self._html_search_regex(r'<span>Runtime:</span> (\d+:\d+)</div>',
                                                           webpage, 'duration', fatal=False))
@@ -97,7 +100,9 @@ class XHamsterIE(InfoExtractor):
 
         hd = is_hd(webpage)
 
-        video_url = extract_video_url(webpage)
+        format_id = 'hd' if hd else 'sd'
+
+        video_url = extract_video_url(webpage, format_id)
         formats = [{
             'url': video_url,
             'format_id': 'hd' if hd else 'sd',
@@ -108,7 +113,7 @@ class XHamsterIE(InfoExtractor):
             mrss_url = self._search_regex(r'<link rel="canonical" href="([^"]+)', webpage, 'mrss_url')
             webpage = self._download_webpage(mrss_url + '?hd', video_id, note='Downloading HD webpage')
             if is_hd(webpage):
-                video_url = extract_video_url(webpage)
+                video_url = extract_video_url(webpage, 'hd')
                 formats.append({
                     'url': video_url,
                     'format_id': 'hd',
@@ -122,7 +127,7 @@ class XHamsterIE(InfoExtractor):
             'title': title,
             'description': description,
             'upload_date': upload_date,
-            'uploader_id': uploader_id,
+            'uploader': uploader,
             'thumbnail': thumbnail,
             'duration': duration,
             'view_count': view_count,