Update changelog with bugs being fixed.

[youtubedl] / youtube_dl / extractor / izlesene.py

diff --git a/youtube_dl/extractor/izlesene.py b/youtube_dl/extractor/izlesene.py
index b1d72177d5acef2c48a82f7df18081005199b47e..f8fca6c8f47c9978e78d1c2c76d0dcb84a8d7f9d 100644 (file)
--- a/youtube_dl/extractor/izlesene.py
+++ b/youtube_dl/extractor/izlesene.py
@@ -1,10 +1,11 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
 # coding: utf-8
 from __future__ import unicode_literals
 

-import re
-

 from .common import InfoExtractor
 from .common import InfoExtractor

-from ..compat import compat_urllib_parse_unquote
+from ..compat import (
+    compat_str,
+    compat_urllib_parse_unquote,
+)

 from ..utils import (
     determine_ext,
     float_or_none,
 from ..utils import (
     determine_ext,
     float_or_none,


@@ -57,12 +58,33 @@ class IzleseneIE(InfoExtractor):
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 

-        url = 'http://www.izlesene.com/video/%s' % video_id
-        webpage = self._download_webpage(url, video_id)
+        webpage = self._download_webpage('http://www.izlesene.com/video/%s' % video_id, video_id)
+
+        video = self._parse_json(
+            self._search_regex(
+                r'videoObj\s*=\s*({.+?})\s*;\s*\n', webpage, 'streams'),
+            video_id)
+
+        title = video.get('videoTitle') or self._og_search_title(webpage)
+
+        formats = []
+        for stream in video['media']['level']:
+            source_url = stream.get('source')
+            if not source_url or not isinstance(source_url, compat_str):
+                continue
+            ext = determine_ext(url, 'mp4')
+            quality = stream.get('value')
+            height = int_or_none(quality)
+            formats.append({
+                'format_id': '%sp' % quality if quality else 'sd',
+                'url': compat_urllib_parse_unquote(source_url),
+                'ext': ext,
+                'height': height,
+            })
+        self._sort_formats(formats)

 
 

-        title = self._og_search_title(webpage)

         description = self._og_search_description(webpage, default=None)
         description = self._og_search_description(webpage, default=None)

-        thumbnail = self._proto_relative_url(
+        thumbnail = video.get('posterURL') or self._proto_relative_url(

             self._og_search_thumbnail(webpage), scheme='http:')
 
         uploader = self._html_search_regex(
             self._og_search_thumbnail(webpage), scheme='http:')
 
         uploader = self._html_search_regex(


@@ -71,41 +93,15 @@ class IzleseneIE(InfoExtractor):
         timestamp = parse_iso8601(self._html_search_meta(
             'uploadDate', webpage, 'upload date'))
 
         timestamp = parse_iso8601(self._html_search_meta(
             'uploadDate', webpage, 'upload date'))
 

-        duration = float_or_none(self._html_search_regex(
-            r'"videoduration"\s*:\s*"([^"]+)"',
-            webpage, 'duration', fatal=False), scale=1000)
+        duration = float_or_none(video.get('duration') or self._html_search_regex(
+            r'videoduration["\']?\s*=\s*(["\'])(?P<value>(?:(?!\1).)+)\1',
+            webpage, 'duration', fatal=False, group='value'), scale=1000)

 
         view_count = str_to_int(get_element_by_id('videoViewCount', webpage))
         comment_count = self._html_search_regex(
             r'comment_count\s*=\s*\'([^\']+)\';',
             webpage, 'comment_count', fatal=False)
 
 
         view_count = str_to_int(get_element_by_id('videoViewCount', webpage))
         comment_count = self._html_search_regex(
             r'comment_count\s*=\s*\'([^\']+)\';',
             webpage, 'comment_count', fatal=False)
 

-        content_url = self._html_search_meta(
-            'contentURL', webpage, 'content URL', fatal=False)
-        ext = determine_ext(content_url, 'mp4')
-
-        # Might be empty for some videos.
-        streams = self._html_search_regex(
-            r'"qualitylevel"\s*:\s*"([^"]+)"', webpage, 'streams', default='')
-
-        formats = []
-        if streams:
-            for stream in streams.split('|'):
-                quality, url = re.search(r'\[(\w+)\](.+)', stream).groups()
-                formats.append({
-                    'format_id': '%sp' % quality if quality else 'sd',
-                    'url': compat_urllib_parse_unquote(url),
-                    'ext': ext,
-                })
-        else:
-            stream_url = self._search_regex(
-                r'"streamurl"\s*:\s*"([^"]+)"', webpage, 'stream URL')
-            formats.append({
-                'format_id': 'sd',
-                'url': compat_urllib_parse_unquote(stream_url),
-                'ext': ext,
-            })
-

         return {
             'id': video_id,
             'title': title,
         return {
             'id': video_id,
             'title': title,