[youtube] Fix extraction.

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

diff --git a/youtube_dl/extractor/playvid.py b/youtube_dl/extractor/playvid.py
index a74f8f675b660855967c3498197da9e6aae65a85..4aef186ea22b4dab1be50a0bdd6dbcbbcae1e2b1 100644 (file)
--- a/youtube_dl/extractor/playvid.py
+++ b/youtube_dl/extractor/playvid.py
@@ -3,32 +3,49 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..compat import (
+    compat_urllib_parse_unquote,
+    compat_urllib_parse_unquote_plus,
+)
 from ..utils import (
-    compat_urllib_parse,
-    determine_ext,
+    clean_html,
+    ExtractorError,
 )
 
-class PlayvidIE(InfoExtractor):
 
-    _VALID_URL = r'^(?:https?://)?www\.playvid\.com/watch(\?v=|/)(?P<id>.+?)(#|$)'
-    _TEST = {
-        'url': 'http://www.playvid.com/watch/agbDDi7WZTV',
-        'file': 'agbDDi7WZTV.mp4',
-        'md5': '44930f8afa616efdf9482daf4fe53e1e',
+class PlayvidIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?playvid\.com/watch(\?v=|/)(?P<id>.+?)(?:#|$)'
+    _TESTS = [{
+        'url': 'http://www.playvid.com/watch/RnmBNgtrrJu',
+        'md5': 'ffa2f6b2119af359f544388d8c01eb6c',
         'info_dict': {
-            'title': 'Michelle Lewin in Miami Beach',
-            'duration': 240,
+            'id': 'RnmBNgtrrJu',
+            'ext': 'mp4',
+            'title': 'md5:9256d01c6317e3f703848b5906880dc8',
+            'duration': 82,
             'age_limit': 18,
-        }
-    }
+        },
+        'skip': 'Video removed due to ToS',
+    }, {
+        'url': 'http://www.playvid.com/watch/hwb0GpNkzgH',
+        'md5': '39d49df503ad7b8f23a4432cbf046477',
+        'info_dict': {
+            'id': 'hwb0GpNkzgH',
+            'ext': 'mp4',
+            'title': 'Ellen Euro Cutie Blond Takes a Sexy Survey Get Facial in The Park',
+            'age_limit': 18,
+            'thumbnail': r're:^https?://.*\.jpg$',
+        },
+    }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-
+        video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
-        self.report_extraction(video_id)
+        m_error = re.search(
+            r'<div class="block-error">\s*<div class="heading">\s*<div>(?P<msg>.+?)</div>\s*</div>', webpage)
+        if m_error:
+            raise ExtractorError(clean_html(m_error.group('msg')), expected=True)
 
         video_title = None
         duration = None
@@ -36,38 +53,35 @@ class PlayvidIE(InfoExtractor):
         formats = []
 
         # most of the information is stored in the flashvars
-        flashvars_match = re.search(r'flashvars="(.+?)"',webpage)
-
-        if flashvars_match:
-            infos = compat_urllib_parse.unquote(flashvars_match.group(1)).split(r'&amp;')
-            for info in infos:
-                videovars_match = re.match(r'^video_vars\[(.+?)\]=(.+?)$',info)
-                if videovars_match:
-                    key = videovars_match.group(1)
-                    val = videovars_match.group(2)
+        flashvars = self._html_search_regex(
+            r'flashvars="(.+?)"', webpage, 'flashvars')
 
-                    if key == 'title':
-                        video_title = val.replace('+',' ')
-                    if key == 'duration':
-                        try:
-                            duration = val
-                        except ValueError:
-                            duration = None
-                    if key == 'big_thumb':
-                        video_thumbnail = val
+        infos = compat_urllib_parse_unquote(flashvars).split(r'&')
+        for info in infos:
+            videovars_match = re.match(r'^video_vars\[(.+?)\]=(.+?)$', info)
+            if videovars_match:
+                key = videovars_match.group(1)
+                val = videovars_match.group(2)
 
-                    videourl_match = re.match(r'^video_urls\]\[(?P<resolution>\d+)p',key)
-                    if videourl_match:
-                        resolution = int(videourl_match.group('resolution'))
-                        formats.append({
-                            'resolution': resolution,            # 360, 480, ...
-                            'ext': determine_ext(val),
-                            'url': val
-                        })
+                if key == 'title':
+                    video_title = compat_urllib_parse_unquote_plus(val)
+                if key == 'duration':
+                    try:
+                        duration = int(val)
+                    except ValueError:
+                        pass
+                if key == 'big_thumb':
+                    video_thumbnail = val
 
-        # fatal error, if no download url is found
-        if len(formats) == 0:
-            raise ExtractorError,'no video url found'
+                videourl_match = re.match(
+                    r'^video_urls\]\[(?P<resolution>[0-9]+)p', key)
+                if videourl_match:
+                    height = int(videourl_match.group('resolution'))
+                    formats.append({
+                        'height': height,
+                        'url': val,
+                    })
+        self._sort_formats(formats)
 
         # Extract title - should be in the flashvars; if not, look elsewhere
         if video_title is None: