Fix W504 and disable W503 (closes #20863)

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

diff --git a/youtube_dl/extractor/motherless.py b/youtube_dl/extractor/motherless.py
index 90ed91ba6353e12964882aefb65236c91942bafd..43fd70f112005a893377c8e5cf489291fb8cc812 100644 (file)
--- a/youtube_dl/extractor/motherless.py
+++ b/youtube_dl/extractor/motherless.py
@@ -77,8 +77,11 @@ class MotherlessIE(InfoExtractor):
 
         title = self._html_search_regex(
             r'id="view-upload-title">\s+([^<]+)<', webpage, 'title')
-        video_url = self._html_search_regex(
-            r'setup\(\{\s+"file".+: "([^"]+)",', webpage, 'video URL')
+        video_url = (self._html_search_regex(
+            (r'setup\(\{\s*["\']file["\']\s*:\s*(["\'])(?P<url>(?:(?!\1).)+)\1',
+             r'fileurl\s*=\s*(["\'])(?P<url>(?:(?!\1).)+)\1'),
+            webpage, 'video URL', default=None, group='url')
+            or 'http://cdn4.videos.motherlessmedia.com/videos/%s.mp4?fs=opencloud' % video_id)
         age_limit = self._rta_search(webpage)
         view_count = str_to_int(self._html_search_regex(
             r'<strong>Views</strong>\s+([^<]+)<',
@@ -120,7 +123,7 @@ class MotherlessIE(InfoExtractor):
 
 
 class MotherlessGroupIE(InfoExtractor):
-    _VALID_URL = 'https?://(?:www\.)?motherless\.com/gv?/(?P<id>[a-z0-9_]+)'
+    _VALID_URL = r'https?://(?:www\.)?motherless\.com/gv?/(?P<id>[a-z0-9_]+)'
     _TESTS = [{
         'url': 'http://motherless.com/g/movie_scenes',
         'info_dict': {
@@ -148,14 +151,27 @@ class MotherlessGroupIE(InfoExtractor):
                 else super(MotherlessGroupIE, cls).suitable(url))
 
     def _extract_entries(self, webpage, base):
-        return [
-            self.url_result(
-                compat_urlparse.urljoin(base, video_path),
-                MotherlessIE.ie_key(), video_title=title)
-            for video_path, title in orderedSet(re.findall(
-                r'href="/([^"]+)"[^>]+>\s+<img[^>]+alt="[^-]+-\s([^"]+)"',
-                webpage))
-        ]
+        entries = []
+        for mobj in re.finditer(
+                r'href="(?P<href>/[^"]+)"[^>]*>(?:\s*<img[^>]+alt="[^-]+-\s(?P<title>[^"]+)")?',
+                webpage):
+            video_url = compat_urlparse.urljoin(base, mobj.group('href'))
+            if not MotherlessIE.suitable(video_url):
+                continue
+            video_id = MotherlessIE._match_id(video_url)
+            title = mobj.group('title')
+            entries.append(self.url_result(
+                video_url, ie=MotherlessIE.ie_key(), video_id=video_id,
+                video_title=title))
+        # Alternative fallback
+        if not entries:
+            entries = [
+                self.url_result(
+                    compat_urlparse.urljoin(base, '/' + entry_id),
+                    ie=MotherlessIE.ie_key(), video_id=entry_id)
+                for entry_id in orderedSet(re.findall(
+                    r'data-codename=["\']([A-Z0-9]+)', webpage))]
+        return entries
 
     def _real_extract(self, url):
         group_id = self._match_id(url)