Set upstream metadata fields: Bug-Database, Bug-Submit, Repository, Repository-Browse.

[youtubedl] / youtube_dl / extractor / liveleak.py

diff --git a/youtube_dl/extractor/liveleak.py b/youtube_dl/extractor/liveleak.py
index c54519636d064ea7a0d9b0ac543c5ec190f60306..4ac437c8b7a4b4affd6a9a2175dc807e4e964f7f 100644 (file)
--- a/youtube_dl/extractor/liveleak.py
+++ b/youtube_dl/extractor/liveleak.py
@@ -7,7 +7,7 @@ from ..utils import int_or_none
 
 
 class LiveLeakIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:\w+\.)?liveleak\.com/view\?(?:.*?)i=(?P<id>[\w_]+)(?:.*)'
+    _VALID_URL = r'https?://(?:\w+\.)?liveleak\.com/view\?.*?\b[it]=(?P<id>[\w_]+)'
     _TESTS = [{
         'url': 'http://www.liveleak.com/view?i=757_1364311680',
         'md5': '0813c2430bea7a46bf13acf3406992f4',
@@ -44,7 +44,7 @@ class LiveLeakIE(InfoExtractor):
         },
         'skip': 'Video is dead',
     }, {
-        # Covers https://github.com/rg3/youtube-dl/pull/5983
+        # Covers https://github.com/ytdl-org/youtube-dl/pull/5983
         # Multiple resolutions
         'url': 'http://www.liveleak.com/view?i=801_1409392012',
         'md5': 'c3a449dbaca5c0d1825caecd52a57d7b',
@@ -57,7 +57,7 @@ class LiveLeakIE(InfoExtractor):
             'thumbnail': r're:^https?://.*\.jpg$'
         }
     }, {
-        # Covers https://github.com/rg3/youtube-dl/pull/10664#issuecomment-247439521
+        # Covers https://github.com/ytdl-org/youtube-dl/pull/10664#issuecomment-247439521
         'url': 'http://m.liveleak.com/view?i=763_1473349649',
         'add_ie': ['Youtube'],
         'info_dict': {
@@ -72,15 +72,27 @@ class LiveLeakIE(InfoExtractor):
         'params': {
             'skip_download': True,
         },
+    }, {
+        'url': 'https://www.liveleak.com/view?i=677_1439397581',
+        'info_dict': {
+            'id': '677_1439397581',
+            'title': 'Fuel Depot in China Explosion caught on video',
+        },
+        'playlist_count': 3,
+    }, {
+        'url': 'https://www.liveleak.com/view?t=HvHi_1523016227',
+        'only_matching': True,
+    }, {
+        # No original video
+        'url': 'https://www.liveleak.com/view?t=C26ZZ_1558612804',
+        'only_matching': True,
     }]
 
     @staticmethod
-    def _extract_url(webpage):
-        mobj = re.search(
-            r'<iframe[^>]+src="https?://(?:\w+\.)?liveleak\.com/ll_embed\?(?:.*?)i=(?P<id>[\w_]+)(?:.*)',
+    def _extract_urls(webpage):
+        return re.findall(
+            r'<iframe[^>]+src="(https?://(?:\w+\.)?liveleak\.com/ll_embed\?[^"]*[ift]=[\w_]+[^"]+)"',
             webpage)
-        if mobj:
-            return 'http://www.liveleak.com/view?i=%s' % mobj.group('id')
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
@@ -111,22 +123,69 @@ class LiveLeakIE(InfoExtractor):
                 'age_limit': age_limit,
             }
 
-        info_dict = entries[0]
+        for idx, info_dict in enumerate(entries):
+            formats = []
+            for a_format in info_dict['formats']:
+                if not a_format.get('height'):
+                    a_format['height'] = int_or_none(self._search_regex(
+                        r'([0-9]+)p\.mp4', a_format['url'], 'height label',
+                        default=None))
+                formats.append(a_format)
 
-        for a_format in info_dict['formats']:
-            if not a_format.get('height'):
-                a_format['height'] = self._search_regex(
-                    r'([0-9]+)p\.mp4', a_format['url'], 'height label', default=None)
+                # Removing '.*.mp4' gives the raw video, which is essentially
+                # the same video without the LiveLeak logo at the top (see
+                # https://github.com/ytdl-org/youtube-dl/pull/4768)
+                orig_url = re.sub(r'\.mp4\.[^.]+', '', a_format['url'])
+                if a_format['url'] != orig_url:
+                    format_id = a_format.get('format_id')
+                    format_id = 'original' + ('-' + format_id if format_id else '')
+                    if self._is_valid_url(orig_url, video_id, format_id):
+                        formats.append({
+                            'format_id': format_id,
+                            'url': orig_url,
+                            'preference': 1,
+                        })
+            self._sort_formats(formats)
+            info_dict['formats'] = formats
 
-        self._sort_formats(info_dict['formats'])
+            # Don't append entry ID for one-video pages to keep backward compatibility
+            if len(entries) > 1:
+                info_dict['id'] = '%s_%s' % (video_id, idx + 1)
+            else:
+                info_dict['id'] = video_id
+
+            info_dict.update({
+                'title': video_title,
+                'description': video_description,
+                'uploader': video_uploader,
+                'age_limit': age_limit,
+                'thumbnail': video_thumbnail,
+            })
+
+        return self.playlist_result(entries, video_id, video_title)
+
+
+class LiveLeakEmbedIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?liveleak\.com/ll_embed\?.*?\b(?P<kind>[ift])=(?P<id>[\w_]+)'
+
+    # See generic.py for actual test cases
+    _TESTS = [{
+        'url': 'https://www.liveleak.com/ll_embed?i=874_1459135191',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.liveleak.com/ll_embed?f=ab065df993c1',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        kind, video_id = re.match(self._VALID_URL, url).groups()
 
-        info_dict.update({
-            'id': video_id,
-            'title': video_title,
-            'description': video_description,
-            'uploader': video_uploader,
-            'age_limit': age_limit,
-            'thumbnail': video_thumbnail,
-        })
+        if kind == 'f':
+            webpage = self._download_webpage(url, video_id)
+            liveleak_url = self._search_regex(
+                r'(?:logourl\s*:\s*|window\.open\()(?P<q1>[\'"])(?P<url>%s)(?P=q1)' % LiveLeakIE._VALID_URL,
+                webpage, 'LiveLeak URL', group='url')
+        else:
+            liveleak_url = 'http://www.liveleak.com/view?%s=%s' % (kind, video_id)
 
-        return info_dict
+        return self.url_result(liveleak_url, ie=LiveLeakIE.ie_key())