Update changelog.

[youtubedl] / youtube_dl / extractor / liveleak.py

diff --git a/youtube_dl/extractor/liveleak.py b/youtube_dl/extractor/liveleak.py
index d4b142ea09766b96bf38d35d21e36371b748718e..246aac576a2c6b275ed38ef614eb7a569b9e2ffb 100644 (file)
--- a/youtube_dl/extractor/liveleak.py
+++ b/youtube_dl/extractor/liveleak.py
@@ -1,44 +1,169 @@
+from __future__ import unicode_literals
+

 import re
 
 from .common import InfoExtractor
 import re
 
 from .common import InfoExtractor

-from ..utils import (
-    ExtractorError,
-)
+from ..utils import int_or_none

 
 
 class LiveLeakIE(InfoExtractor):
 
 
 class LiveLeakIE(InfoExtractor):

+    _VALID_URL = r'https?://(?:\w+\.)?liveleak\.com/view\?(?:.*?)i=(?P<id>[\w_]+)(?:.*)'
+    _TESTS = [{
+        'url': 'http://www.liveleak.com/view?i=757_1364311680',
+        'md5': '0813c2430bea7a46bf13acf3406992f4',
+        'info_dict': {
+            'id': '757_1364311680',
+            'ext': 'mp4',
+            'description': 'extremely bad day for this guy..!',
+            'uploader': 'ljfriel2',
+            'title': 'Most unlucky car accident',
+            'thumbnail': r're:^https?://.*\.jpg$'
+        }
+    }, {
+        'url': 'http://www.liveleak.com/view?i=f93_1390833151',
+        'md5': 'd3f1367d14cc3c15bf24fbfbe04b9abf',
+        'info_dict': {
+            'id': 'f93_1390833151',
+            'ext': 'mp4',
+            'description': 'German Television Channel NDR does an exclusive interview with Edward Snowden.\r\nUploaded on LiveLeak cause German Television thinks the rest of the world isn\'t intereseted in Edward Snowden.',
+            'uploader': 'ARD_Stinkt',
+            'title': 'German Television does first Edward Snowden Interview (ENGLISH)',
+            'thumbnail': r're:^https?://.*\.jpg$'
+        }
+    }, {
+        # Prochan embed
+        'url': 'http://www.liveleak.com/view?i=4f7_1392687779',
+        'md5': '42c6d97d54f1db107958760788c5f48f',
+        'info_dict': {
+            'id': '4f7_1392687779',
+            'ext': 'mp4',
+            'description': "The guy with the cigarette seems amazingly nonchalant about the whole thing...  I really hope my friends' reactions would be a bit stronger.\r\n\r\nAction-go to 0:55.",
+            'uploader': 'CapObveus',
+            'title': 'Man is Fatally Struck by Reckless Car While Packing up a Moving Truck',
+            'age_limit': 18,
+        },
+        'skip': 'Video is dead',
+    }, {
+        # Covers https://github.com/rg3/youtube-dl/pull/5983
+        # Multiple resolutions
+        'url': 'http://www.liveleak.com/view?i=801_1409392012',
+        'md5': 'c3a449dbaca5c0d1825caecd52a57d7b',
+        'info_dict': {
+            'id': '801_1409392012',
+            'ext': 'mp4',
+            'description': 'Happened on 27.7.2014. \r\nAt 0:53 you can see people still swimming at near beach.',
+            'uploader': 'bony333',
+            'title': 'Crazy Hungarian tourist films close call waterspout in Croatia',
+            'thumbnail': r're:^https?://.*\.jpg$'
+        }
+    }, {
+        # Covers https://github.com/rg3/youtube-dl/pull/10664#issuecomment-247439521
+        'url': 'http://m.liveleak.com/view?i=763_1473349649',
+        'add_ie': ['Youtube'],
+        'info_dict': {
+            'id': '763_1473349649',
+            'ext': 'mp4',
+            'title': 'Reporters and public officials ignore epidemic of black on asian violence in Sacramento | Colin Flaherty',
+            'description': 'Colin being the warrior he is and showing the injustice Asians in Sacramento are being subjected to.',
+            'uploader': 'Ziz',
+            'upload_date': '20160908',
+            'uploader_id': 'UCEbta5E_jqlZmEJsriTEtnw'
+        },
+        '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,
+    }]

 
 

-    _VALID_URL = r'^(?:http?://)?(?:\w+\.)?liveleak\.com/view\?(?:.*?)i=(?P<video_id>[\w_]+)(?:.*)'
-    IE_NAME = u'liveleak'
+    @staticmethod
+    def _extract_urls(webpage):
+        return re.findall(
+            r'<iframe[^>]+src="(https?://(?:\w+\.)?liveleak\.com/ll_embed\?[^"]*[if]=[\w_]+[^"]+)"',
+            webpage)

 
     def _real_extract(self, url):
 
     def _real_extract(self, url):

-        mobj = re.match(self._VALID_URL, url)
-        if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)

 
 

-        video_id = mobj.group('video_id')
+        video_title = self._og_search_title(webpage).replace('LiveLeak.com -', '').strip()
+        video_description = self._og_search_description(webpage)
+        video_uploader = self._html_search_regex(
+            r'By:.*?(\w+)</a>', webpage, 'uploader', fatal=False)
+        age_limit = int_or_none(self._search_regex(
+            r'you confirm that you are ([0-9]+) years and over.',
+            webpage, 'age limit', default=None))
+        video_thumbnail = self._og_search_thumbnail(webpage)

 
 

-        webpage = self._download_webpage(url, video_id)
+        entries = self._parse_html5_media_entries(url, webpage, video_id)
+        if not entries:
+            # Maybe an embed?
+            embed_url = self._search_regex(
+                r'<iframe[^>]+src="((?:https?:)?//(?:www\.)?(?:prochan|youtube)\.com/embed[^"]+)"',
+                webpage, 'embed URL')
+            return {
+                '_type': 'url_transparent',
+                'url': embed_url,
+                'id': video_id,
+                'title': video_title,
+                'description': video_description,
+                'uploader': video_uploader,
+                'age_limit': age_limit,
+            }

 
 

-        video_url = self._search_regex(r'file: "(.*?)",',
-            webpage, u'video URL')
+        for idx, info_dict in enumerate(entries):
+            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))

 
 

-        video_title = self._html_search_regex(r'<meta property="og:title" content="(?P<title>.*?)"',
-            webpage, u'title').replace('LiveLeak.com -', '').strip()
+            self._sort_formats(info_dict['formats'])

 
 

-        video_description = self._html_search_regex(r'<meta property="og:description" content="(?P<desc>.*?)"',
-            webpage, u'description', fatal=False)
+            # 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

 
 

-        video_uploader = self._html_search_regex(r'By:.*?(\w+)</a>',
-            webpage, u'uploader', fatal=False)
+            info_dict.update({
+                'title': video_title,
+                'description': video_description,
+                'uploader': video_uploader,
+                'age_limit': age_limit,
+                'thumbnail': video_thumbnail,
+            })

 
 

-        info = {
-            'id':  video_id,
-            'url': video_url,
-            'ext': 'mp4',
-            'title': video_title,
-            'description': video_description,
-            'uploader': video_uploader
-        }
+        return self.playlist_result(entries, video_id, video_title)
+
+
+class LiveLeakEmbedIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?liveleak\.com/ll_embed\?.*?\b(?P<kind>[if])=(?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):
+        mobj = re.match(self._VALID_URL, url)
+        kind, video_id = mobj.group('kind', 'id')
+
+        if kind == 'f':
+            webpage = self._download_webpage(url, video_id)
+            liveleak_url = self._search_regex(
+                r'logourl\s*:\s*(?P<q1>[\'"])(?P<url>%s)(?P=q1)' % LiveLeakIE._VALID_URL,
+                webpage, 'LiveLeak URL', group='url')
+        elif kind == 'i':
+            liveleak_url = 'http://www.liveleak.com/view?i=%s' % video_id

 
 

-        return [info]
+        return self.url_result(liveleak_url, ie=LiveLeakIE.ie_key())