Update upstream source from tag 'upstream/2019.09.28'

[youtubedl] / youtube_dl / extractor / liveleak.py

diff --git a/youtube_dl/extractor/liveleak.py b/youtube_dl/extractor/liveleak.py
index b04be1e8cfda94addca26a1d1e3731ce61519dc1..4ac437c8b7a4b4affd6a9a2175dc807e4e964f7f 100644 (file)
--- a/youtube_dl/extractor/liveleak.py
+++ b/youtube_dl/extractor/liveleak.py
@@ -1,6 +1,5 @@
 from __future__ import unicode_literals
 
 from __future__ import unicode_literals
 

-import json

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


@@ -8,7 +7,7 @@ from ..utils import int_or_none
 
 
 class LiveLeakIE(InfoExtractor):
 
 
 class LiveLeakIE(InfoExtractor):

-    _VALID_URL = r'^(?:http://)?(?:\w+\.)?liveleak\.com/view\?(?:.*?)i=(?P<video_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',
     _TESTS = [{
         'url': 'http://www.liveleak.com/view?i=757_1364311680',
         'md5': '0813c2430bea7a46bf13acf3406992f4',


@@ -17,7 +16,8 @@ class LiveLeakIE(InfoExtractor):
             'ext': 'mp4',
             'description': 'extremely bad day for this guy..!',
             'uploader': 'ljfriel2',
             'ext': 'mp4',
             'description': 'extremely bad day for this guy..!',
             'uploader': 'ljfriel2',

-            'title': 'Most unlucky car accident'
+            'title': 'Most unlucky car accident',
+            'thumbnail': r're:^https?://.*\.jpg$'

         }
     }, {
         'url': 'http://www.liveleak.com/view?i=f93_1390833151',
         }
     }, {
         'url': 'http://www.liveleak.com/view?i=f93_1390833151',


@@ -28,8 +28,10 @@ class LiveLeakIE(InfoExtractor):
             '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)',
             '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': {
         'url': 'http://www.liveleak.com/view?i=4f7_1392687779',
         'md5': '42c6d97d54f1db107958760788c5f48f',
         'info_dict': {


@@ -39,12 +41,61 @@ class LiveLeakIE(InfoExtractor):
             'uploader': 'CapObveus',
             'title': 'Man is Fatally Struck by Reckless Car While Packing up a Moving Truck',
             'age_limit': 18,
             '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/ytdl-org/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/ytdl-org/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,
+    }, {
+        '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_urls(webpage):
+        return re.findall(
+            r'<iframe[^>]+src="(https?://(?:\w+\.)?liveleak\.com/ll_embed\?[^"]*[ift]=[\w_]+[^"]+)"',
+            webpage)
+

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

-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('video_id')
+        video_id = self._match_id(url)

         webpage = self._download_webpage(url, video_id)
 
         video_title = self._og_search_title(webpage).replace('LiveLeak.com -', '').strip()
         webpage = self._download_webpage(url, video_id)
 
         video_title = self._og_search_title(webpage).replace('LiveLeak.com -', '').strip()


@@ -54,43 +105,87 @@ class LiveLeakIE(InfoExtractor):
         age_limit = int_or_none(self._search_regex(
             r'you confirm that you are ([0-9]+) years and over.',
             webpage, 'age limit', default=None))
         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)
+
+        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,
+            }
+
+        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)
+
+                # 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

 
 

-        sources_raw = self._search_regex(
-            r'(?s)sources:\s*(\[.*?\]),', webpage, 'video URLs', default=None)
-        if sources_raw is None:
-            alt_source = self._search_regex(
-                r'(file: ".*?"),', webpage, 'video URL', default=None)
-            if alt_source:
-                sources_raw = '[{ %s}]' % alt_source
+            # 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:
             else:

-                # Maybe an embed?
-                embed_url = self._search_regex(
-                    r'<iframe[^>]+src="(http://www.prochan.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,
-                }
-
-        sources_json = re.sub(r'\s([a-z]+):\s', r'"\1": ', sources_raw)
-        sources = json.loads(sources_json)
-
-        formats = [{
-            'format_note': s.get('label'),
-            'url': s['file'],
-        } for s in sources]
-        self._sort_formats(formats)
-
-        return {
-            'id': video_id,
-            'title': video_title,
-            'description': video_description,
-            'uploader': video_uploader,
-            'formats': formats,
-            'age_limit': age_limit,
-        }
+                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()
+
+        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 self.url_result(liveleak_url, ie=LiveLeakIE.ie_key())