X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/9a117f94b4bfe84cfe1d904d5132aefcf41511c9..e76f531201cd41dfc0ce00be28bcc5c575c7acc5:/youtube_dl/extractor/liveleak.py

diff --git a/youtube_dl/extractor/liveleak.py b/youtube_dl/extractor/liveleak.py
index cf8a2c9..857edfd 100644
--- a/youtube_dl/extractor/liveleak.py
+++ b/youtube_dl/extractor/liveleak.py
@@ -1,54 +1,119 @@
+from __future__ import unicode_literals
+
+import json
 import re
 
 from .common import InfoExtractor
-from ..utils import (
-    ExtractorError,
-)
+from ..utils import int_or_none
 
 
 class LiveLeakIE(InfoExtractor):
-
-    _VALID_URL = r'^(?:http?://)?(?:\w+\.)?liveleak\.com/view\?(?:.*?)i=(?P<video_id>[\w_]+)(?:.*)'
-    IE_NAME = u'liveleak'
-    _TEST = {
-        u'url': u'http://www.liveleak.com/view?i=757_1364311680',
-        u'file': u'757_1364311680.mp4',
-        u'md5': u'0813c2430bea7a46bf13acf3406992f4',
-        u'info_dict': {
-            u"description": u"extremely bad day for this guy..!", 
-            u"uploader": u"ljfriel2", 
-            u"title": u"Most unlucky car accident"
+    _VALID_URL = r'https?://(?:\w+\.)?liveleak\.com/view\?(?:.*?)i=(?P<id>[\w_]+)(?:.*)'
+    _TESTS = [{
+        'url': 'http://www.liveleak.com/view?i=757_1364311680',
+        'md5': '50f79e05ba149149c1b4ea961223d5b3',
+        'info_dict': {
+            'id': '757_1364311680',
+            'ext': 'flv',
+            'description': 'extremely bad day for this guy..!',
+            'uploader': 'ljfriel2',
+            'title': 'Most unlucky car accident'
+        }
+    }, {
+        'url': 'http://www.liveleak.com/view?i=f93_1390833151',
+        'md5': 'b13a29626183c9d33944e6a04f41aafc',
+        '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)',
+        }
+    }, {
+        '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,
         }
-    }
+    }, {
+        # Covers https://github.com/rg3/youtube-dl/pull/5983
+        'url': 'http://www.liveleak.com/view?i=801_1409392012',
+        'md5': '0b3bec2d888c20728ca2ad3642f0ef15',
+        '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'
+        }
+    }]
 
     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 = mobj.group('video_id')
-
+        video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
-        video_url = self._search_regex(r'file: "(.*?)",',
-            webpage, u'video URL')
+        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_title = self._html_search_regex(r'<meta property="og:title" content="(?P<title>.*?)"',
-            webpage, u'title').replace('LiveLeak.com -', '').strip()
+        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
+            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,
+                }
 
-        video_description = self._html_search_regex(r'<meta property="og:description" content="(?P<desc>.*?)"',
-            webpage, u'description', fatal=False)
+        sources_json = re.sub(r'\s([a-z]+):\s', r'"\1": ', sources_raw)
+        sources = json.loads(sources_json)
 
-        video_uploader = self._html_search_regex(r'By:.*?(\w+)</a>',
-            webpage, u'uploader', fatal=False)
+        formats = [{
+            'format_id': '%s' % i,
+            'format_note': s.get('label'),
+            'url': s['file'],
+        } for i, s in enumerate(sources)]
+        for i, s in enumerate(sources):
+            # Removing '.h264_*.mp4' gives the raw video, which is essentially
+            # the same video without the LiveLeak logo at the top (see
+            # https://github.com/rg3/youtube-dl/pull/4768)
+            orig_url = re.sub(r'\.h264_.+?\.mp4', '', s['file'])
+            if s['file'] != orig_url:
+                formats.append({
+                    'format_id': 'original-%s' % i,
+                    'format_note': s.get('label'),
+                    'url': orig_url,
+                    'preference': 1,
+                })
+        self._sort_formats(formats)
 
-        info = {
-            'id':  video_id,
-            'url': video_url,
-            'ext': 'mp4',
+        return {
+            'id': video_id,
             'title': video_title,
             'description': video_description,
-            'uploader': video_uploader
+            'uploader': video_uploader,
+            'formats': formats,
+            'age_limit': age_limit,
         }
-
-        return [info]