X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/139d14b198add5b26a11ed5dca022feaa6581d98..e22397a1c5329e12f3719961c9456e378f94dbb1:/youtube_dl/extractor/tf1.py

diff --git a/youtube_dl/extractor/tf1.py b/youtube_dl/extractor/tf1.py
index 3a68eaa..55e2a07 100644
--- a/youtube_dl/extractor/tf1.py
+++ b/youtube_dl/extractor/tf1.py
@@ -2,11 +2,12 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
+from ..compat import compat_str
 
 
 class TF1IE(InfoExtractor):
     """TF1 uses the wat.tv player."""
-    _VALID_URL = r'http://(?:(?:videos|www|lci)\.tf1|www\.tfou)\.fr/.*?-(?P<id>\d+)(?:-\d+)?\.html'
+    _VALID_URL = r'https?://(?:(?:videos|www|lci)\.tf1|(?:www\.)?(?:tfou|ushuaiatv|histoire|tvbreizh))\.fr/(?:[^/]+/)*(?P<id>[^/?#.]+)'
     _TESTS = [{
         'url': 'http://videos.tf1.fr/auto-moto/citroen-grand-c4-picasso-2013-presentation-officielle-8062060.html',
         'info_dict': {
@@ -19,10 +20,11 @@ class TF1IE(InfoExtractor):
             # Sometimes wat serves the whole file with the --test option
             'skip_download': True,
         },
+        'expected_warnings': ['HTTP Error 404'],
     }, {
         'url': 'http://www.tfou.fr/chuggington/videos/le-grand-mysterioso-chuggington-7085291-739.html',
         'info_dict': {
-            'id': '12043945',
+            'id': 'le-grand-mysterioso-chuggington-7085291-739',
             'ext': 'mp4',
             'title': 'Le grand Mystérioso - Chuggington',
             'description': 'Le grand Mystérioso - Emery rêve qu\'un article lui soit consacré dans le journal.',
@@ -32,22 +34,59 @@ class TF1IE(InfoExtractor):
             # Sometimes wat serves the whole file with the --test option
             'skip_download': True,
         },
+        'skip': 'HTTP Error 410: Gone',
     }, {
         'url': 'http://www.tf1.fr/tf1/koh-lanta/videos/replay-koh-lanta-22-mai-2015.html',
         'only_matching': True,
     }, {
         'url': 'http://lci.tf1.fr/sept-a-huit/videos/sept-a-huit-du-24-mai-2015-8611550.html',
         'only_matching': True,
+    }, {
+        'url': 'http://www.tf1.fr/hd1/documentaire/videos/mylene-farmer-d-une-icone.html',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.tf1.fr/tmc/quotidien-avec-yann-barthes/videos/quotidien-premiere-partie-11-juin-2019.html',
+        'info_dict': {
+            'id': '13641379',
+            'ext': 'mp4',
+            'title': 'md5:f392bc52245dc5ad43771650c96fb620',
+            'description': 'md5:44bc54f0a21322f5b91d68e76a544eae',
+            'upload_date': '20190611',
+        },
+        'params': {
+            # Sometimes wat serves the whole file with the --test option
+            'skip_download': True,
+        },
     }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
+
         webpage = self._download_webpage(url, video_id)
-        embed_url = self._html_search_regex(
-            r'["\'](https?://www.wat.tv/embedframe/.*?)["\']', webpage, 'embed url')
-        embed_page = self._download_webpage(embed_url, video_id,
-                                            'Downloading embed player page')
-        wat_id = self._search_regex(r'UVID=(.*?)&', embed_page, 'wat id')
-        wat_info = self._download_json(
-            'http://www.wat.tv/interface/contentv3/%s' % wat_id, video_id)
-        return self.url_result(wat_info['media']['url'], 'Wat')
+
+        wat_id = None
+
+        data = self._parse_json(
+            self._search_regex(
+                r'__APOLLO_STATE__\s*=\s*({.+?})\s*(?:;|</script>)', webpage,
+                'data', default='{}'), video_id, fatal=False)
+
+        if data:
+            try:
+                wat_id = next(
+                    video.get('streamId')
+                    for key, video in data.items()
+                    if isinstance(video, dict)
+                    and video.get('slug') == video_id)
+                if not isinstance(wat_id, compat_str) or not wat_id.isdigit():
+                    wat_id = None
+            except StopIteration:
+                pass
+
+        if not wat_id:
+            wat_id = self._html_search_regex(
+                (r'(["\'])(?:https?:)?//www\.wat\.tv/embedframe/.*?(?P<id>\d{8})\1',
+                 r'(["\']?)streamId\1\s*:\s*(["\']?)(?P<id>\d+)\2'),
+                webpage, 'wat id', group='id')
+
+        return self.url_result('wat:%s' % wat_id, 'Wat')