[youtube] Fix extraction.

[youtube-dl] / youtube_dl / extractor / telebruxelles.py

diff --git a/youtube_dl/extractor/telebruxelles.py b/youtube_dl/extractor/telebruxelles.py
index a3d05f97d681b6cb4da6adf179a4f0a5744e5123..a0353fe3ae30ab089d846f8360e89d21602c1446 100644 (file)
--- a/youtube_dl/extractor/telebruxelles.py
+++ b/youtube_dl/extractor/telebruxelles.py
@@ -1,37 +1,46 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 
 
 class TeleBruxellesIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?telebruxelles\.be/(news|sport|dernier-jt)/?(?P<id>[^/#?]+)'
+    _VALID_URL = r'https?://(?:www\.)?(?:telebruxelles|bx1)\.be/(?:[^/]+/)*(?P<id>[^/#?]+)'
     _TESTS = [{
-        'url': 'http://www.telebruxelles.be/news/auditions-devant-parlement-francken-galant-tres-attendus/',
-        'md5': '59439e568c9ee42fb77588b2096b214f',
+        'url': 'http://bx1.be/news/que-risque-lauteur-dune-fausse-alerte-a-la-bombe/',
+        'md5': 'a2a67a5b1c3e8c9d33109b902f474fd9',
         'info_dict': {
-            'id': '11942',
-            'display_id': 'auditions-devant-parlement-francken-galant-tres-attendus',
-            'ext': 'flv',
-            'title': 'Parlement : Francken et Galant répondent aux interpellations de l’opposition',
-            'description': 're:Les auditions des ministres se poursuivent*'
-        },
-        'params': {
-            'skip_download': 'requires rtmpdump'
+            'id': '158856',
+            'display_id': 'que-risque-lauteur-dune-fausse-alerte-a-la-bombe',
+            'ext': 'mp4',
+            'title': 'Que risque l’auteur d’une fausse alerte à la bombe ?',
+            'description': 'md5:3cf8df235d44ebc5426373050840e466',
         },
     }, {
-        'url': 'http://www.telebruxelles.be/sport/basket-brussels-bat-mons-80-74/',
-        'md5': '181d3fbdcf20b909309e5aef5c6c6047',
+        'url': 'http://bx1.be/sport/futsal-schaerbeek-sincline-5-3-a-thulin/',
+        'md5': 'dfe07ecc9c153ceba8582ac912687675',
         'info_dict': {
-            'id': '10091',
-            'display_id': 'basket-brussels-bat-mons-80-74',
-            'ext': 'flv',
-            'title': 'Basket : le Brussels bat Mons 80-74',
-            'description': 're:^Ils l\u2019on fait ! En basket, le B*',
-        },
-        'params': {
-            'skip_download': 'requires rtmpdump'
+            'id': '158433',
+            'display_id': 'futsal-schaerbeek-sincline-5-3-a-thulin',
+            'ext': 'mp4',
+            'title': 'Futsal : Schaerbeek s’incline 5-3 à Thulin',
+            'description': 'md5:fd013f1488d5e2dceb9cebe39e2d569b',
         },
+    }, {
+        'url': 'http://bx1.be/emission/bxenf1-gastronomie/',
+        'only_matching': True,
+    }, {
+        'url': 'https://bx1.be/berchem-sainte-agathe/personnel-carrefour-de-berchem-sainte-agathe-inquiet/',
+        'only_matching': True,
+    }, {
+        'url': 'https://bx1.be/dernier-jt/',
+        'only_matching': True,
+    }, {
+        # live stream
+        'url': 'https://bx1.be/lives/direct-tv/',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -39,22 +48,29 @@ class TeleBruxellesIE(InfoExtractor):
         webpage = self._download_webpage(url, display_id)
 
         article_id = self._html_search_regex(
-            r"<article id=\"post-(\d+)\"", webpage, 'article ID')
+            r'<article[^>]+\bid=["\']post-(\d+)', webpage, 'article ID', default=None)
         title = self._html_search_regex(
-            r'<h1 class=\"entry-title\">(.*?)</h1>', webpage, 'title')
-        description = self._og_search_description(webpage)
+            r'<h1[^>]*>(.+?)</h1>', webpage, 'title',
+            default=None) or self._og_search_title(webpage)
+        description = self._og_search_description(webpage, default=None)
 
         rtmp_url = self._html_search_regex(
-            r"file: \"(rtmp://\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3}:\d{1,5}/vod/mp4:\" \+ \"\w+\" \+ \".mp4)\"",
+            r'file["\']?\s*:\s*"(r(?:tm|mt)ps?://[^/]+/(?:vod/mp4:"\s*\+\s*"[^"]+"\s*\+\s*"\.mp4|stream/live))"',
             webpage, 'RTMP url')
-        rtmp_url = rtmp_url.replace("\" + \"", "")
+        # Yes, they have a typo in scheme name for live stream URLs (e.g.
+        # https://bx1.be/lives/direct-tv/)
+        rtmp_url = re.sub(r'^rmtp', 'rtmp', rtmp_url)
+        rtmp_url = re.sub(r'"\s*\+\s*"', '', rtmp_url)
+        formats = self._extract_wowza_formats(rtmp_url, article_id or display_id)
+        self._sort_formats(formats)
+
+        is_live = 'stream/live' in rtmp_url
 
         return {
-            'id': article_id,
+            'id': article_id or display_id,
             'display_id': display_id,
-            'title': title,
+            'title': self._live_title(title) if is_live else title,
             'description': description,
-            'url': rtmp_url,
-            'ext': 'flv',
-            'rtmp_live': True  # if rtmpdump is not called with "--live" argument, the download is blocked and can be completed
+            'formats': formats,
+            'is_live': is_live,
         }