Merge branch 'pornovoisines' of https://github.com/Roman2K/youtube-dl into Roman2K...

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

diff --git a/youtube_dl/extractor/npo.py b/youtube_dl/extractor/npo.py
index 54be06a4edc375f736cc038961eac67e73eecf85..5d84485714b9f360d47c8676710e9c3e6d9578c7 100644 (file)
--- a/youtube_dl/extractor/npo.py
+++ b/youtube_dl/extractor/npo.py
@@ -1,6 +1,6 @@
 from __future__ import unicode_literals
 
-from .subtitles import SubtitlesInfoExtractor
+from .common import InfoExtractor
 from ..utils import (
     fix_xml_ampersands,
     parse_duration,
@@ -11,7 +11,7 @@ from ..utils import (
 )
 
 
-class NPOBaseIE(SubtitlesInfoExtractor):
+class NPOBaseIE(InfoExtractor):
     def _get_token(self, video_id):
         token_page = self._download_webpage(
             'http://ida.omroep.nl/npoplayer/i.js',
@@ -22,7 +22,7 @@ class NPOBaseIE(SubtitlesInfoExtractor):
 
 class NPOIE(NPOBaseIE):
     IE_NAME = 'npo.nl'
-    _VALID_URL = r'https?://www\.npo\.nl/[^/]+/[^/]+/(?P<id>[^/?]+)'
+    _VALID_URL = r'https?://(?:www\.)?npo\.nl/(?!live|radio)[^/]+/[^/]+/(?P<id>[^/?]+)'
 
     _TESTS = [
         {
@@ -163,13 +163,10 @@ class NPOIE(NPOBaseIE):
 
         subtitles = {}
         if metadata.get('tt888') == 'ja':
-            subtitles['nl'] = 'http://e.omroep.nl/tt888/%s' % video_id
-
-        if self._downloader.params.get('listsubtitles', False):
-            self._list_available_subtitles(video_id, subtitles)
-            return
-
-        subtitles = self.extract_subtitles(video_id, subtitles)
+            subtitles['nl'] = [{
+                'ext': 'vtt',
+                'url': 'http://e.omroep.nl/tt888/%s' % video_id,
+            }]
 
         return {
             'id': video_id,
@@ -185,7 +182,7 @@ class NPOIE(NPOBaseIE):
 
 class NPOLiveIE(NPOBaseIE):
     IE_NAME = 'npo.nl:live'
-    _VALID_URL = r'https?://www\.npo\.nl/live/(?P<id>.+)'
+    _VALID_URL = r'https?://(?:www\.)?npo\.nl/live/(?P<id>.+)'
 
     _TEST = {
         'url': 'http://www.npo.nl/live/npo-1',
@@ -222,7 +219,8 @@ class NPOLiveIE(NPOBaseIE):
         if streams:
             for stream in streams:
                 stream_type = stream.get('type').lower()
-                if stream_type == 'ss':
+                # smooth streaming is not supported
+                if stream_type in ['ss', 'ms']:
                     continue
                 stream_info = self._download_json(
                     'http://ida.omroep.nl/aapi/?stream=%s&token=%s&type=jsonp'
@@ -233,7 +231,10 @@ class NPOLiveIE(NPOBaseIE):
                 stream_url = self._download_json(
                     stream_info['stream'], display_id,
                     'Downloading %s URL' % stream_type,
-                    transform_source=strip_jsonp)
+                    'Unable to download %s URL' % stream_type,
+                    transform_source=strip_jsonp, fatal=False)
+                if not stream_url:
+                    continue
                 if stream_type == 'hds':
                     f4m_formats = self._extract_f4m_formats(stream_url, display_id)
                     # f4m downloader downloads only piece of live stream
@@ -245,6 +246,7 @@ class NPOLiveIE(NPOBaseIE):
                 else:
                     formats.append({
                         'url': stream_url,
+                        'preference': -10,
                     })
 
         self._sort_formats(formats)
@@ -260,6 +262,84 @@ class NPOLiveIE(NPOBaseIE):
         }
 
 
+class NPORadioIE(InfoExtractor):
+    IE_NAME = 'npo.nl:radio'
+    _VALID_URL = r'https?://(?:www\.)?npo\.nl/radio/(?P<id>[^/]+)/?$'
+
+    _TEST = {
+        'url': 'http://www.npo.nl/radio/radio-1',
+        'info_dict': {
+            'id': 'radio-1',
+            'ext': 'mp3',
+            'title': 're:^NPO Radio 1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
+            'is_live': True,
+        },
+        'params': {
+            'skip_download': True,
+        }
+    }
+
+    @staticmethod
+    def _html_get_attribute_regex(attribute):
+        return r'{0}\s*=\s*\'([^\']+)\''.format(attribute)
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        title = self._html_search_regex(
+            self._html_get_attribute_regex('data-channel'), webpage, 'title')
+
+        stream = self._parse_json(
+            self._html_search_regex(self._html_get_attribute_regex('data-streams'), webpage, 'data-streams'),
+            video_id)
+
+        codec = stream.get('codec')
+
+        return {
+            'id': video_id,
+            'url': stream['url'],
+            'title': self._live_title(title),
+            'acodec': codec,
+            'ext': codec,
+            'is_live': True,
+        }
+
+
+class NPORadioFragmentIE(InfoExtractor):
+    IE_NAME = 'npo.nl:radio:fragment'
+    _VALID_URL = r'https?://(?:www\.)?npo\.nl/radio/[^/]+/fragment/(?P<id>\d+)'
+
+    _TEST = {
+        'url': 'http://www.npo.nl/radio/radio-5/fragment/174356',
+        'md5': 'dd8cc470dad764d0fdc70a9a1e2d18c2',
+        'info_dict': {
+            'id': '174356',
+            'ext': 'mp3',
+            'title': 'Jubileumconcert Willeke Alberti',
+        },
+    }
+
+    def _real_extract(self, url):
+        audio_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, audio_id)
+
+        title = self._html_search_regex(
+            r'href="/radio/[^/]+/fragment/%s" title="([^"]+)"' % audio_id,
+            webpage, 'title')
+
+        audio_url = self._search_regex(
+            r"data-streams='([^']+)'", webpage, 'audio url')
+
+        return {
+            'id': audio_id,
+            'url': audio_url,
+            'title': title,
+        }
+
+
 class TegenlichtVproIE(NPOIE):
     IE_NAME = 'tegenlicht.vpro.nl'
     _VALID_URL = r'https?://tegenlicht\.vpro\.nl/afleveringen/.*?'