Merge branch 'akamai_pv' of https://github.com/remitamine/youtube-dl into remitamine...

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

diff --git a/youtube_dl/extractor/nrk.py b/youtube_dl/extractor/nrk.py
index 5ed235f1ea4d22b3a4d7a8801b4162a88f0dc16b..9df20082224f84099657d2c2415cb9b2e66df8b6 100644 (file)
--- a/youtube_dl/extractor/nrk.py
+++ b/youtube_dl/extractor/nrk.py
@@ -4,7 +4,12 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..compat import (
+    compat_urlparse,
+    compat_urllib_parse_unquote,
+)
 from ..utils import (
+    determine_ext,
     ExtractorError,
     float_or_none,
     parse_duration,
@@ -47,12 +52,23 @@ class NRKIE(InfoExtractor):
             'http://v8.psapi.nrk.no/mediaelement/%s' % video_id,
             video_id, 'Downloading media JSON')
 
-        if data['usageRights']['isGeoBlocked']:
-            raise ExtractorError(
-                'NRK har ikke rettigheter til å vise dette programmet utenfor Norge',
-                expected=True)
+        media_url = data.get('mediaUrl')
+
+        if not media_url:
+            if data['usageRights']['isGeoBlocked']:
+                raise ExtractorError(
+                    'NRK har ikke rettigheter til å vise dette programmet utenfor Norge',
+                    expected=True)
 
-        video_url = data['mediaUrl'] + '?hdcore=3.5.0&plugin=aasp-3.5.0.151.81'
+        if determine_ext(media_url) == 'f4m':
+            formats = self._extract_f4m_formats(
+                media_url + '?hdcore=3.5.0&plugin=aasp-3.5.0.151.81', video_id, f4m_id='hds')
+            self._sort_formats(formats)
+        else:
+            formats = [{
+                'url': media_url,
+                'ext': 'flv',
+            }]
 
         duration = parse_duration(data.get('duration'))
 
@@ -66,17 +82,16 @@ class NRKIE(InfoExtractor):
 
         return {
             'id': video_id,
-            'url': video_url,
-            'ext': 'flv',
             'title': data['title'],
             'description': data['description'],
             'duration': duration,
             'thumbnail': thumbnail,
+            'formats': formats,
         }
 
 
 class NRKPlaylistIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?nrk\.no/(?!video)(?:[^/]+/)+(?P<id>[^/]+)'
+    _VALID_URL = r'https?://(?:www\.)?nrk\.no/(?!video|skole)(?:[^/]+/)+(?P<id>[^/]+)'
 
     _TESTS = [{
         'url': 'http://www.nrk.no/troms/gjenopplev-den-historiske-solformorkelsen-1.12270763',
@@ -115,6 +130,37 @@ class NRKPlaylistIE(InfoExtractor):
             entries, playlist_id, playlist_title, playlist_description)
 
 
+class NRKSkoleIE(InfoExtractor):
+    IE_DESC = 'NRK Skole'
+    _VALID_URL = r'https?://(?:www\.)?nrk\.no/skole/klippdetalj?.*\btopic=(?P<id>[^/?#&]+)'
+
+    _TESTS = [{
+        'url': 'http://nrk.no/skole/klippdetalj?topic=nrk:klipp/616532',
+        'md5': '04cd85877cc1913bce73c5d28a47e00f',
+        'info_dict': {
+            'id': '6021',
+            'ext': 'flv',
+            'title': 'Genetikk og eneggede tvillinger',
+            'description': 'md5:3aca25dcf38ec30f0363428d2b265f8d',
+            'duration': 399,
+        },
+    }, {
+        'url': 'http://www.nrk.no/skole/klippdetalj?topic=nrk%3Aklipp%2F616532#embed',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.nrk.no/skole/klippdetalj?topic=urn:x-mediadb:21379',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        video_id = compat_urllib_parse_unquote(self._match_id(url))
+
+        webpage = self._download_webpage(url, video_id)
+
+        nrk_id = self._search_regex(r'data-nrk-id=["\'](\d+)', webpage, 'nrk id')
+        return self.url_result('nrk:%s' % nrk_id)
+
+
 class NRKTVIE(InfoExtractor):
     IE_DESC = 'NRK TV and NRK Radio'
     _VALID_URL = r'(?P<baseurl>https?://(?:tv|radio)\.nrk(?:super)?\.no/)(?:serie/[^/]+|program)/(?P<id>[a-zA-Z]{4}\d{8})(?:/\d{2}-\d{2}-\d{4})?(?:#del=(?P<part_id>\d+))?'
@@ -122,26 +168,32 @@ class NRKTVIE(InfoExtractor):
     _TESTS = [
         {
             'url': 'https://tv.nrk.no/serie/20-spoersmaal-tv/MUHH48000314/23-05-2014',
-            'md5': 'adf2c5454fa2bf032f47a9f8fb351342',
             'info_dict': {
                 'id': 'MUHH48000314',
-                'ext': 'flv',
+                'ext': 'mp4',
                 'title': '20 spørsmål',
                 'description': 'md5:bdea103bc35494c143c6a9acdd84887a',
                 'upload_date': '20140523',
                 'duration': 1741.52,
             },
+            'params': {
+                # m3u8 download
+                'skip_download': True,
+            },
         },
         {
             'url': 'https://tv.nrk.no/program/mdfp15000514',
-            'md5': '383650ece2b25ecec996ad7b5bb2a384',
             'info_dict': {
                 'id': 'mdfp15000514',
-                'ext': 'flv',
-                'title': 'Kunnskapskanalen: Grunnlovsjubiléet - Stor ståhei for ingenting',
+                'ext': 'mp4',
+                'title': 'Grunnlovsjubiléet - Stor ståhei for ingenting',
                 'description': 'md5:654c12511f035aed1e42bdf5db3b206a',
                 'upload_date': '20140524',
-                'duration': 4605.0,
+                'duration': 4605.08,
+            },
+            'params': {
+                # m3u8 download
+                'skip_download': True,
             },
         },
         {
@@ -196,20 +248,6 @@ class NRKTVIE(InfoExtractor):
         }
     ]
 
-    def _debug_print(self, txt):
-        if self._downloader.params.get('verbose', False):
-            self.to_screen('[debug] %s' % txt)
-
-    def _get_subtitles(self, subtitlesurl, video_id, baseurl):
-        url = "%s%s" % (baseurl, subtitlesurl)
-        self._debug_print('%s: Subtitle url: %s' % (video_id, url))
-        captions = self._download_xml(
-            url, video_id, 'Downloading subtitles')
-        lang = captions.get('lang', 'no')
-        return {lang: [
-            {'ext': 'ttml', 'url': url},
-        ]}
-
     def _extract_f4m(self, manifest_url, video_id):
         return self._extract_f4m_formats(
             manifest_url + '?hdcore=3.1.1&plugin=aasp-3.1.1.69.124', video_id, f4m_id='hds')
@@ -218,7 +256,7 @@ class NRKTVIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
         part_id = mobj.group('part_id')
-        baseurl = mobj.group('baseurl')
+        base_url = mobj.group('baseurl')
 
         webpage = self._download_webpage(url, video_id)
 
@@ -278,11 +316,14 @@ class NRKTVIE(InfoExtractor):
         self._sort_formats(formats)
 
         subtitles_url = self._html_search_regex(
-            r'data-subtitlesurl[ ]*=[ ]*"([^"]+)"',
-            webpage, 'subtitle URL', default=None)
-        subtitles = None
+            r'data-subtitlesurl\s*=\s*(["\'])(?P<url>.+?)\1',
+            webpage, 'subtitle URL', default=None, group='url')
+        subtitles = {}
         if subtitles_url:
-            subtitles = self.extract_subtitles(subtitles_url, video_id, baseurl)
+            subtitles['no'] = [{
+                'ext': 'ttml',
+                'url': compat_urlparse.urljoin(base_url, subtitles_url),
+            }]
 
         return {
             'id': video_id,