Merge branch 'remitamine-nowness'

youtube_dl/extractor/__init__.py

@@ -405,7 +405,11 @@ from .normalboots import NormalbootsIE
 from .nosvideo import NosVideoIE
 from .nova import NovaIE
 from .novamov import NovaMovIE
-from .nowness import NownessIE
+from .nowness import (
+    NownessIE,
+    NownessPlaylistIE,
+    NownessSeriesIE,
+)
 from .nowtv import NowTVIE
 from .nowvideo import NowVideoIE
 from .npo import (

youtube_dl/extractor/nowness.py

@@ -1,19 +1,55 @@
 # encoding: utf-8
 from __future__ import unicode_literals
 
-import re
-
 from .brightcove import BrightcoveIE
 from .common import InfoExtractor
 from ..utils import ExtractorError
+from ..compat import (
+    compat_str,
+    compat_urllib_request,
+)
+
 
+class NownessBaseIE(InfoExtractor):
+    def _extract_url_result(self, post):
+        if post['type'] == 'video':
+            for media in post['media']:
+                if media['type'] == 'video':
+                    video_id = media['content']
+                    source = media['source']
+                    if source == 'brightcove':
+                        player_code = self._download_webpage(
+                            'http://www.nowness.com/iframe?id=%s' % video_id, video_id,
+                            note='Downloading player JavaScript',
+                            errnote='Unable to download player JavaScript')
+                        bc_url = BrightcoveIE._extract_brightcove_url(player_code)
+                        if bc_url is None:
+                            raise ExtractorError('Could not find player definition')
+                        return self.url_result(bc_url, 'Brightcove')
+                    elif source == 'vimeo':
+                        return self.url_result('http://vimeo.com/%s' % video_id, 'Vimeo')
+                    elif source == 'youtube':
+                        return self.url_result(video_id, 'Youtube')
+                    elif source == 'cinematique':
+                        # youtube-dl currently doesn't support cinematique
+                        # return self.url_result('http://cinematique.com/embed/%s' % video_id, 'Cinematique')
+                        pass
 
-class NownessIE(InfoExtractor):
+    def _api_request(self, url, request_path):
-    _VALID_URL = r'https?://(?:(?:www|cn)\.)?nowness\.com/[^?#]*?/(?P<id>[0-9]+)/(?P<slug>[^/]+?)(?:$|[?#])'
+        display_id = self._match_id(url)
+        request = compat_urllib_request.Request(
+            'http://api.nowness.com/api/' + request_path % display_id,
+            headers={
+                'X-Nowness-Language': 'zh-cn' if 'cn.nowness.com' in url else 'en-us',
+            })
+        return display_id, self._download_json(request, display_id)
 
-    _TESTS = [
+
-        {
+class NownessIE(NownessBaseIE):
-            'url': 'http://www.nowness.com/day/2013/6/27/3131/candor--the-art-of-gesticulation',
+    IE_NAME = 'nowness'
+    _VALID_URL = r'https?://(?:(?:www|cn)\.)?nowness\.com/(?:story|(?:series|category)/[^/]+)/(?P<id>[^/]+?)(?:$|[?#])'
+    _TESTS = [{
+        'url': 'https://www.nowness.com/story/candor-the-art-of-gesticulation',
         'md5': '068bc0202558c2e391924cb8cc470676',
         'info_dict': {
             'id': '2520295746001',
@@ -23,9 +59,8 @@ class NownessIE(InfoExtractor):
             'thumbnail': 're:^https?://.*\.jpg',
             'uploader': 'Nowness',
         }
-        },
+    }, {
-        {
+        'url': 'https://cn.nowness.com/story/kasper-bjorke-ft-jaakko-eino-kalevi-tnr',
-            'url': 'http://cn.nowness.com/day/2014/8/7/4069/kasper-bj-rke-ft-jaakko-eino-kalevi--tnr',
         'md5': 'e79cf125e387216f86b2e0a5b5c63aa3',
         'info_dict': {
             'id': '3716354522001',
@@ -35,30 +70,51 @@ class NownessIE(InfoExtractor):
             'thumbnail': 're:^https?://.*\.jpg',
             'uploader': 'Nowness',
         }
-        },
+    }]
-    ]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        _, post = self._api_request(url, 'post/getBySlug/%s')
-        video_id = mobj.group('slug')
+        return self._extract_url_result(post)
 
-        webpage = self._download_webpage(url, video_id)
-        player_url = self._search_regex(
-            r'"([^"]+/content/issue-[0-9.]+.js)"', webpage, 'player URL')
-        real_id = self._search_regex(
-            r'\sdata-videoId="([0-9]+)"', webpage, 'internal video ID')
 
-        player_code = self._download_webpage(
+class NownessPlaylistIE(NownessBaseIE):
-            player_url, video_id,
+    IE_NAME = 'nowness:playlist'
-            note='Downloading player JavaScript',
+    _VALID_URL = r'https?://(?:(?:www|cn)\.)?nowness\.com/playlist/(?P<id>\d+)'
-            errnote='Player download failed')
+    _TEST = {
-        player_code = player_code.replace("'+d+'", real_id)
+        'url': 'https://www.nowness.com/playlist/3286/i-guess-thats-why-they-call-it-the-blues',
+        'info_dict': {
+            'id': '3286',
+        },
+        'playlist_mincount': 8,
+    }
 
-        bc_url = BrightcoveIE._extract_brightcove_url(player_code)
+    def _real_extract(self, url):
-        if bc_url is None:
+        playlist_id, playlist = self._api_request(url, 'post?PlaylistId=%s')
-            raise ExtractorError('Could not find player definition')
+        entries = [self._extract_url_result(item) for item in playlist['items']]
-        return {
+        return self.playlist_result(entries, playlist_id)
-            '_type': 'url',
+
-            'url': bc_url,
+
-            'ie_key': 'Brightcove',
+class NownessSeriesIE(NownessBaseIE):
+    IE_NAME = 'nowness:series'
+    _VALID_URL = r'https?://(?:(?:www|cn)\.)?nowness\.com/series/(?P<id>[^/]+?)(?:$|[?#])'
+    _TEST = {
+        'url': 'https://www.nowness.com/series/60-seconds',
+        'info_dict': {
+            'id': '60',
+            'title': '60 Seconds',
+            'description': 'One-minute wisdom in a new NOWNESS series',
+        },
+        'playlist_mincount': 4,
     }
+
+    def _real_extract(self, url):
+        display_id, series = self._api_request(url, 'series/getBySlug/%s')
+        entries = [self._extract_url_result(post) for post in series['posts']]
+        series_title = None
+        series_description = None
+        translations = series.get('translations', [])
+        if translations:
+            series_title = translations[0].get('title') or translations[0]['seoTitle']
+            series_description = translations[0].get('seoDescription')
+        return self.playlist_result(
+            entries, compat_str(series['id']), series_title, series_description)