[refactor] Single quotes consistency

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

diff --git a/youtube_dl/extractor/rte.py b/youtube_dl/extractor/rte.py
index c17ef2cc2f31dca76b7f56dfb69fcf582089a863..042bc8dab9a1d6b273c94e03b88f152a49217e89 100644 (file)
--- a/youtube_dl/extractor/rte.py
+++ b/youtube_dl/extractor/rte.py
@@ -1,9 +1,12 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 from ..utils import (
     float_or_none,
+    parse_iso8601,
     unescapeHTML,
 )
 
@@ -40,7 +43,7 @@ class RteIE(InfoExtractor):
             r'<meta name="thumbnail" content="uri:irus:(.*?)" />', webpage, 'thumbnail')
         thumbnail = 'http://img.rasset.ie/' + thumbnail_id + '.jpg'
 
-        feeds_url = self._html_search_meta("feeds-prefix", webpage, 'feeds url') + video_id
+        feeds_url = self._html_search_meta('feeds-prefix', webpage, 'feeds url') + video_id
         json_string = self._download_json(feeds_url, video_id)
 
         # f4m_url = server + relative_url
@@ -60,30 +63,53 @@ class RteIE(InfoExtractor):
 class RteRadioIE(InfoExtractor):
     IE_NAME = 'rte:radio'
     IE_DESC = 'Raidió Teilifís Éireann radio'
-    # Radioplayer URLs have the specifier #!rii=<channel_id>:<id>:<playable_item_id>:<date>:
+    # Radioplayer URLs have two distinct specifier formats,
+    # the old format #!rii=<channel_id>:<id>:<playable_item_id>:<date>:
+    # the new format #!rii=b<channel_id>_<id>_<playable_item_id>_<date>_
     # where the IDs are int/empty, the date is DD-MM-YYYY, and the specifier may be truncated.
     # An <id> uniquely defines an individual recording, and is the only part we require.
-    _VALID_URL = r'https?://(?:www\.)?rte\.ie/radio/utils/radioplayer/rteradioweb\.html#!rii=(?:[0-9]*)(?:%3A|:)(?P<id>[0-9]+)'
+    _VALID_URL = r'https?://(?:www\.)?rte\.ie/radio/utils/radioplayer/rteradioweb\.html#!rii=(?:b?[0-9]*)(?:%3A|:|%5F|_)(?P<id>[0-9]+)'
 
-    _TEST = {
+    _TESTS = [{
+        # Old-style player URL; HLS and RTMPE formats
         'url': 'http://www.rte.ie/radio/utils/radioplayer/rteradioweb.html#!rii=16:10507902:2414:27-12-2015:',
         'info_dict': {
             'id': '10507902',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'Gloria',
             'thumbnail': 're:^https?://.*\.jpg$',
-            'description': 'Tim Thurston guides you through a millennium of sacred music featuring Gregorian chant, pure solo voices and choral masterpieces, framed around the glorious music of J.S. Bach.',
+            'description': 'md5:9ce124a7fb41559ec68f06387cabddf0',
+            'timestamp': 1451203200,
+            'upload_date': '20151227',
             'duration': 7230.0,
         },
         'params': {
             'skip_download': 'f4m fails with --test atm'
         }
-    }
+    }, {
+        # New-style player URL; RTMPE formats only
+        'url': 'http://rte.ie/radio/utils/radioplayer/rteradioweb.html#!rii=b16_3250678_8861_06-04-2012_',
+        'info_dict': {
+            'id': '3250678',
+            'ext': 'flv',
+            'title': 'The Lyric Concert with Paul Herriott',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'description': '',
+            'timestamp': 1333742400,
+            'upload_date': '20120406',
+            'duration': 7199.016,
+        },
+        'params': {
+            'skip_download': 'f4m fails with --test atm'
+        }
+    }]
 
     def _real_extract(self, url):
         item_id = self._match_id(url)
-        feeds_url = 'http://www.rte.ie/rteavgen/getplaylist/?type=web&format=json&id=' + item_id
-        json_string = self._download_json(feeds_url, item_id)
+
+        json_string = self._download_json(
+            'http://www.rte.ie/rteavgen/getplaylist/?type=web&format=json&id=' + item_id,
+            item_id)
 
         # NB the string values in the JSON are stored using XML escaping(!)
         show = json_string['shows'][0]
@@ -91,13 +117,24 @@ class RteRadioIE(InfoExtractor):
         description = unescapeHTML(show.get('description'))
         thumbnail = show.get('thumbnail')
         duration = float_or_none(show.get('duration'), 1000)
+        timestamp = parse_iso8601(show.get('published'))
 
         mg = show['media:group'][0]
 
         formats = []
 
-        if mg.get('url') and not mg['url'].startswith('rtmpe:'):
-            formats.append({'url': mg['url']})
+        if mg.get('url'):
+            m = re.match(r'(?P<url>rtmpe?://[^/]+)/(?P<app>.+)/(?P<playpath>mp4:.*)', mg['url'])
+            if m:
+                m = m.groupdict()
+                formats.append({
+                    'url': m['url'] + '/' + m['app'],
+                    'app': m['app'],
+                    'play_path': m['playpath'],
+                    'player_url': url,
+                    'ext': 'flv',
+                    'format_id': 'rtmp',
+                })
 
         if mg.get('hls_server') and mg.get('hls_url'):
             formats.extend(self._extract_m3u8_formats(
@@ -114,8 +151,9 @@ class RteRadioIE(InfoExtractor):
         return {
             'id': item_id,
             'title': title,
-            'formats': formats,
             'description': description,
             'thumbnail': thumbnail,
+            'timestamp': timestamp,
             'duration': duration,
+            'formats': formats,
         }