X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/af478477605bdf3f5d57562035885cfee905f379..0fada1afdee4b9d13610291943d8b2d9ea7283a2:/youtube_dl/extractor/freesound.py?ds=sidebyside

diff --git a/youtube_dl/extractor/freesound.py b/youtube_dl/extractor/freesound.py
index 5ff62af..138b6bc 100644
--- a/youtube_dl/extractor/freesound.py
+++ b/youtube_dl/extractor/freesound.py
@@ -3,10 +3,16 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..utils import (
+    float_or_none,
+    get_element_by_class,
+    get_element_by_id,
+    unified_strdate,
+)
 
 
 class FreesoundIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?freesound\.org/people/([^/]+)/sounds/(?P<id>[^/]+)'
+    _VALID_URL = r'https?://(?:www\.)?freesound\.org/people/[^/]+/sounds/(?P<id>[^/]+)'
     _TEST = {
         'url': 'http://www.freesound.org/people/miklovan/sounds/194503/',
         'md5': '12280ceb42c81f19a515c745eae07650',
@@ -14,26 +20,60 @@ class FreesoundIE(InfoExtractor):
             'id': '194503',
             'ext': 'mp3',
             'title': 'gulls in the city.wav',
-            'uploader': 'miklovan',
             'description': 'the sounds of seagulls in the city',
+            'duration': 130.233,
+            'uploader': 'miklovan',
+            'upload_date': '20130715',
+            'tags': list,
         }
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        music_id = mobj.group('id')
-        webpage = self._download_webpage(url, music_id)
-        title = self._html_search_regex(
-            r'<div id="single_sample_header">.*?<a href="#">(.+?)</a>',
-            webpage, 'music title', flags=re.DOTALL)
+        audio_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, audio_id)
+
+        audio_url = self._og_search_property('audio', webpage, 'song url')
+        title = self._og_search_property('audio:title', webpage, 'song title')
+
         description = self._html_search_regex(
-            r'<div id="sound_description">(.*?)</div>', webpage, 'description',
-            fatal=False, flags=re.DOTALL)
+            r'(?s)id=["\']sound_description["\'][^>]*>(.+?)</div>',
+            webpage, 'description', fatal=False)
+
+        duration = float_or_none(
+            get_element_by_class('duration', webpage), scale=1000)
+
+        upload_date = unified_strdate(get_element_by_id('sound_date', webpage))
+        uploader = self._og_search_property(
+            'audio:artist', webpage, 'uploader', fatal=False)
+
+        channels = self._html_search_regex(
+            r'Channels</dt><dd>(.+?)</dd>', webpage,
+            'channels info', fatal=False)
+
+        tags_str = get_element_by_class('tags', webpage)
+        tags = re.findall(r'<a[^>]+>([^<]+)', tags_str) if tags_str else None
+
+        audio_urls = [audio_url]
+
+        LQ_FORMAT = '-lq.mp3'
+        if LQ_FORMAT in audio_url:
+            audio_urls.append(audio_url.replace(LQ_FORMAT, '-hq.mp3'))
+
+        formats = [{
+            'url': format_url,
+            'format_note': channels,
+            'quality': quality,
+        } for quality, format_url in enumerate(audio_urls)]
+        self._sort_formats(formats)
 
         return {
-            'id': music_id,
+            'id': audio_id,
             'title': title,
-            'url': self._og_search_property('audio', webpage, 'music url'),
-            'uploader': self._og_search_property('audio:artist', webpage, 'music uploader'),
             'description': description,
+            'duration': duration,
+            'uploader': uploader,
+            'upload_date': upload_date,
+            'tags': tags,
+            'formats': formats,
         }