[freesound] Improve and remove unrelated metadata (closes #11608)

youtube_dl/extractor/freesound.py

@@ -4,18 +4,15 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
-    determine_ext,
     float_or_none,
     get_element_by_class,
     get_element_by_id,
-    int_or_none,
-    parse_filesize,
     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',
@@ -23,62 +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)
+        audio_id = self._match_id(url)
-        music_id = mobj.group('id')
+
-        webpage = self._download_webpage(url, music_id)
+        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')
-        duration = float_or_none(get_element_by_class('duration', webpage), scale=1000)
-        tags = get_element_by_class('tags', webpage)
-        sound_info = get_element_by_id('sound_information_box', webpage)
-        release_date = get_element_by_id('sound_date', webpage)
 
         description = self._html_search_regex(
-            r'<div id="sound_description">(.*?)</div>', webpage, 'description',
+            r'(?s)id=["\']sound_description["\'][^>]*>(.+?)</div>',
-            fatal=False, flags=re.DOTALL)
+            webpage, 'description', fatal=False)
 
-        download_count = int_or_none(self._html_search_regex(
+        duration = float_or_none(
-            r'Downloaded.*>(\d+)<', webpage, 'downloaded', fatal=False))
+            get_element_by_class('duration', webpage), scale=1000)
 
-        filesize = float_or_none(parse_filesize(self._search_regex(
+        upload_date = unified_strdate(get_element_by_id('sound_date', webpage))
-            r'Filesize</dt><dd>(.*)</dd>', sound_info, 'file size (approx)', fatal=False)))
+        uploader = self._og_search_property(
+            'audio:artist', webpage, 'uploader', fatal=False)
 
-        if release_date:
+        channels = self._html_search_regex(
-            release_date = unified_strdate(release_date.replace('th', ''))
+            r'Channels</dt><dd>(.+?)</dd>', webpage,
+            'channels info', fatal=False)
 
-        bitdepth = self._html_search_regex(
+        tags_str = get_element_by_class('tags', webpage)
-            r'Bitdepth</dt><dd>(.*)</dd>', sound_info, 'Bitdepth', fatal=False)
+        tags = re.findall(r'<a[^>]+>([^<]+)', tags_str) if tags_str else None
 
-        channels = self._html_search_regex(
+        audio_urls = [audio_url]
-            r'Channels</dt><dd>(.*)</dd>', sound_info, 'Channels info', fatal=False)
+
+        LQ_FORMAT = '-lq.mp3'
+        if LQ_FORMAT in audio_url:
+            audio_urls.append(audio_url.replace(LQ_FORMAT, '-hq.mp3'))
 
         formats = [{
-            'url': audio_url,
+            'url': format_url,
-            'id': music_id,
+            'format_note': channels,
-            'format_id': self._og_search_property('audio:type', webpage, 'audio format', fatal=False),
+            'quality': quality,
-            'format_note': '{0} {1} {2}'.format(determine_ext(audio_url), bitdepth, channels),
+        } for quality, format_url in enumerate(audio_urls)]
-            'filesize_approx': filesize,
+        self._sort_formats(formats)
-            'asr': int_or_none(self._html_search_regex(
-                r'Samplerate</dt><dd>(\d+).*</dd>',
-                sound_info, 'samplerate', fatal=False)),
-        }]
 
         return {
-            'id': music_id,
+            'id': audio_id,
             'title': title,
-            'uploader': self._og_search_property('audio:artist', webpage, 'music uploader', fatal=False),
             'description': description,
             'duration': duration,
-            'tags': [self._html_search_regex(r'>(.*)</a>', t, 'tag', fatal=False)
+            'uploader': uploader,
-                     for t in tags.split('\n') if t.strip()],
+            'upload_date': upload_date,
+            'tags': tags,
             'formats': formats,
-            'release_date': release_date,
-            'likes_count': download_count,
         }