From 3a407e707ac96bc082fd82325e916802a3b55d36 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Thu, 12 Jan 2017 23:03:53 +0700 Subject: [PATCH] [freesound] Improve and remove unrelated metadata (closes #11608) --- youtube_dl/extractor/freesound.py | 75 +++++++++++++++---------------- 1 file changed, 35 insertions(+), 40 deletions(-) diff --git a/youtube_dl/extractor/freesound.py b/youtube_dl/extractor/freesound.py index f0b2400cf..138b6bc58 100644 --- a/youtube_dl/extractor/freesound.py +++ b/youtube_dl/extractor/freesound.py @@ -4,18 +4,15 @@ 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[^/]+)' + _VALID_URL = r'https?://(?:www\.)?freesound\.org/people/[^/]+/sounds/(?P[^/]+)' _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) - music_id = mobj.group('id') - webpage = self._download_webpage(url, music_id) + 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') - 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'
(.*?)
', webpage, 'description', - fatal=False, flags=re.DOTALL) + r'(?s)id=["\']sound_description["\'][^>]*>(.+?)', + webpage, 'description', fatal=False) - download_count = int_or_none(self._html_search_regex( - r'Downloaded.*>(\d+)<', webpage, 'downloaded', fatal=False)) + duration = float_or_none( + get_element_by_class('duration', webpage), scale=1000) - filesize = float_or_none(parse_filesize(self._search_regex( - r'Filesize
(.*)
', sound_info, 'file size (approx)', fatal=False))) - - if release_date: - release_date = unified_strdate(release_date.replace('th', '')) - - bitdepth = self._html_search_regex( - r'Bitdepth
(.*)
', sound_info, 'Bitdepth', fatal=False) + 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
(.*)
', sound_info, 'Channels info', fatal=False) + r'Channels
(.+?)
', webpage, + 'channels info', fatal=False) + + tags_str = get_element_by_class('tags', webpage) + tags = re.findall(r']+>([^<]+)', 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': audio_url, - 'id': music_id, - 'format_id': self._og_search_property('audio:type', webpage, 'audio format', fatal=False), - 'format_note': '{0} {1} {2}'.format(determine_ext(audio_url), bitdepth, channels), - 'filesize_approx': filesize, - 'asr': int_or_none(self._html_search_regex( - r'Samplerate
(\d+).*
', - sound_info, 'samplerate', fatal=False)), - }] + '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, - 'uploader': self._og_search_property('audio:artist', webpage, 'music uploader', fatal=False), 'description': description, 'duration': duration, - 'tags': [self._html_search_regex(r'>(.*)', t, 'tag', fatal=False) - for t in tags.split('\n') if t.strip()], + 'uploader': uploader, + 'upload_date': upload_date, + 'tags': tags, 'formats': formats, - 'release_date': release_date, - 'likes_count': download_count, }