From 9d22a7dfb07ad6ff23680e11238744b2bb1735aa Mon Sep 17 00:00:00 2001 From: Philipp Hagemeister Date: Fri, 16 Jan 2015 13:44:44 +0100 Subject: [PATCH] [fourtube] Fix extraction --- youtube_dl/extractor/fourtube.py | 76 ++++++++++++++++++-------------- youtube_dl/utils.py | 2 +- 2 files changed, 45 insertions(+), 33 deletions(-) diff --git a/youtube_dl/extractor/fourtube.py b/youtube_dl/extractor/fourtube.py index 7187e0752..b2284ab01 100644 --- a/youtube_dl/extractor/fourtube.py +++ b/youtube_dl/extractor/fourtube.py @@ -7,10 +7,9 @@ compat_urllib_request, ) from ..utils import ( - clean_html, parse_duration, + parse_iso8601, str_to_int, - unified_strdate, ) @@ -28,68 +27,81 @@ class FourTubeIE(InfoExtractor): 'uploader': 'WCP Club', 'uploader_id': 'wcp-club', 'upload_date': '20131031', + 'timestamp': 1383263892, 'duration': 583, + 'view_count': int, + 'like_count': int, + 'categories': list, } } def _real_extract(self, url): video_id = self._match_id(url) - webpage_url = 'http://www.4tube.com/videos/' + video_id - webpage = self._download_webpage(webpage_url, video_id) + webpage = self._download_webpage(url, video_id) - self.report_extraction(video_id) + title = self._html_search_meta('name', webpage) + timestamp = parse_iso8601(self._html_search_meta( + 'uploadDate', webpage)) + thumbnail = self._html_search_meta('thumbnailUrl', webpage) + uploader_id = self._html_search_regex( + r'', + webpage, 'uploader id') + uploader = self._html_search_regex( + r'', + webpage, 'uploader') - playlist_json = self._html_search_regex(r'var playerConfigPlaylist\s+=\s+([^;]+)', webpage, 'Playlist') - media_id = self._search_regex(r'idMedia:\s*(\d+)', playlist_json, 'Media Id') - sources = self._search_regex(r'sources:\s*\[([^\]]*)\]', playlist_json, 'Sources').split(',') - title = self._search_regex(r'title:\s*"([^"]*)', playlist_json, 'Title') - thumbnail_url = self._search_regex(r'image:\s*"([^"]*)', playlist_json, 'Thumbnail', fatal=False) + categories_html = self._search_regex( + r'(?s)>\s*Categories / Tags\s*.*?', + webpage, 'categories', fatal=False) + categories = None + if categories_html: + categories = [ + c.strip() for c in re.findall( + r'(?s)
  • (.*?)', categories_html)] - uploader_str = self._search_regex(r'Uploaded by(.*?)', webpage, 'uploader', fatal=False) - mobj = re.search(r'(?P[^<]+)', uploader_str) - (uploader, uploader_id) = (mobj.group('name'), mobj.group('id')) if mobj else (clean_html(uploader_str), None) + view_count = str_to_int(self._search_regex( + r'', + webpage, 'view count', fatal=False)) + like_count = str_to_int(self._search_regex( + r'', + webpage, 'like count', fatal=False)) + duration = parse_duration(self._html_search_meta('duration', webpage)) - upload_date = None - view_count = None - duration = None - description = self._html_search_meta('description', webpage, 'description') - if description: - upload_date = self._search_regex(r'Published Date: (\d{2} [a-zA-Z]{3} \d{4})', description, 'upload date', - fatal=False) - if upload_date: - upload_date = unified_strdate(upload_date) - view_count = self._search_regex(r'Views: ([\d,\.]+)', description, 'view count', fatal=False) - if view_count: - view_count = str_to_int(view_count) - duration = parse_duration(self._search_regex(r'Length: (\d+m\d+s)', description, 'duration', fatal=False)) + params_js = self._search_regex( + r'\$\.ajax\(url,\ opts\);\s*\}\s*\}\)\(([0-9,\[\] ]+)\)', + webpage, 'initialization parameters' + ) + params = self._parse_json('[%s]' % params_js, video_id) + media_id = params[0] + sources = ['%s' % p for p in params[2]] - token_url = "http://tkn.4tube.com/{0}/desktop/{1}".format(media_id, "+".join(sources)) + token_url = 'http://tkn.4tube.com/{0}/desktop/{1}'.format( + media_id, '+'.join(sources)) headers = { b'Content-Type': b'application/x-www-form-urlencoded', b'Origin': b'http://www.4tube.com', } token_req = compat_urllib_request.Request(token_url, b'{}', headers) tokens = self._download_json(token_req, video_id) - formats = [{ 'url': tokens[format]['token'], 'format_id': format + 'p', 'resolution': format + 'p', 'quality': int(format), } for format in sources] - self._sort_formats(formats) return { 'id': video_id, 'title': title, 'formats': formats, - 'thumbnail': thumbnail_url, + 'categories': categories, + 'thumbnail': thumbnail, 'uploader': uploader, 'uploader_id': uploader_id, - 'upload_date': upload_date, + 'timestamp': timestamp, + 'like_count': like_count, 'view_count': view_count, 'duration': duration, 'age_limit': 18, - 'webpage_url': webpage_url, } diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py index 2546fa45d..7832ed87f 100644 --- a/youtube_dl/utils.py +++ b/youtube_dl/utils.py @@ -1277,7 +1277,7 @@ def parse_duration(s): s = s.strip() m = re.match( - r'''(?ix)T? + r'''(?ix)(?:P?T)? (?: (?P[0-9.]+)\s*(?:mins?|minutes?)\s*| (?P[0-9.]+)\s*(?:hours?)|