# coding: utf-8 from __future__ import unicode_literals import re from .common import InfoExtractor from ..utils import ( determine_ext, float_or_none, get_element_by_id, int_or_none, parse_iso8601, str_to_int, ) class IzleseneIE(InfoExtractor): _VALID_URL = r'''(?x) https?://(?:(?:www|m)\.)?izlesene\.com/ (?:video|embedplayer)/(?:[^/]+/)?(?P<id>[0-9]+) ''' _TESTS = [ { 'url': 'http://www.izlesene.com/video/sevincten-cildirtan-dogum-gunu-hediyesi/7599694', 'md5': '4384f9f0ea65086734b881085ee05ac2', 'info_dict': { 'id': '7599694', 'ext': 'mp4', 'title': 'Sevinçten Çıldırtan Doğum Günü Hediyesi', 'description': 'md5:253753e2655dde93f59f74b572454f6d', 'thumbnail': 're:^http://.*\.jpg', 'uploader_id': 'pelikzzle', 'timestamp': 1404302298, 'upload_date': '20140702', 'duration': 95.395, 'age_limit': 0, } }, { 'url': 'http://www.izlesene.com/video/tarkan-dortmund-2006-konseri/17997', 'md5': '97f09b6872bffa284cb7fa4f6910cb72', 'info_dict': { 'id': '17997', 'ext': 'mp4', 'title': 'Tarkan Dortmund 2006 Konseri', 'description': 'Tarkan Dortmund 2006 Konseri', 'thumbnail': 're:^http://.*\.jpg', 'uploader_id': 'parlayankiz', 'timestamp': 1163322193, 'upload_date': '20061112', 'duration': 253.666, 'age_limit': 0, } }, ] def _real_extract(self, url): video_id = self._match_id(url) url = 'http://www.izlesene.com/video/%s' % video_id webpage = self._download_webpage(url, video_id) title = self._og_search_title(webpage) description = self._og_search_description(webpage) thumbnail = self._proto_relative_url( self._og_search_thumbnail(webpage), scheme='http:') uploader = self._html_search_regex( r"adduserUsername\s*=\s*'([^']+)';", webpage, 'uploader', fatal=False, default='') timestamp = parse_iso8601(self._html_search_meta( 'uploadDate', webpage, 'upload date', fatal=False)) duration = float_or_none(self._html_search_regex( r'"videoduration"\s*:\s*"([^"]+)"', webpage, 'duration', fatal=False), scale=1000) view_count = str_to_int(get_element_by_id('videoViewCount', webpage)) comment_count = self._html_search_regex( r'comment_count\s*=\s*\'([^\']+)\';', webpage, 'comment_count', fatal=False) family_friendly = self._html_search_meta( 'isFamilyFriendly', webpage, 'age limit', fatal=False) content_url = self._html_search_meta( 'contentURL', webpage, 'content URL', fatal=False) ext = determine_ext(content_url, 'mp4') # Might be empty for some videos. streams = self._html_search_regex( r'"qualitylevel"\s*:\s*"([^"]+)"', webpage, 'streams', fatal=False, default='') formats = [] if streams: for stream in streams.split('|'): quality, url = re.search(r'\[(\w+)\](.+)', stream).groups() formats.append({ 'format_id': '%sp' % quality if quality else 'sd', 'url': url, 'ext': ext, }) else: stream_url = self._search_regex( r'"streamurl"\s?:\s?"([^"]+)"', webpage, 'stream URL') formats.append({ 'format_id': 'sd', 'url': stream_url, 'ext': ext, }) return { 'id': video_id, 'title': title, 'description': description, 'thumbnail': thumbnail, 'uploader_id': uploader, 'timestamp': timestamp, 'duration': duration, 'view_count': int_or_none(view_count), 'comment_count': int_or_none(comment_count), 'age_limit': 18 if family_friendly == 'False' else 0, 'formats': formats, }