diff --git a/youtube_dl/extractor/lrt.py b/youtube_dl/extractor/lrt.py index f5c997ef4..70b21ff0a 100644 --- a/youtube_dl/extractor/lrt.py +++ b/youtube_dl/extractor/lrt.py @@ -6,8 +6,6 @@ import re from .common import InfoExtractor from ..utils import ( determine_ext, - int_or_none, - parse_duration, remove_end, ) @@ -17,70 +15,67 @@ class LRTIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?lrt\.lt/mediateka/irasas/(?P[0-9]+)' _TESTS = [{ # m3u8 download - 'url': 'http://www.lrt.lt/mediateka/irasas/54391/', - 'md5': 'fe44cf7e4ab3198055f2c598fc175cb0', + 'url': 'https://www.lrt.lt/mediateka/irasas/54391/septynios-kauno-dienos', + 'md5': '50c84b875e930784daa66fa8c2617664', 'info_dict': { 'id': '54391', 'ext': 'mp4', 'title': 'Septynios Kauno dienos', 'description': 'md5:24d84534c7dc76581e59f5689462411a', - 'duration': 1783, - 'view_count': int, - 'like_count': int, }, }, { - # direct mp3 download - 'url': 'http://www.lrt.lt/mediateka/irasas/1013074524/', - 'md5': '389da8ca3cad0f51d12bed0c844f6a0a', + # after the redesign - everything is mp4 file (some of the shows used to be mp3) + 'url': 'https://www.lrt.lt/mediateka/irasas/1013074524/kita-tema-2016-09-05-15-05', + 'md5': 'e286cd475ad2839ab90c2ec0be57930b', 'info_dict': { 'id': '1013074524', - 'ext': 'mp3', + 'ext': 'mp4', 'title': 'Kita tema 2016-09-05 15:05', 'description': 'md5:1b295a8fc7219ed0d543fc228c931fb5', - 'duration': 3008, - 'view_count': int, - 'like_count': int, }, }] def _real_extract(self, url): video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) - title = remove_end(self._og_search_title(webpage), ' - LRT') + file_url1 = "" + urlinfo = "" + formats = [] for _, file_url in re.findall( - r'file\s*:\s*(["\'])(?P(?:(?!\1).)+)\1', webpage): - ext = determine_ext(file_url) + r'(main_url) = "(.*)";', webpage): + urlinfo = self._download_json("https://www.lrt.lt/servisai/stream_url/vod/media_info/?url=" + file_url, video_id) + + file_url1 = urlinfo.get("playlist_item").get("file") + + release_date_string_temp = urlinfo.get("date") + release_date = release_date_string_temp[0:4] + release_date_string_temp[5:7] + release_date_string_temp[8:10] + + ext = determine_ext(file_url1) if ext not in ('m3u8', 'mp3'): continue # mp3 served as m3u8 produces stuttered media file - if ext == 'm3u8' and '.mp3' in file_url: + if ext == 'm3u8' and '.mp3' in file_url1: continue if ext == 'm3u8': formats.extend(self._extract_m3u8_formats( - file_url, video_id, 'mp4', entry_protocol='m3u8_native', + file_url1, video_id, 'mp4', entry_protocol='m3u8_native', fatal=False)) elif ext == 'mp3': formats.append({ - 'url': file_url, + 'url': file_url1, 'vcodec': 'none', }) self._sort_formats(formats) - thumbnail = self._og_search_thumbnail(webpage) - description = self._og_search_description(webpage) - duration = parse_duration(self._search_regex( - r'var\s+record_len\s*=\s*(["\'])(?P[0-9]+:[0-9]+:[0-9]+)\1', - webpage, 'duration', default=None, group='duration')) + try: + thumbnail = self._og_search_thumbnail(webpage) + except: + thumbnail = 0 - view_count = int_or_none(self._html_search_regex( - r']+class=(["\']).*?record-desc-seen.*?\1[^>]*>(?P.+?)', - webpage, 'view count', fatal=False, group='count')) - like_count = int_or_none(self._search_regex( - r']+id=(["\'])flikesCount.*?\1>(?P\d+)<', - webpage, 'like count', fatal=False, group='count')) + description = urlinfo.get("content") return { 'id': video_id, @@ -88,7 +83,5 @@ class LRTIE(InfoExtractor): 'formats': formats, 'thumbnail': thumbnail, 'description': description, - 'duration': duration, - 'view_count': view_count, - 'like_count': like_count, + 'release_date': release_date }