From c63ca0eef8ac147b3f2a39ba7265ad1b3c11d516 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Mon, 14 May 2018 23:27:56 +0700 Subject: [PATCH] [youtube] Improve format filesize extraction (#16453) --- youtube_dl/extractor/youtube.py | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py index 897398d208..7f4298c088 100644 --- a/youtube_dl/extractor/youtube.py +++ b/youtube_dl/extractor/youtube.py @@ -1815,6 +1815,10 @@ def _extract_count(count_name): chapters = self._extract_chapters(description_original, video_duration) + def _extract_filesize(media_url): + return int_or_none(self._search_regex( + r'\bclen[=/](\d+)', media_url, 'filesize', default=None)) + if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'): self.report_rtmp_download() formats = [{ @@ -1919,8 +1923,11 @@ def _extract_count(count_name): mobj = re.search(r'^(?P\d+)[xX](?P\d+)$', url_data.get('size', [''])[0]) width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None) + filesize = int_or_none(url_data.get( + 'clen', [None])[0]) or _extract_filesize(url) + more_fields = { - 'filesize': int_or_none(url_data.get('clen', [None])[0]), + 'filesize': filesize, 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000), 'width': width, 'height': height, @@ -1994,6 +2001,8 @@ def decrypt_sig(mobj): for df in self._extract_mpd_formats( mpd_url, video_id, fatal=dash_mpd_fatal, formats_dict=self._formats): + if not df.get('filesize'): + df['filesize'] = _extract_filesize(df['url']) # Do not overwrite DASH format found in some previous DASH manifest if df['format_id'] not in dash_formats: dash_formats[df['format_id']] = df