From e88c9ef62a4a26cc77370b741a4244d298c7d45a Mon Sep 17 00:00:00 2001 From: Remita Amine Date: Mon, 4 Jan 2021 00:51:55 +0100 Subject: [utils] add a function to clean podcast URLs --- youtube_dl/utils.py | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) (limited to 'youtube_dl') diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py index d5fb6fd24..8e4d144c9 100644 --- a/youtube_dl/utils.py +++ b/youtube_dl/utils.py @@ -5706,3 +5706,20 @@ def random_birthday(year_field, month_field, day_field): month_field: str(random_date.month), day_field: str(random_date.day), } + + +def clean_podcast_url(url): + return re.sub(r'''(?x) + (?: + (?: + chtbl\.com/track| + media\.blubrry\.com| # https://create.blubrry.com/resources/podcast-media-download-statistics/getting-started/ + play\.podtrac\.com + )/[^/]+| + (?:dts|www)\.podtrac\.com/(?:pts/)?redirect\.[0-9a-z]{3,4}| # http://analytics.podtrac.com/how-to-measure + flex\.acast\.com| + pd(?: + cn\.co| # https://podcorn.com/analytics-prefix/ + st\.fm # https://podsights.com/docs/ + )/e + )/''', '', url) -- cgit v1.2.3