sjw/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-11-10 12:49:02 +01:00
Code Issues Packages Projects Releases Wiki Activity

[utils] Strip double spaces in clean_html

Browse Source 
Closes #2497
Authored by: dirkf
This commit is contained in:
pukkandan 2022-02-03 19:45:57 +05:30
parent 65662dffb1
commit 4918522735
No known key found for this signature in database
GPG Key ID: 0F00D95A001F4698
1 changed files with 3 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
yt_dlp/utils.py
View File

@ -639,10 +639,9 @@ def clean_html(html):
if html is None: # Convenience for sanitizing descriptions etc.
return html
# Newline vs <br />
html = html.replace('\n', ' ')
html = re.sub(r'(?u)\s*<\s*br\s*/?\s*>\s*', '\n', html)
html = re.sub(r'(?u)<\s*/\s*p\s*>\s*<\s*p[^>]*>', '\n', html)
html = re.sub(r'\s+', ' ', html)
html = re.sub(r'(?u)\s?<\s?br\s?/?\s?>\s?', '\n', html)
html = re.sub(r'(?u)<\s?/\s?p\s?>\s?<\s?p[^>]*>', '\n', html)
# Strip html tags
html = re.sub('<.*?>', '', html)
# Replace html entities