gitlab-com-dstftw/youtube-dl
1
0
Fork 0
mirror of https://gitlab.com/dstftw/youtube-dl.git synced 2020-11-16 09:42:26 +00:00
Code Issues Releases Wiki Activity

[YoutubeDL] Skip malformed formats for better extraction robustness

Browse source
This commit is contained in:
Sergey M․ 2017-06-23 21:18:33 +07:00
parent b5f523ed62
commit 73af5cc817
No known key found for this signature in database
GPG key ID: 2C393E0F18A9236D
1 changed files with 12 additions and 4 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

16
youtube_dl/YoutubeDL.py
View file

@ -1448,17 +1448,25 @@ class YoutubeDL(object):
if not formats: if not formats:
raise ExtractorError('No video formats found!') raise ExtractorError('No video formats found!')
def is_wellformed(f):
url = f.get('url')
valid_url = url and isinstance(url, compat_str)
if not valid_url:
self.report_warning(
'"url" field is missing or empty - skipping format, '
'there is an error in extractor')
return valid_url
# Filter out malformed formats for better extraction robustness
formats = list(filter(is_wellformed, formats))
formats_dict = {} formats_dict = {}
# We check that all the formats have the format and format_id fields # We check that all the formats have the format and format_id fields
for i, format in enumerate(formats): for i, format in enumerate(formats):
if 'url' not in format:
raise ExtractorError('Missing "url" key in result (index %d)' % i)
sanitize_string_field(format, 'format_id') sanitize_string_field(format, 'format_id')
sanitize_numeric_fields(format) sanitize_numeric_fields(format)
format['url'] = sanitize_url(format['url']) format['url'] = sanitize_url(format['url'])
if format.get('format_id') is None: if format.get('format_id') is None:
format['format_id'] = compat_str(i) format['format_id'] = compat_str(i)
else: else: