mirrors/ArchiveBox
1
0
Fork 0
mirror of https://github.com/ArchiveBox/ArchiveBox.git synced 2025-05-14 07:04:27 -04:00
Code Issues Projects Releases Packages Wiki Activity Actions

remove derived link info in favor of derived Link properties

Browse source
This commit is contained in:
Nick Sweeting 2019-04-16 23:21:24 -04:00
parent 50d368b1bc
commit 717e390ef6
4 changed files with 40 additions and 34 deletions
Download patch file Download diff file Expand all files Collapse all files

2
archivebox/legacy/util.py
View file

@ -60,7 +60,6 @@ base_url = lambda url: without_scheme(url) # uniq base url used to dedupe links
without_www = lambda url: url.replace('://www.', '://', 1)
without_trailing_slash = lambda url: url[:-1] if url[-1] == '/' else url.replace('/?', '?')
fuzzy_url = lambda url: without_trailing_slash(without_www(without_scheme(url.lower())))
hashurl = lambda url: base32_encode(int(sha256(base_url(url).encode('utf-8')).hexdigest(), 16))[:20]
urlencode = lambda s: s and quote(s, encoding='utf-8', errors='replace')
@ -393,6 +392,7 @@ def parse_date(date: Any) -> Optional[datetime]:
pass
if '-' in date:
# 2019-04-07T05:44:39.227520
try:
return datetime.fromisoformat(date)
except Exception: