Mirrors/ArchiveBox
2
0
Fork 0
mirror of https://github.com/ArchiveBox/ArchiveBox synced 2024-11-23 04:33:11 +00:00
Code Issues Projects Releases Packages Wiki Activity

dont remove query when uniquifying links

Browse source
This commit is contained in:
Nick Sweeting 2018-04-17 03:54:59 -04:00
parent 542b1d14f6
commit 0099849df9
1 changed files with 1 additions and 1 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
util.py
View file

@ -33,7 +33,7 @@ without_query = lambda url: url.split('?', 1)[0]
without_hash = lambda url: url.split('#', 1)[0] without_hash = lambda url: url.split('#', 1)[0]
without_path = lambda url: url.split('/', 1)[0] without_path = lambda url: url.split('/', 1)[0]
domain = lambda url: without_hash(without_query(without_path(without_scheme(url)))) domain = lambda url: without_hash(without_query(without_path(without_scheme(url))))
base_url = lambda url: without_query(without_scheme(url)) base_url = lambda url: without_scheme(url) # uniq base url used to dedupe links
short_ts = lambda ts: ts.split('.')[0] short_ts = lambda ts: ts.split('.')[0]