Tal Einat <taleinat+pyt...@gmail.com> added the comment: Putting aside backwards compatibility, I would argue the opposite: Since consecutive slashes are valid, I suggest we would need to see that collapsing them into a single slash is the status quo, otherwise we should avoid such collapsing.
Here's some evidence that we should keep consecutive slashes: 1. The curl cli does not appear to collapse consecutive slashes in URLs before sending them: $ nc -l localhost 8000 & [1] 39380 $ curl --dump-header - --proxy localhost:8000 --silent --max-time 1 http://www.example.com/this//double/path GET http://www.example.com/this//double/path HTTP/1.1 Host: www.example.com User-Agent: curl/7.64.1 Accept: */* Proxy-Connection: Keep-Alive [1] + 39380 done nc -l localhost 8000 2. With NodeJS v10.18.0 and v12.16.1: > const urllib = require('url'); > new url.URL("this//double/path", "http://www.example.com/";).href 'http://www.example.com/this//double/path' For me this is evidence enough that urljoin should *not* be collapsing consecutive slashes. ---------- _______________________________________ Python tracker <rep...@bugs.python.org> <https://bugs.python.org/issue40594> _______________________________________ _______________________________________________ Python-bugs-list mailing list Unsubscribe: https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
