On 11/05/13 14:44, Stefan Sperling wrote:
On Tue, Nov 05, 2013 at 02:08:21PM +0100, Alexander Hall wrote:
On 11/05/13 13:56, Stefan Sperling wrote:
Before:
$ ftp ' http://localhost/snap/INSTALL.amd64'
ftp: http: no address associated with name
ftp: Can't connect or login to host ` http'
After:
$ ftp ' http://localhost/snap/INSTALL.amd64'
Trying ::1...
Trying 127.0.0.1...
Requesting http://localhost/snap/INSTALL.amd64
100% |**************************************************| 84357 00:00
84357 bytes received in 0.00 seconds (267.27 MB/s)
Do others think this useful? I hit this because I made copy/paste errors.
Unless you can give a reference saying an URL may start with arbitrary
whitespace, I don't want this to go in.
Well, I would bet anyone can easily find claims that go either
way in some RFC. Here's one that's in favour of this change, FWIW.
https://tools.ietf.org/html/rfc3986#appendix-C
[[[
In practice, URIs are delimited in a variety of ways, but usually
within double-quotes "http://example.com/", angle brackets
<http://example.com/>, or just by using whitespace
[...]
For robustness, software that accepts user-typed URI should attempt
to recognize and strip both delimiters and embedded whitespace."
]]
Not sure if a program argument qualifies as "user typed" (the example
extracts urls from a text).
That said, jj@ seems to be on your line, so there are at least 2-1 votes
in your favor.
Just my 2 cents.
/Alexander