Re: problem with spaces in urls

Ah. Just not enough coffee.

You should look in: parseHttpRequest() in icp.c

You should also look at the url_escape() function in url.c

Actually, I just tried this request on Squid: GET http://www.brisnet.org.au/a b c d
HTTP/1.0

The result was an unexpected bad treatment:
98/03/07 10:17:46| parseHttpRequest: Request is 'http://www.brisnet.org.au/a'
98/03/07 10:17:46| parseHttpRequest: HTTP version is 'b c d HTTP/1.0
'

(I expected it to be rejected outright. I'd be happy with 'converted into a working URL,
but failing that I'd expect rejection as an HTTP violation, and not an attempt to fetch
it)

D

rosenblg@nyu.edu wrote:

> thanks for the reply; i'm aware of the spaces not being legal (which is
> why i mentioned my not having control over what's generated). to make
> matters worse, it's RLIN (library) software, which is almost impossible
> to get changed.
>
> if anyone has any hack to fix it (or where i should look) in the source,
> please let me know - i'm holding off hacking anything for a bit!
>
> gary
>
> Dancer said:
> >
> > Then it is not a legal URL. In order to be a legal URL the spaces should be
> > represented by '+' or '%20' (both of which are legal encodings for spaces). Putting
> > a real space in the URL is a good way to get it to not work properly, unless the
> > browser is smart enough to see it and do the encoding itself (which many are).
> >
> > D

--
Did you read the documentation AND the FAQ?
If not, I'll probably still answer your question, but my patience will
be limited, and you take the risk of sarcasm and ridicule.