https://issues.apache.org/bugzilla/show_bug.cgi?id=55323

--- Comment #3 from Eric Covener <[email protected]> ---
(In reply to Simon Klinkert from comment #2)
> (In reply to Eric Covener from comment #1)
> > That looks too general, your symptom is that you pass file%#12file into a
> > rewriterule (-U check?) and it needs to be encoded at that point, but the
> > patch would re-escape  anyone who sent in a properly escaped URL.
> 
> First of all, thanks for your feedback!
> 
> How do I encode the uri at "that point"?
> 
> There is no -U involved. My rewrite rule looks like this:
> 
> RewriteRule ^(/.*)? %{LA-U:ENV:storage_path}$1 [L]

$1 has captured a URL-escaped string. I think you can re-escabe it with the [B]
flag, or pull the unescaped version out of %{THE_REQUEST} in a rewritecond.

> In addition, I do not really see why this should be too general. The
> function ap_sub_req_method_uri does already a re-escaping if new_uri does
> not start with '/'. Why is that bad for the other case?

Sorry, did not look to closely and not too familiar with it. Was just concerned
generally about the scope of the change.

-- 
You are receiving this mail because:
You are the assignee for the bug.

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to