You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@httpd.apache.org by "Jeffrey W. Baker" <jw...@acm.org> on 2001/03/01 00:49:14 UTC

[PATCH] update to ap_unparse_uri_components()

Please add this latest patch to the Apache 1.3 tree.  It fixes two
failures in the current 1.3.19 candidate:

1) the URI "#frag" would be emitted as "", because of a NULL path
2) sometimes an "@" was present in the site part even in the absence of
user info

For reference, the 1.3.17 version of this function could (aside from
dumoing core) emit bogosities such as:

http://hostpath
user@path?name=value#frag
http://@host/path?name=value#frag

Okay I guess that last one is valid but it still looks silly.  I have also
attached my updated battery of test cases.

Warm regards,
Jeffrey Baker

Re: [PATCH] update to ap_unparse_uri_components()

Posted by "Jeffrey W. Baker" <jw...@acm.org>.
On Thu, 1 Mar 2001, Martin Kraemer wrote:

> On Wed, Feb 28, 2001 at 03:49:14PM -0800, Jeffrey W. Baker wrote:
> > Please add this latest patch to the Apache 1.3 tree.  It fixes two
> > failures in the current 1.3.19 candidate:
> >
> > 1) the URI "#frag" would be emitted as "", because of a NULL path
> > 2) sometimes an "@" was present in the site part even in the absence of
> > user info
> >
>
> Thanks (for both the patch and the test suite).
>
> I wonder if case 1) and 2) can occur in real-world apache use.
> Though I am with you that the error should be fixed, I think 1.3.19
> is not at risk without the fix. Or what do you think?

I agree.  Case 1 is something that I bet nobody is programmatically
generating, and case 2 is technically a valid URI.  No risk to 1.3.19.

-jwb

Re: [PATCH] update to ap_unparse_uri_components()

Posted by Martin Kraemer <Ma...@Fujitsu-Siemens.com>.
On Wed, Feb 28, 2001 at 03:49:14PM -0800, Jeffrey W. Baker wrote:
> Please add this latest patch to the Apache 1.3 tree.  It fixes two
> failures in the current 1.3.19 candidate:
> 
> 1) the URI "#frag" would be emitted as "", because of a NULL path
> 2) sometimes an "@" was present in the site part even in the absence of
> user info
> 

Thanks (for both the patch and the test suite).

I wonder if case 1) and 2) can occur in real-world apache use.
Though I am with you that the error should be fixed, I think 1.3.19
is not at risk without the fix. Or what do you think?

   Martin
-- 
<Ma...@Fujitsu-Siemens.com>    |       Fujitsu Siemens
       <ma...@apache.org>              |   81730  Munich,  Germany