On 17 Mar 2013, at 6:54 PM, Eric Covener <[email protected]> wrote:

> If we maintain the use of a password here, like mod_ssl does, wouldn't
> we need to make sure it doesn't come in over the wire?

We use apr_table_setn() which replaces anything that is there already, although 
if either the username or the password resolve to an empty string it is 
possible for a user to inject their own.

I think to be safe, we should unset the header in the two empty string cases. 
Done in r1457504.

Regards,
Graham
--

Attachment: smime.p7s
Description: S/MIME cryptographic signature

Reply via email to