[EMAIL PROTECTED] - Thu Apr  6 14:55:42 2006]:

> 
> I'm still not convinced that all of the other assumptions made in the
> X509 name comparison routines are valid for subsequent use of the sorted
> list in a binary search.  We are not seeing any more problems though.
> 
>  

Yes the problem is more complex that that. As things stand the new
X509_NAME_cmp() function is no longer a valid metric.

It violates the requirement that if a == b and b == c then a == c

To see this consider the case of a DN field with a = "HELLO WORLD" and b
= "hello world" both PRINTABLESTRINGS and c = "hello world" is a T61.

Case insensitive comparision will make a and b equal. Memory comparision
will occur between a or b and c. That will make a != c but b == c.

Can you disable the cross string type comparision and see if that also
resolves your problems?


______________________________________________________________________
OpenSSL Project                                 http://www.openssl.org
Development Mailing List                       [email protected]
Automated List Manager                           [EMAIL PROTECTED]

Reply via email to