[Date Prev][Date Next]
DN normalization (was: Please comment)
At 09:19 PM 6/4/00 -0400, Mark Valence wrote:
>>A note: we should be a bit careful of over optimizing DN
>>processing as we will likely have to rework DNs later to
>>fully implement RFC 2253.
>I'll take a look at 2253 and see if that's something I can do at the
>same time (even though it's sort of tangential).
Basically, we need to explode the DN into component RDN,
then explode each RDN, and then generate a string which is truely
normalized. This would deal with escaping/quoting issues,
base64 values, multi-valued RDNs, base64 DN, attribute type
name normalization, etc. Should be built into the new
dnNormalize() routine, replacing old dn_normalize().
For now, I'm fine with our oversimplified DN normalization
(as I avoid all these nasties in my DNs).