[Date Prev][Date Next] [Chronological] [Thread] [Top]

Re: Openldap ldbm cache corruption

At 01:58 AM 1/13/99 +0100, Harald Tveit Alvestrand wrote:
>it might be nice to downcase rather than upcase - seems weird, but
>there's a couple of cases (German double S, French (France) accents)
>where a downcase preserves information that upcase loses, if done
>according to a native locale.

We currently do everything in the "C" locale.  I'd prefer using
uppercase characters because they are (generally) more noticeably
different from the 'provided' DN.

Note: DNs and attributes should be T.61 encoded when using LDAPv2
(OpenLDAP).  Of course, I'm sure we have a few problems correctly
handling non-ASCII DNs and attributes.

We might actually want to leave the case alone so we can later
replace the strcasecmp() with ldapcmp()/ldapcasecmp() functions
(T.61 for LDAPv2, UTF-8 for LDAPv3).