I'm not familiar with the 'LDAP Data Interchange Format' (to say the least) but is 'uid' always bound to follow 'cn'?

I don't know either, but if it varies, reading the data one whole record at a time will make it a lot easier to handle the variation.

Hey guys,
Thanks for all the help, I've almost got it working now. UID is not a necessary attribute of an LDAP record, however to use Oracle Portal it is. I'm trying to move from a Oracles 902 ldap directory to a 10g directory and this is my final step.

My only problem now is I have 4 records with cn on the last line of the record and for some reason its not find it in the search. i.e.

cn=HELPDESK,cn=users,dc=myco,dc=com
orclactivestartdate: 20031202043257Z
objectclass: top
objectclass: person
objectclass: inetOrgPerson
objectclass: organizationalPerson
objectclass: orclUser
objectclass: orclUserV2
o: MyOrg
orcldefaultprofilegroup: cn=user_grp,cn=portal_groups,cn=groups,dc=myco,dc=com
givenname: Joe
sn: Schmo
telephonenumber: 800-999-9999
userpassword: {MD4}1a4QEWWP+nCZ19uAy9w==
mail: joe_blow@myco.com
cn: HELPDESK
Any ideas? Thanks again Matt

My only problem now is I have 4 records with cn on the last line of the record and for some reason its not find it in the search.

If you're using the code that I originally proposed, then the match is failing because there is nothing that matches the "$3" ("$nexttag") portion of the regex. If you make that part of the match optional, it should work fine -- just put a "?" quantifier on the part that would be $3:

   if ( /(\ncn: (\S+))\n(\S+)?/ ) {
[download]

Now, if you need to make sure that all the lines are in a specific order (e.g., if the "cn:/uid:" pair should not be the last two lines of the record), you would probably be best off splitting the record on "\n" (after you've fixed the uid line), assigning the lines to an array (or better yet, a hash, keyed by the first field on each line), and then printing the lines out in the desired order.