https://bugs.openldap.org/show_bug.cgi?id=9257
Bug ID: 9257 Summary: Abstract attribute types can be instantiated Product: OpenLDAP Version: unspecified Hardware: All OS: All Status: UNCONFIRMED Severity: normal Priority: --- Component: slapd Assignee: bugs@openldap.org Reporter: grapvar@gmail.com Target Milestone: ---
If you add 'name' or 'distinguishedName' attributes somewhere they are added seamlessly.
However, these attributes are SLAP_AT_ABSTRACT, which
slap.h: #define SLAP_AT_ABSTRACT 0x0100U /* cannot be instantiated */
https://bugs.openldap.org/show_bug.cgi?id=9257
--- Comment #1 from Michael Ströder michael@stroeder.com --- AFAICS there is no such thing like an "abstract" attribute type in RFC 4512. So if 'name' and 'distinguishedName' are in the subschema subentry they should be usable just like any other attribute type.
https://bugs.openldap.org/show_bug.cgi?id=9257
--- Comment #2 from Konstantin Andreev grapvar@gmail.com --- rfc4519/2.7,2.18 strongly discourages using these attributes on it's own, although does not prohibit such use. What is described there closely matches `abstract' concept in, e.g. c++, so `slap_at_abstract' makes sense.
However, I do no argue neither pro nor contra slap_at_abstract. There is a conflict between the claim [/* cannot be instantiated */] and reality, and this conflict needs resolution.
https://bugs.openldap.org/show_bug.cgi?id=9257
--- Comment #3 from Howard Chu hyc@openldap.org --- (In reply to Konstantin Andreev from comment #2)
rfc4519/2.7,2.18 strongly discourages using these attributes on it's own, although does not prohibit such use. What is described there closely matches `abstract' concept in, e.g. c++, so `slap_at_abstract' makes sense.
However, I do no argue neither pro nor contra slap_at_abstract. There is a conflict between the claim [/* cannot be instantiated */] and reality, and this conflict needs resolution.
This claim is not published in any user documentation, so I don't see any pressing need to correct it. Certainly, to enforce it at this point in time is likely to break someone's DIT if they ever used these attribute types directly. As such, there is risk of harm from changing it, and no risk from leaving it alone.
The flag was added in 55e925abe63b49cdf6a5ddcd88240b2259eb7dc1 so it's been around for a long time without any issue.
I see no bug to fix here.
https://bugs.openldap.org/show_bug.cgi?id=9257
--- Comment #4 from Konstantin Andreev grapvar@gmail.com --- That was me who was caught in this tender snare. I had tried to use slap_at_abstract, and failed because it is ... not implemented 8-O
-
openldap-its＠openldap.org