Dear ConnId community,
at Evolveum, we identified a need to expand current schema support in ConnId framework, which improves end-user experience and performance when processing schema:
We at Evolveum are very interested in these features, and would like to develop it, if it makes sense to you. What do you think?
Below you can find more detailed proposals for each of these features.
1. Support for description attribute for objectClass and attribute definition
ObjectClassInfo
and AttributeInfo
objects.
Rationale: The “description” parameters would provide information about the intended usage of the Object Classes and the attributes right in the schema.
Examples:
"User" Object Class description: "A User object is a digital identity object that represents a single human user or a non-human agent (e.g., service account) authorized to access digital resources."
"User" Object class "member_of" attribute description: "Unique identifiers of groups represented as a list of memberships for policy inheritance."
Design and Implementation
Extension of the ObjectClassInfo class and addition of "Description" field with accessor methods. Similar approach for the AttributeInfo
class.
(Also builder classes, encoder/decoder and other updated accordingly.)
2. Support for schema generation constraints
A set of new features enabling the client to request from the connector only a part of the full schema provided by the resource
Rationale: The schema of some remote systems can contain a large number of object classes, which can also contain numerous information regarding attributes.
This information has to be handled by the connector client, even if there are often use cases when only a small subset of the object classes present in the resource schema is actually needed.
This could be mitigated by providing the connector client with a list of Object class information objects, the client would then request the schema with only a handful of object classes, which were selected from the list.
Example: LDAP servers and AD contain large schemas, out of which only a couple of object classes tend to be managed by IAM systems, usually those representing some sort of group or account. Also, large numbers of cloud services provide API endpoints, which quite often have a large number of Object classes to choose from.
Design and Implementation
The feature consists of a new API class containing two methods. One would provide the connector client with a set of “Lightweight” ConnectorObjectInfo
objects, which would be lacking Attribute
information. This is to make the amount of data reasonably small, yet provide other information that might be needed by the client. A second method would then return a Schema object constructed based on the list of Object classes chosen by the client, provided by the client as an input parameter.
As mentioned earlier, we at Evolveum are very interested in these features and would like to develop them if it makes sense to you. What do you think?
Thanks for your feedback.
--
You received this message because you are subscribed to the Google Groups "connid-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to connid-dev+...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/connid-dev/2122219629.2312764.1753268218186.JavaMail.zimbra%40evolveum.com.
-- Francesco Chicchiriccò Tirasa - Open Source Excellence http://www.tirasa.net/ Member at The Apache Software Foundation Syncope, Cocoon, Olingo, CXF, OpenJPA https://about.me/ilgrosso
-- Francesco Chicchiriccò Tirasa - Open Source Excellence