Incremental custom model

[heiko_robert]

You could store your model in the Alfresco repo (/Data Dictionary/Models) which can be easily modified from Share-UI or WebDAV but I would not recommend this in production systems. You don't have the same flexibility and how would you implement deployments / lifecycle management, QM? Why not creating traditional spring based models packaged in jars which can be easily deployed by the server team? Of course you need a dev system with server filesystem access to develop and test your changes ...

Could you please give an example for the model extensions you want to implement?

We tried to implement a shared model (GitHub - ecm4u/businessdocumentmodel: ecm4u Alfresco Business Document Model ) to be used by as many Alfresco users as possible to avoid the problems you are facing with but we didn't find enough combatants. We should update this model since we already refactored this to a more flexible aproach which allows online changes on generic types referencing relational and/or external data.

[rsclark3]

First problem is editing an existing content model. I need to add new values to the selection list e.g., 'As-Built' and 'Specifications' so these can be selected. At this point, I can only have a vendor come in and make the change.

Second problem is extending this model to include new properties. I want to add 'Company' as a multi-line text entry later on in this complex model.

The other problem is property layout in a form but that's beyond this question. I had hoped to use Model Manager for all content model attributes but now I'm stuck with engaging expensive consultants to make minor changes as we expand our models for searching.

Ultimately, it would be best if a robust Model Manager was added to Alfresco that allows both file system models to be exposed and new models to be added.

[javiercasfer]

Hi, thanks for all the answers. 

In my case, your recomendations do not work for me.

What I have is a document manager wich document types have some properties and, maybe, on two or five month, that document type will have two more properties. That could happen to all types i have. 

What I need is to import all those documents (only those that are not going to be modified) and import them to Alfresco, to create a historic document storage. 

After studying what could be the best solution, I think that an aggregation model is the solution. I made a software that can create a model with all properties for all documents extracted and create CMM file for them all, but, in the future, will be more extractions, so I'll need to add all those new properties and types. The old model will be update with the new model generated on the extraction. 

There will be a moment that I'll have an active model that I have to update. 

Finally, my questions are:

   - Having the model created, wich options do i have to import it to alfresco?

   - Do I have to create manually all cofigurations files and update them all (config, message, etc) or there is an alternative (like an amp file) to import it automatically?

Hope there is someone in my situation with an answer

[javiercasfer · Accepted solution]

Hi everybody!

I've solved my problem. 

What I do is, when I do a content extraction from the source, i generate a model content xml with the types and properties i get. Then I use the xsd definition to compare my old model with my new one and add all new types and properties i have to the new one. 

Then, from may aggregation model (the old updated), i generate my own share-config-custom.xml with the forms and properties I need and model.propreties files. Then, replace these 3 files on my alfresco instalation and restart it. It works perfectly. Keep in mind that these only will work if you don't remove types or properties when you are updating your custom model. 

Maybe that could help someone. Thanks for all the answers.