About the CMIS Python Library

The goal of this project is to create a CMIS client for Python that can be used to work with any CMIS-compliant repository.

The library is being developed with the following guidelines:
  • Developers using this API should be able to work with CMIS domain objects without having to worry about the underlying implementation details.
  • The library will use the Resftul AtomPub Binding.
  • The library will conform to the CMIS spec. As the current Apache Chemistry test server fails the Apache Chemistry TCK, Alfresco will be used as the primary test repository.
  • The library should have no hard-coded URL’s. It should be able to get everything it needs regarding how to work with the CMIS service from the CMIS service URL response and subsequent calls.
  • There shouldn’t have to be a vendor-specific version of this library. The goal is for it to be interoperable with CMIS-compliant providers.

Example

This should give you an idea of how easy and natural it is to work with the API:
>>> cmisClient = CmisClient('http://localhost:8080/alfresco/s/cmis', 'admin', 'admin')
>>> repo = cmisClient.getDefaultRepository()
>>> rootFolder = repo.getRootFolder()
>>> children = rootFolder.getChildren()
>>> newFolder = rootFolder.createFolder({'cmis:name': 'testDeleteFolder folder'})
>>> props = newFolder.getProperties()
>>> newFolder.delete()

Note

I realize the createFolder call currently fails the “hide CMIS details from the user” directive.

To-Do’s

The library is not currently production ready. There are many outstanding todo’s:
  • Updating a private working copy (PWC) with a file
  • Allowable actions
  • More/better error-handling
  • Implementing optional arguments/filters, paging
  • Type object methods
  • Relationships, Policies, ACLs
  • More/better unit tests
  • Testing against multiple repositories

Table Of Contents

Previous topic

Welcome to CMIS Library’s documentation!

Next topic

Code

This Page