Kolibri channel content collections

A Kolibri channel consists of two parts:

• Metadata stored in sqlite3 DB file. The channel DB file can downloaded from URL like /content/databases/{{channel_id}}.sqlite3 from any instance of the Kolibri application, Kolibri Studio, or obtained through direct file transfer.

• A collection of files stored in /content/storage/{x}/{y}/{xyz.......}.ext

For the purpose of importing Kolibri channels as content collections, we need to import only the content metadata from the Kolibri database file.

Usage

1. Obtain the channel_id of the channel you want to import. You can find the a channel’s ID from the URL when viewing the channel in Kolirbi or Kolibri Studio.

2. Run the script in the rocdata/contentcollections-kolibri repository to obtain the kolibritree JSON dump of the channel database:./kolibri_db/reader.py --channel_id=<channel_id>

3. Import the channel as a content collection using the rocserver management command:

   ./manage.py ccimport_kolibri  \
      --jurisdiction LE \
      --country US \
      --name <shortname> \
      --source_domain <sourcewebsiteurl> \
      --source_url <weburlofcollection> \
      --kolibritree_url=<urlwherejsoncanbedownloadedfrom>
   

   where <shortname> must be a unique, short identifier for the collection, e.g. KA-en.

The following properties of the content collection will be set by default:

• collection_id: set to channel_id

• version: taken from imported JSON [TODO]

• publication_status="publicdraft". This can be chanced to “public” through admin.

• subjects=[]: can add subject references (ManyToManyField relations to terms in vocabularies of kind subjects.

• education_levels=[]: can add grade levels references (relations to terms in vocabularies of kind education_levels.