gStudio-Lite¶

Elastic Search & RCS implementation of gStudio

Update following files & modules:¶

requirements.txt:
- Strip down to only necessary ones.
- Are we going with oac/oat? Needs to take decision to strikeout qbank requirements accordingly.
Update INSTALL.txt
SQL User DB
- Are we continuing with sql db for user?
- If yes, then are we going with sqlite of postgreSQL ?
fab commands
Development possibilities?
- If yes, then we can continue with development requirements else we can remove them.
Code to update in following modules
- benchmarker
  - We can remove benchmarker or do changes.
- settings.py and local_settings.py
- ndf
  - commands
  - models:
    - Override following collection instances with equivalent elastic search (ES) implementation:
      - node_collection
      - triple_collection
      - benchmark_collection
      - filehive_collection
      - buddy_collection
      - counter_collection
    - Override following for ES and call existing for mongo:
      - one()
      - find()
      - find_one()
      - update()
      - <collection instances>.collection.<classname>() # to create new instance
    - These overriding can be taken to separate file next to models or at root level.
    - While overriding take care that we are supporting all mongo queries, operators and it has equivalent implementation in ES.
  - Check code in following:
    - templatetags
    - views

Script to inject existing data objects to ES container
Update save() in models to inject new and updated object to ES container
Implement update method to:
- Call default mongo update()
- Implement custom for ES
...
...