TODO

CitationEngine

Short term todo list:

• Be ready for people to help.
  • Get some working examples of how our code works
  • Build a large test bank for people to work with and build grammars for.

Long term todo list:

• Document class
  • Create a pretty print function for displaying Document contents
• Parser classes
  • Complete "abbreviated journal" to "unabbreviated journal" tool.
    • NOTE: Look into what pubmed has done toward this already
• Fetcher classes
  • Pubmed
    • Need to determine resolutionToken lifespan
  • Store meta-collection data i.e. query used, source obtained from, and timestamp
    • Need to know who, when, and from which batch. I.E.: user.datetime.query.pointers to all documents/raw data collected.
      • Think about saving all captured information to disk -- json?
      • local storage?
      • database?
      • used in conjunction with merges/conflict resolutions
• CorpusController class
  • Setup logging?
• Database stuff
  • Think about an optimistic insert or random ID
  • Improve document merging / conflict resolution
  • Consider Bloom filter vs hash for DB queries
  • Modify db class to accept which database to interface with
  • Modify db.add_or_update() to return the objectID of the Document inserted into the DB?
• Task Queue
  • Implement and test Celery/RabbitMQ with citation engine