Skip to content

Latest commit

 

History

History
77 lines (58 loc) · 2.45 KB

README.rst

File metadata and controls

77 lines (58 loc) · 2.45 KB
Raw

pydatomic

Python library for accessing the datomic DBMS via the REST API. Includes a reader for edn.

REST client

Connections are instances of datomic.Datomic:

>>> from datomic import Datomic
>>> conn = Datomic('http://localhost:3000/', 'tdb')

The method create_database(name) returns a database object which can be used for queries. It has the same methods as the Datomic connection instance, but you don't pass the database name as the first argument.

>>> db = conn.create_database('cms')
>>> db.transact(["""{
...  :db/id #db/id[:db.part/db]
...  :db/ident :person/name
...  :db/valueType :db.type/string
...  :db/cardinality :db.cardinality/one
...  :db/doc "A person's name"
...  :db.install/_attribute :db.part/db}"""])   #doctest: +ELLIPSIS
{':db-after':...
>>> db.transact(['{:db/id #db/id[:db.part/user] :person/name "Peter"}'])  #doctest: +ELLIPSIS
{':db-after':...
>>> r = db.query('[:find ?e ?n :where [?e :person/name ?n]]')
>>> print r  #doctest: +ELLIPSIS
((... u'Peter'))
>>> eid = r[0][0]

The query function optionally takes arguments to apply to the query and has a keyword argument history for querying the history database:

>>> print db.query('[:find ?n :in $ ?e :where [?e :person/name ?n]]', [eid], history=True)
((u'Peter',),)
>>> print db.entity(eid)  #doctest: +ELLIPSIS
{':person/name': u'Peter', ':db/id': ...}

TBD

  • Support for as-of and since
  • Support for data-structure queries instead of just textual ones (need to implement an EDN encoder for that).

edn parser

Includes a parser for most of EDN (https://github.com/edn-format/edn), featuring:

  • Coroutine-based interface for streaming data
  • loads() interface for the rest of the time!
  • Strings and characters are converted to unicode before passing to application
  • Support for tags
  • All structures are returned as immutable objects except dicts, as Python still lacks a frozendict type.
  • Symbols and keywords are returned as strings (not unicode)

TBD

  • Encoder!
  • Handle invalid input gracefully
  • Check validity of strings for keywords/symbols
  • Include a frozendict implementation?
  • Create a type for symbols and keywords?
  • Better API for adding tag handlers (currently you need to modify the global dictionary!)
  • Map exact floating point values to Decimal type?
  • Don't call tag handlers whilst parsing the element after a discard

License

Distributed under the MIT license.