CHANGES.txt

v0.0.24, 2015-08-24 -- add sort feature. better read feature (thx to pasdavoine). fix a bug in set operations (thx to clearblank for reporting it)
v0.0.23, 2015-08-04 -- fix sqlite3 backend python3 incompatibility. pep8 cleaning
v0.0.22, 2015-08-04 -- add sqlite3 and shelve backend. thanks to Edwin Cox
v0.0.21, 2015-07-27 -- add link to nidus website
v0.0.20, 2015-07-03 -- tiny write perf improvement. better collect and request method, thanks to pasdavoine.
v0.0.19, 2015-07-01 -- adding set operations (issubset, issuperset, union, intersection, difference, symmetric_difference) which can be used as graph methods or entity iterators operators. overload __contains__ for 'in' and 'not in' test. add an ids method on entity iterators to get internal entity ids (useful in combination with _id filtering). POTENTIAL BREAKING CHANGES : a lot of entity iterators method were previously exhausting iterators, so clauses like it.count()==it.count() was False (with for instances, it=g.V())
v0.0.18, 2015-06-17 -- fix issues #2, #3, #4, #5, #6, #7, #8, #9. thanks to pasdavoine for issue reporting and PR
v0.0.17, 2015-03-17 -- create a zmq context and socket for each client request to prevent 'Operation cannot be accomplished in current state' errors
v0.0.16, 2015-03-04 -- fix a bug in collect (bad use of local cache). improve data method to allow to select a subset of fields, optionnally getting values in same order.
v0.0.15, 2014-12-02 -- add a .all method (to get sth similar to django .all method)
v0.0.14, 2014-12-02 -- now, allow filtering on internal edge and node ids, ex : g.V(_id=1)
v0.0.13, 2014-04-06 -- add a new method on entity and iterator : .random. this helps to do some "random walks".
v0.0.12, 2014-03-25 -- remove gevent dependency, code is now compatible with both Python 2.7 and Python 3.3/3.4
v0.0.11, 2014-03-08 -- add path algorithm (Dijkstra and A*). optimize .aka method performance. add a method to get a networkx graph from grapheekdb graph (provided networkx is installed). fix a bug on export/import (which was using msgpack instead of json).
v0.0.10, 2014-02-25 -- incompatible changes : agg method has been splitted in 2 methods : sum and percent. new agregation methods can now use complex expression with references to current entity or context
v0.0.9, 2014-02-14 -- agg method can now use contextual variable attributes. __in lookup taks advantage of compatible indexes. no longer use zmq.green fro client api.
v0.0.8, 2014-02-12 -- bugfix in edge neighbours retrieval when an edge index exists.
v0.0.7, 2014-02-12 -- bugfixes in index related code. proper db file closure. filter values are now sanitized before being sent to server (allows to use set or tuples when server is expecting list). server scripts now worked on entities in bulk
v0.0.6, 2014-02-09 -- partial indexes now implemented/msgpack used instead of json (add a requirement but reduce file size)
v0.0.5, 2014-02-09 -- indexes are implemented separately. ditto for database operations. better handling of "supernodes" and low discriminative indexes. various fixes
v0.0.4, 2014-02-05 -- EntityIterator agg method now returns an EntityAggregate which is iterable and has 2 methods : limit and entities, useful for method chaining
v0.0.3, 2014-02-02 -- Change "map" method to "data". Update documentation
v0.0.2, 2014-02-02 -- Adding dot export (to use in IPython).
v0.0.1, 2014-01-31 -- Initial release.