Pycassa

Bug Fixes

* Pin Thrift dependency to 0.9.3 to avoid problems with Thrift 0.10.0
(fixes Github issue 245)

Features:

* Add describe_token_map() to SystemManager

Miscellaneous:

* Add TimestampType alias for DateType
* Match Cassandra's sorting of TimeUUIDs in stubs

Features:

* Upgrade Thrift interface to 19.36.1, which adds support for the
LOCAL_ONE consistency level and the populate_io_cache_on_flush
column family attribute.

Bug Fixes

* Return timestamp from remove() in stub ColumnFamily

Miscellaneous

* Upgrade bundled ez_setup.py

This release only adds one feature: support for Cassandra 1.2’s atomic
batches.
Features

* Add support for Cassandra 1.2+ atomic batches through a new atomic
parameter for batch.Mutator, batch.CfMutator, and ColumnFamily.batch().

This release fixes a few edge cases around connection pooling that can affect
long-running applications. It also adds token range support to
ColumnFamily.get_range(), which can be useful for parallelizing full-ring
scans.

Features

* Add token range support to ColumnFamily.get_range()

Bug Fixes

* Prevent possible double connection disposal when recycling connections
* Handle empty strings for IntegerType values
* Prevent closed connections from being returned to the pool.
* Ensure connection count is decremented when pool is disposed

This release adds a couple of minor new features and improves multithreaded
locking behavior in ConnectionPool. There should be no backwards-
compatibility concerns.

Features

* Full support for column_start, column_finish, column_count, and
column_reversed parameters in contrib.stubs
* Addition of an include_ttl parameter to ColumnFamily fetching
methods which works like the existing include_timestamp parameter.

Bug Fixes

* Reduce the locked critical section in ConnectionPool, primarily to
make sure lock aquisition time is not ignored outside of the pool’s
timeout setting.