Cf-python

--------------

**2016-06-23**

* Added reference_datetime option to cf.write
* Fixed bug in `cf.um.read.read` which incorrectly ordered vertical
coordinates

----

-------------

**2016-06-17**

* New methods `cf.Variable.files` and `cf.Data.files`,
`cf.Field.files` which report which files are referenced by the data
array.
* Fix to stop partitions return `numpy.bool_` instead of
`numpy.ndarray`
* Fix to determining cyclicity of regridded fields.
* Functionality to recursively read directories in `cf.read`, ``cfa``
and ``cfump``
* Print warning but carry on when ESMF import fails
* Fixed bug in `cf.Field.subspace` when accessing axes derived from UM
format files

----

-------------

**2016-05-18**

* Slightly changed the compression API to `cf.write`
* Added compression support to the ``cfa`` command line script
* Added functionality to change data type on writing to `cf.write` and
``cfa`` - both in general and for with extra convenience for the
common case of double to single (and vice versa).
* Removed annoying debug print statements from `cf.um.read.read`

----

-------------

**2016-05-04**

* Added fix for change in numpy behaviour (`numpy.number` types do not
support assignment)
* Added capability to load in a user STASH to standard name table:
`cf.um.read.load_stash2standard_name`

----

-------------

**2016-04-27**

* Added --reference_datetime option to ``cfa``
* Bug fix to `cf.Field.collapse` when providing `cf.Query` objects via
the group parameter
* Added auto regridding method, which is now the default


----

-------------

**2016-03-03**

* Bug fix in `cf.Field.where` when using `cf.masked`
* conda installation (with thanks to Andy Heaps)
* Bug fix for type casting in `cf.Field.collapse`
* Display long_name if it exists and there is no standard_name
* Fix for compiling the UM C code on certain OSs (with thanks to Simon Wilson)
* Fixed incorrect assignment of cyclicity in `cf.Field.regrids`
* Nearest neighbour regridding in `cf.Field.regrids`

----