Changelog

 version 1.0.9 (not yet released)
 ================================

 * add .travis.yml (for travis-ci testing on github), silence warnings from
   test output (issue 225).

 * modify __unicode__ for Variable and Dimension to return more useful error
   message when Dataset object has been garbage collected.

 * use weak references to group instances when creating Dimension and Variable
   objects. This prevents cyclic references messing up garbage collection (issue
   218, pull request 219).

 * accessing values from a 0-dimensional Variable now returns a 0-dimensional
   numpy array, not a 1-dimensional array (issue 220). To write code
   compatible with both the old and new (fixed) behavior, wrap values accessed
   from a 0-dimensional Variable with numpy.asscalar.

 * add an __array__ method to Variable to make numpy ufuncs faster (issue 216).

 * change download_url in setup.py to point to pypi instead of googlecode.

 * fix for date2index error when time variable has only one entry (issue 215).

 * silence warnings ("Non-trivial type declarators in shared declaration (e.g.
   mix of pointers and values). Each pointer declaration should be on its own
   line") with Cython 0.2.

 version 1.0.8 (tag v1.0.8rel)
 =============================

 * change file_format Dataset attribute to data_model (keeping file_format
   for backward compatibility).  Add disk_format attribute (underlying
   disk format, one of NETCDF3, HDF4, HDF5, DAP2, DAP4, PNETCDF or UNDEFINED).
   Uses nc_inq_format_extended, added in version 4.3.1 of library.  If using
   earlier version of lib, disk_format will be set to UNDEFINED.

 * default _FillValue now ignored for byte data types (int8 and uint8) as per
   http://www.unidata.ucar.edu/software/netcdf/docs/netcdf-c/Fill-Values.html#Fill-Values
   "If you need a fill value for a byte variable, it is recommended that you
   explicitly define an appropriate _FillValue attribute, as generic utilities
   such as ncdump will not assume a default fill value for byte variables".  
   ncinfo now returns fill mode information (issue 209).

 * check to see if filling was disabled before masking data equal to
   default fill value (issue 209).

 * add variable type information to Dataset.__repr__ (output of ncinfo).

 version 1.0.7 (tag v1.0.7rel)
 =============================

 * add the ability to specify the locations of hdf4,jpeg and curl libs, in
   case netCDF4 was built statically with HDF4 and/or OpenDAP support (issue
   207).

 * add 'ncinfo' utility (like 'ncdump -h' but less verbose).

 * more information displayed when Dataset or Group instance is printed.

 * fix for issue 194 (versions after 1.0.5 fail for netcdf 4.1.1, due to 
   call to nc_inq_path, which was added in netcdf 4.1.2).  Fixed by adding
   compile time API check similar to what was done for nc_rename_grp.  If
   filepath Dataset method is called an exception will be raised at runtime
   if the module was built with netcdf < 4.1.2, or cython was not installed
   at build time.

 * fix for issues 202 and 206 (exception raised by numpy.isnan for character data types).

 * if dateutils not installed and time unit accuracy < 1 second requested, have
   netcdftime raise an ImportError.

 version 1.0.6 (svn revision 1312)
 ================================

 * issue warning of endian-ness of dtype argument does not match endian kwarg in
   createVariable.

 * make sure netcdf type NC_CHAR always returned in numpy array dtype 'S1'
   (sometimes arrays of type 'U1' were being returned). Fixes intermittently failing
   test tst_compoundatt.py on python 3.3.

 * fix for issue 201 (if data associated with numpy array not the 
   same endian-ness as dtype, data was written incorrectly).  Now
   bytes are swapped if necessary.  Variable.endian() now returns
   'native' instead of None for NETCDF3 formatted files.  createVariable
   now enforces endian='native' for NETCDF3 files. Added tst_endian.py
   test case.

 * fix for issue 200 (library version detection failed on cygwin).

 * fix for issue 199 (nc4tonc3 utility not copying global attributes).

 * fix for issue 198 (setup.py chokes when no arguments given).

 * fix for issue 197 (slicing of netCDF variables using lists of integers).

 * create 'path' attribute for group instance using posixpath, instead 
   of os.path (to ensure the unix path is used on all platforms). Issue 196.

 * fix for issue 196 (test failures on win32 due to files being deleted
   before they are closed).

 version 1.0.5 (svn revision 1278)
 ================================

 * change setup.py to compile the Cython sources directly, if
   cython is available. 
   This allows for "ifdef" like capability to modify source at compile time to
   account for changes in netcdf API (e.g. the forthcoming addition of the 
   nc_rename_grp in version 4.3.1).

 * added a "renameGroup" method, which raises an exception if
   the netcdf lib version linked does not support it. Requires
   netcdf >= 4.3.1. 

 * support for more than one missing value (missing_value attribute
   is a vector) when converting to masked array.

 * add 'renameAttribute' method to Dataset, Group and Variable.

 * fix so that var[:] = x works if x is a scalar, and var is a netcdf variable
   with an unlimited dimension that has shape () - i.e. no data has been
   written to it yet.  Before this change, var[:] = x did not write any data.
   Now the scalar x will be written as the first entry in var along the
   unlimited dimension.

 * remove dos line feeds from nc3tonc4 (issue 181).

 * add datatype property for Variable that returns numpy dtype for primitive
   datatypes (same as dtype attribute) but returns CompoundType or VLType 
   instance for compound or vlen variables (issue 178).

 * fix logic for deciding where to look for nc-config in setup.py (issue 177).

 * issue a warning and don't try to apply scale_factor or add_offset if these 
   attributes are not convertible to floats (issue 176). 

 * add filepath method to Dataset instance to return file path (or opendap
   URL) used to create Dataset (issue 172).

 * fix for issue 170 (opening a remote DAP dataset fails after creating
   a NETCDF4 formatted file).

 * fix for issue 169 (error in chartostring function on 64-bit windows).

 * add support for missing_value or _FillValue == NaN (issue 168).

 * added a Dimension.group() method (issue 165).

 version 1.0.4 (svn revision 1229)
 =================================

 * fixed alignment bug that could cause memory corruption
   when reading compound type variables. All users of compound
   types should upgrade.

 version 1.0.3 (svn revision 1219)
 =================================

 * don't try to write empty data array to netcdf file (fixed failing
   test with netcdf 4.3.0rc2).

 * date2num, num2date and date2index now can handle units of microseconds and milliseconds
   (for proleptic_gregorian calendar, or gregorian and standard calendars
    as long as the time origin is after 1582-10-15). Issue 159.

 * Added a _grp attribute to Dimension (issue 165).

 * don't bundle ordereddict (issue 164).

 * support reading of vlen string attributes (issue 156).

 * add --vars option to nc3tonc4 (issue 154).

 * Don't try to set fletcher32 checksum on scalar variables (it causes HDF5
   to crash).  Fixes issue 150.

 * Add --istart/--istop options to nc3tonc4 (issue 148, courtesy of 
   Rich Signell).

 * fix for proleptic_gregorian in netcdftime.py (courtesy of Matthias Cuntz).

 version 1.0.2 (svn revision 1196)
 =================================

 * disable version check for HDF5, which is broken by hdf5 1.8.10.

 * make sure all files have a calendar attribute in MFTime (issue 144).

 * more robust fix to issue 90 (array shape modified by assignment to
   a netCDF variable with one more dimension), including test case.

 version 1.0.1 (svn revision 1190)
 =================================

 * fix error that occurred when retrieving data from a variable that has a
   missing_value attribute specified as a string (issue 142).

 * automatically close netcdf files when there are no references
   left to Dataset object (using __dealloc__ method).  Fixes issue 137.

 * fix for slicing of scalar vlen string variables (issue 140).

 * fix to allow writing of unicode data to a NC_CHAR variable.

 * allow for writing of large variables (> 2**32 elements). Fixes issue 130.

 version 1.0fix1
 ===============
  
 * fix python 3 incompatibility in setup.py (issue 125).

 version 1.0 (svn revision 1164)
 ===============================
 * add 'aggdim' keyword to MFDataset, so the name of the dimension
   to aggregate over can be specified (instead of using the unlimited
   dimension).  aggdim=None by default, which results in the previous
   behavior.  aggdim must be the leftmost dimension of all the variables
   to be aggregated.

 * raise IndexError when indexing a netcdf variable out of range
   so iterating over a variable in a for loop behaves as expected
   (as described in http://effbot.org/zone/python-for-statement.htm).
   Fixes issue 121.

 * added MacPorts portfile (so it can be installed via MacPorts
   on macosx using a "local Portfile repository"). Installs 
   from svn HEAD using 'port install netcdf4-python'.

 * added experimental 'diskless' file capability (only added to the C 
   lib after the 4.2 release).  Controlled by kwarg 'diskless' to
   netCDF4.Dataset (default False). diskless=True when creating a file
   results in a file that exists only in memory, closing the file
   makes the data disapper, except if persist=True keyword given in
   which case it is persisted to a disk file on close.  diskless=True
   when opening a file creates an in-memory copy of the file for faster access.

 * add the ability to specify the location of the required libs 
   (and whether to use nc-config) with setup.cfg, 
   instead of using environment variables.

 * fix ISO9601 date parser so it recognizes time zone offsets in time unit
   strings (contributed by David Hassel, issue 114, r1117).

 * add setncatts Dataset,Group and Variable method to add a bunch of
   attributes (given in a python dictionary) at once.  Speeds things
   up for NETCDF3 and NETCDF4_CLASSIC files a lot, since nc_redef/nc_enddef  
   not need to be called for each attribute (issue 85, r1113). Adding 1000 
   attributes is about 35 times faster using setncatts to add them all at once. 
   Makes no difference for NETCDF4 formatted files, since nc_redef/nc_enddef
   is not called.

 * only round after apply scale_factor and add_offset if variable type is
   integer (issue 111, r1109).

 * Fixed bug with all False Boolean index (r1107).

 * added support for after, before and nearest selection method to date2index
   fast "first guess" indexing (r1106).

 * Remove white space in time units string (netcdftime._parse_date).
   An extra space in the time units of one CMIP3 model caused an error
   (r1105).

 * based on results with examples/bench_compress2.py, change
   default complevel for zlib compression from 6 to 4. If complevel=0,
   turn compression off entirely (set zlib=False) (r1102).

 version 0.9.9 (svn revision 1099)
 ================================
 * changed default unicode encoding from "latin-1" to "utf-8", since
   this is the python 3 default, and the only encoding that appears to
   work for dimension and variable names.

 * added test case for unicode attributes, variable and dimension names.

 * fixes for unicode variable, dimension and group names.

 * fix for unicode attributes in python3 (ncdump did not intrepret them as
   text strings). Issue 107.

 * add --format option to nc4tonc3 utility (can be either NETCDF3_CLASSIC or
   NETCDF3_64BIT). Fixes issue 104.

 version 0.9.8 (svn revision 1080)
 ================================
 * use numpy.ma.isMA to check for masked array (instead of checking
   for presence of 'mask' attribute).

 * fixes for AIX with ibm xlc compiler.

 * make sure unicode attributes don't get converted to ascii strings (issue
   98).

 version 0.9.7 (svn revision 1073)
 ================================
 * Added __str__ methods to Dataset, Variable, Dimension, CompoundType,
   VLType and MFDataset, so useful human-readable information is provided when these
   objects are printed in an interactive session.

 * don't try to apply scale_factor and offset if scale_factor=1 and
   add_offset=0 (to avoid making copies of large arrays).

 * changed netCDF4._default_fillvals to netCDF4.default_fillvals (to make part
   of public API).  Added to docs (issue 94).

 version 0.9.6 (svn revision 1043)
 =================================
 * changed default unicode encoding from "ascii" to "latin-1" (iso-8859-1).

 * add "unicode_error" module variable to control what happens when characters
   cannot be decoded by the encoding specified by the "default_encoding" module
   variable (which is "ascii" by default).  unicode_error = "replace" by default
   which means bad characters are replace by "?".  Previously an error was
   raised, the old behavior can be obtained by setting unicode_error = 'strict'.
   Fixes issue 92.

 * add __enter__ and __exit__ methods so you can do "with Dataset(url) as f:"
   (issue 89).

 * don't add extra singleton dimensions to rhs numpy arrays when assigning to
   a netcdf variable. Fixes issue 90.

 * coerce missing_value attribute to same type as variable (for primitive
   types). Fixes issue 91.

 version 0.9.5 (svn revision 1031)
 ================================
 * fix for compound variables on python 3.2.

 * fix slicing of masked MFDataset variables (issue 83).

 * round to nearest integer after packing with scale_factor and add_offset
   (instead of truncation) (issue 84).

 * if add_offset missing, but scale_factor present, assume add_offset zero.
   if scale_factor missing, but add_offset present, assume scale_factor one.
   (this is consistent with unidata recommendations - issue 86).

 * only try to convert strings to bytes for python 3 so Dataset can
   be subclassed (issue 87).

 version 0.9.4 (svn revision 1018)
 ================================
 * tested with python 2.7.1/3.1.3 using netcdf 4.1.2 and hdf5 1.8.6.

 * Added a 'default_encoding' module variable that controls how unicode
 strings are encoded into bytes. Default is 'ascii'.

 * now works on Python 3.

 * netCDF3 module removed. If you still need it, get it from netCDF4 0.9.3.

 * regenerated C source with Cython 0.14.1.

 * Added a MFTime class. Provide a unified interface to MFDataset time
 variable using different time units.

 * Fixed bug in netcdftime (issue 75) that occurs when time specified is within
 one second of the end of the month.

 * on unix-like systems, the environment variable USE_NCCONFIG can be set to
 tell setup.py to use the nc-config script installed by netcdf to figure out
 where all the libs and headers are (without having to specify NETCDF_DIR,
 HDF5_DIR, etc).  Only works with netcdf 4.1.2.

 version 0.9.3 (svn revision 930)
 ================================
 * fix chunk sizes bug (chunk sizes pointer should be size_t, not int).  Fixes
 issue 66. Added test in tst_compression.py

 * fixed writing of data with missing values with scale/offset packing. Added
 test (tst_masked2.py).

 * fix iso8601 regex in netcdftime date parser so it can parse 'hours since 1-1-1 ...'
 (year had to be 4 digits previously)

 version 0.9.2 (svn revision 907)
 ================================
 * fix netcdftime bug with '360_day' calendar. Fixes issue 59.

 * make sure scalar slice of 1d variable returns array scalar (not array of
   shape (1,)). Fixes issue 57.

 * updated date parser in netcdftime.  Can now handle units like
   "seconds since 1970-01-01T00:00:00Z".

 * added support in setup.py for specifying the locations of the HDF5/netcdf-4 headers
   and libs separately with environment variables (HDF5_INCDIR, HDF5_LIBDIR).i
   Patch contributed by Patrice Dumas.

 * add masked array support to num2date (dates for missing times set to None).

 * add chunk_cache keyword to createVariable.  HDF5 default is 1mb, which can
   cause problems when creating 1000's of variables.  In such cases,
   chunk_cache can be reduced, or set to zero.  

 * add set_var_chunk_cache and get_var_chunk_cache Variable methods.

 * raise AttributeError when trying to set _FillValue attribute (it can only
   be reliably set on variable creation, using the fill_value keyword to
   createVariable).

version 0.9.1 (svn revision 879)
================================
 * raise ImportError if netcdf-4 < 4.1.1 or hdf5 <= 1.8.4.

 * add __netcdf4libversion__ and __hdf5libversion__ module variables.

 * make sure data is not truncated to integers before scale_factor and
   add_offset is applied (issue 46).

 * fix bug in date2num with noleap calendar in netcdftime (issue 45).

 * fix bug in 360day calendar in netcdftime (issue 44).

 * python 2.4 compatibility restored (by modifying OrderedDict).  Fixes issue
   37.

 * make sure int64 attributes cast to int32 when format=NETCDF4_CLASSIC. This
   was causing tst_multifile.py to fail on 64-bit platforms.

 * fix tutorial.py to cast 64 bit integers to 32 bit when writing to 32-bit
   integer vlen (was causing tutorial.py to fail on 64-bit platforms).

 * remove nose dependency from tst_netcdftime.py.

version 0.9 (svn revision 846)
==============================
 * fixed bug (issue 30) with date2index occurring with dates outside the support.

 * make sure that auto masking works with MFDataset.

 * fix bug (issue 34) when slicing MFDataset variables with dimensions of
   length 1.

 * used ordered dictionaries for variables, dimensions, groups etc to
   preserve creation order (makes it easier to copy files, fixes issue 28).

 * change auto_maskandscale default to True.  This means data will
   automatically be converted to and from masked arrays.  Data scaled as short
   integers using the scale_factor and add_offset attributes will also be
   automatically converted to/from float arrays.

 * add setncattr, getncattr, delncattr methods (for setting/getting/deleting netcdf
   attributes with names that clash with the reserved python attributes).

version 0.8.2 (svn revision 769)
================================
 * compound type tests re-enabled. Compound and vlen types now fully
   supported in netcdf-4.1-beta2.

 * make sure data retrieved from a netCDF variable is not coerced to a python
   scalar (it should remain a numpy scalar array).

 * fix docs to point out that an unlimited dimension can be created by setting
   size to *either* None or 0 in createDimension.

 * fix another slicing corner case.

 * remove auto pickling/unpickling into vlen strings (too cute, sometimes
   produced surprising results).

version 0.8.1 (svn revision 744)
================================
 * added 'cmptypes' and 'vltypes' Group/Dataset attributes, which contain
   dictionaries that map the names of compound and vlen types to 
   CompoundType and VLType instances.

 * Experimental variable-length (vlen) data type support added.

 * changes to accomodate compound types in netcdf-4.1-beta snapshots.
   Compound types now work correctly for snapshots >= 20090603.

 * Added __len__ method and 'size' property to Variable class.

 * In date2index, replaced the brute force method by the bisection method and
   added a 'select' keyword to find the index of the date before, after or
   nearest the given date. 

 * Fixed bug occurring when indexing with a numpy array of length 1.
   
 * Fixed bug that occured when -1 was used as a variable index.

 * enabled 'shared access' mode for NETCDF3 formatted files (mode='ws',
   'r+s' or 'as'). Writes in shared mode are unbuffered, which can 
   improve performance for non-sequential access.

 * fixed bug in renameVariable that caused failure when new name is longer
   than old name, and file format is NETCDF3_64BIT or NETCDF3_CLASSIC.

version 0.8 (svn revision 685)
==============================
 * added 'stringtoarr' utility function for converting python strings
   to numpy character arrays of a specified size.

 * initial support for compound data types (which are mapped to structured
   numpy arrays). Compound data types are created with the createCompoundTYpe
   Dataset or Group method. Both attributes and variables can be compound
   types.

 * make sure 64-bit integer attributes converted to 32 bits when writing to
   a NETCDF3 formatted file.

 * added nc4tonc3 utility for converted NETCDF4_CLASSIC files to NETCDF3_64BIT
   files (useful for sharing data with colleagues that don't have netcdf-4
   capable clients).

version 0.7.7 (svn revision 626)
================================

 * David Huard reworked fancy indexing - it is now much more efficient and
   less of a memory hog. Now works differently than numpy fancy indexing - 1d
   arrays of boolean or integer indices work independently on each dimension.
   This enables things like:
   >>> tempdat = temp[[0,1,3],lats>0,lons>0]
   (retrieves 1st, 2nd and 4th levels, all Northern Hem. and Eastern
    Hem. grid points - note that this would raise an IndexError in numpy)

* added opendap test (tst_dap.py).

* bugfix for nc3tonc4 utility.

* fix MFDataset.Variable. __getattr__ to raise AttributeError instead of
  KeyError when attribute not found.

* netcdftime version number upped to 0.7.

version 0.7.6 (svn revision 574)
================================

* added date2index function, courtesy of David Huard, which finds the indices
  in a netCDF time variable corresponding to a sequence of datetime instances.

* make _get_att/_set_att raise AttributeError instead of RuntimeError, so that
  getattr(object, 'nonexistantattribute', None) works.  (thanks David Huard)

* v[:] = data now works along unlim dim, i.e. you can do this:

  file = Dataset('test.nc', "w")
  file.createDimension("time", None)     # unlimited dimension
  var = file.createVariable("var", 'd', ("time",))
  # you used to have to do this
  #var[0:10] = numpy.arange(10)
  # but now you can simply do this
  var[:] = numpy.arange(10)  

version 0.7.5 (svn revision 549)
================================
* return a scalar array, not a python scalar, when a slice returns a single
  number. This is more consistent with numpy behavior, and fixes a bug
  in MFDataset slicing.
* added 'exclude' parameter to MFDataset.__init__
* added set_auto_maskandscale method to MFDataset variables.

version 0.7.4 (svn revision 540)
================================
* ensure all arithmetic is done with float64 in netcdftime (Rob Hetland).
* fixes for netcdf-4.0-beta2 ('chunking' keyword to createVariable
  replaced by 'contiguous').  Now works with netcdf-4.0-beta2 and hdf5-1.8.0
  final, but is incompatible with netcdf-4.0-beta1.

version 0.7.3.1 (svn revision 507)
==================================
* netCDF3 docs were missing from 0.7.3.
* make sure quantization function preserves fill_value of masked arrays.

version 0.7.3 (svn revision 501)
================================
* MFnetCDF4 module merged into netCDF4 and netCDF3 (now
  called MFDataset).
* added netCDF3 module for those who can't install the netCDF 4 lib.
* added set_auto_maskandscale Variable method to enable automatic
  packing and unpacking of short integers (using scale_factor
  and add_offset attributes) and automatic conversion to/from
  masked arrays (using missing_value or _FillValue attribute)
  on a per-variable basis.
  var.set_auto_maskandscale(True) turns automatic
  conversion on (it is off by default).
* automatically pack/unpack short integer variables 
  if scale_factor and add_offset variable attributes are set.
* added support for masked arrays.  If you try to write a masked
  array to a variable with the missing_value or _FillValue attributes
  set, the masked array is filled with that value before being written
  to the file.  If you read data from a variable with the missing_value
  or _FillValue attribute set, a masked array is returned with the
  appropriate values masked.
* added date2num and num2date functions.
* added capability to use 'fancy indexing' with variable objects 
  (i.e. using sequences of integers or booleans in slices). WARNING:
  if a sequence of integers or booleans is used to slice a netCDF4
  variable, all of the data in that dimension is read into a numpy
  array, and then the sequence is used to slice the numpy array,
  returning just the requested elements to the user.  This can
  potentially gobble a lot of memory and degrade performance 
  (especially if 'fancy indexing' is done on the left-most dimension).
* added convenience functions stringtochar and chartostring for
  converting character arrays to arrays of fixed-length strings and
  vice-versa.   Example usage in examples/test_stringarr.py.

20070826 - version 0.7.1 (svn revision 400)
===========================================
* added 'endian()' and 'chunking()' Variable methods (to inquire about
  endian and chunking variable settings).

* 'ndim' attribute was not public (so it couldn't be accessed from python).
  Fixed.

* added 'endian' kwarg to createVariable (to set the endian-ness
  used in the HDF5 file).

* can now manually set HDF5 chunksizes for each dimension at
  variable creation, using 'chunksizes' kwarg to createVariable.

* added "getlibversion()" function to get info about version
  of netcdf-4 library used to build module.

* if a variable has an unsupported datatype (such as 'compound', or
  'vlen'), then instead of raising an exception, just skip it. 
  Print a useful error message when an attribute with an unsupported
  datatype is accessed.

* if variable dimension is specified as 'dimname' or ('dimname')
  in createVariable, it is automatically converted to a tuple ('dimname',).
  Better error messages when specified dimension can't be found.

* createVariable accepts numpy dtype object as datatype.  dtype variable
  attribute is now a numpy dtype object.

20070723 - version 0.7 (svn revision 361)
=========================================
* renamed MFnetCDF4_classic --> MFnetCDF4.

* eliminated netCDF4_classic module (all file formats handled by
  netCDF4 module now).

* removed all user-defined data type stuff (it was hacky and made 
  the code too complex - wait till there is a real use case to 
  refactor and put back in).

* added 'ndim' variable attribute (number of variable dimensions).

20070424 - version 0.6.3 (svn revision 302)
===========================================
* passes all tests with netcdf-4.0-beta1/hdf5-1.8.0-beta1.

* if slice index is not a slice object, assume it's an integer (and
  try to convert to one if it is not).  This allows numpy scalar arrays
  to work as slice indices.

* (netCDF4_classic only) try to make sure file is not left in 'define mode' 
  when execption is raised.

* if slicing a variable results in a array with shape (1,), just return 
  a scalar (except for compound types).

* added instructions for using the netCDF4_classic module to serve
  data over http with the DAP using pydap (http://pydap.org).

* added --quiet and --chunk options to nc3tonc4.

* Turned off zlib compression by default so as not to violate the
  'principle of least surprise'.  Shuffle filter still activated 
  by default when zlib compression turned on.

* Fixed bug in fletcher32 checksum activation call.  Renamed compression()
  variable method to filters(), include fletcher32 checksum flag in output.

* added utility for converting GRIB1 files to compressed
  NETCDF4_CLASSIC files (requires PyNIO).

* added 'compression()' variable method that returns a dict with
  compression filter parameter settings for that variable. (rev 237)

* reimplemented 'shape' and 'dimensions' variable attributes as
  properties.

* fixed bug when 'chunking' keyword in createVariable was set to 'sub' 
  (caused Bus Error on MacOS X).

* Setting 'shuffle=0' keyword in createVariable was turning off
  zlib compression filter instead of shuffle filter.  Fixed.

20070213 - version 0.6.2
========================
* updated for compatibility with netcdf-4.0-alpha18 and hdf5 1.8.0alpha5
  (shared dimensions actually work now).

* netCDF4.createVariable can now use old single character Numeric typecodes
  for datatype specification.

* Improvements to MFDataset (now called MFnetCDF4_classic) by Rob Hetland.
 
20061121 - version 0.6.1
========================
* bugfixes for negative strides.

* bugfix for empty string attributes.

* support for shared dimensions (variables can use dimensions defined 
  only in a parent group).  This doesn't actually work yet, because of
  a bug in netcdf-4.0-alpha17.

* now requires Pyrex (C source files generated on the fly when setup.py
  is run).

20061003 - version 0.6
======================
* if fill_value keyword to createVariable is set to the Boolean
  False (not an integer that evaluates to False), no pre-filling
  is done for that variable.

* updated to be compatible with netcdf-4.0-alpha17.
  Can now install pure-python netcdftime separately with setup-netcdftime.py.
  netcdftime will try to use numpy, but fall back to Numeric if numpy 
  not installed.

* generated source files with a version of pyrex
  (from http://codespeak.net/svn/lxml/pyrex/) that produces
  extensions compatible with python 2.5.

* added new module for multi-file access of NETCDF3 and NETCDF4_CLASSIC
  files (MFDataset). Based on CDFMF from pycdf.

* implement negative strides in variable slicing
  (feature missing from Scientific.IO.NetCDF). Now variables support
  full python extended slicing syntax.

20060925 - version 0.5.1
========================
* on 64-bit systems integer attributes in netCDF4_classic failed, since there
  is no 64-bit integer data type. Fixed by downcasting to 32-bit integer.

20060920 - version 0.5
======================
* Compound type support! (members must be fixed data primitive types -
  no user-defined types or NC_STRING variables allowed).  Attributes
  are still restricted to primitive data types (no vlen or compound
  type attributes).

* Assigning single values to a slice now does the Right Thing, i.e.
  >>> data[:] = 1
  fills all the elements with 1 (instead of raising an IndexError).

* Tested with numpy 1.0b5, netcdf-4.0-alpha16, HDF5 1.7.52 alpha.

* Added renameDimension and renameVariable methods to Dataset and Group classes.

* netCDF attributes can be deleted using python del (i.e. 'del dset.foo').

* Moved examples from test and test_classic to examples and 
  examples_classic directories.

* Added proper unit tests (in test and test_classic directories).

* NULL characters are removed from text attributes.

* Variable _FillValue can be set using new keyword argument 'fill_value' 
  to createVariable Dataset and Group method.

* docstrings now formatted with epydoc (http://epydoc.sf.net).

* improved Scientific.IO.NetCDF compatibility for netCDF4_classic
  (typecode method, ability to use old Numeric typecodes).

* zlib=False or complevel=0 disables shuffle filter in createVariable.

* subversion repository hosted on Google projects
  (http://code.google.com/p/netcdf4-python/).

* examples_classic/bench2.py is a performance comparison with
  Scientific.IO.NetCDF (the numpy version provided by pynetcdf).

* __dict__ attribute of Dataset, Group or Variable provides a python
  dictionary with all netCDF attribute name/value pairs (just like
  Scientific.IO.NetCDF).

20060710 - version 0.4.5
========================
* fixed to work with recent svn versions of numpy

* Now requires at least numpy 0.9.8.

* Raise a AttributeError if user tries to rebind a private attribute
  (like 'variables', 'dimensions' or 'dtype').

20060629 - version 0.4.4
========================
* fixed to work with netcdf-4.0-alpha14.

* automatically cast _FillValue attribute to variable type, to
  avoid surprising error message.

20060320 - version 0.4.3
========================
updated netcdftime module yet again
added 'all_leap'/'366_day' and '360_day' calendars.
netCDFTime class renamed utime, fwd and inv methods
renamed date2num and num2date. These methods can now handle
numpy arrays as well as scalars.
a 'real' python datetime instance is returned if calendar
is gregorian, otherwise a 'datetime-like' instance is returned
(python datetime can't handle funky dates in 'all_leap' and '360_day'
calendars).

20060316 - version 0.4.2
========================
udunits module replaced by pure python version, renamed 'netcdftime'
No longer requires udunits library. Includes 4 calendars
('julian','standard'/'gregorian','proleptic_gregorian','noleap'/'365_day').
Calendar names and their interpretations follow the CF metadata convention.

20060310 - version 0.4.1
========================
udunits module included for doing time conversions.

20060306 - version 0.4
======================

netCDF4_classic module can now write NETCDF3_CLASSIC, NETCDF4_64BIT
as well as NETCDF4_CLASSIC files.  The file format is given as
an optional keyword to the Dataset constructor ('NETCDF4_CLASSIC'
is the default).  Preliminary work on compound types done - but
awaiting the next alpha of the netCDF 4 library to complete (bugs
in alpha12 prevent it from working properly if the compound type
has fields which are arrays).

20060217 - version 0.3.1
========================
refactored user-defined data type support - user-defined
data types are now described by an instance of the class
UserType.  usertype and usertype_name keyword args
eliminated from createVariable.

20060214 - version 0.3
======================
support for variable length strengths (typecode = 'S') and
variable-length, or 'ragged' arrays (vlen user-defined datatype).
Arrays of python objects can be saved as pickled strings with
datatype = 'S'.

20050128 - version 0.2.5
========================
added support for scalar variables (and assignValue, getValue
Variable methods for Scientific.IO.NetCDF compatibility).

20051123 - version 0.2.4
========================
numpy 0.9.4 compatibility
Changed data type codes from ('d', 'f', 'i', 'h', ...) to
('f8', 'f4', 'i4', 'i2', ...).

20050110 - version 0.2.3
========================
added ellipsis slicing capability

20050106 - version 0.2.2
========================
changed scipy_core to numpy.

20051228 - version 0.2.1
========================
bugfixes, added 'nc3tonc4' utility to convert netCDF version 3 files
to NETCDF4_CLASSIC files (with compression).  The converted files
can be read from netCDF 3 clients that have been re-linked to the netCDF 4
library. 'chunking' keyword added to createVariable in netCDF4 module.

20051224 - version 0.2
======================
Added netCDF4_classic module - which creates files in NETCDF4_CLASSIC
format.  These files are compatible with netCDF 3 clients which have
been linked against the netCDF 4 lib.  This module does not use any
new features of the netCDF 4 API except zlib compression.  Unlike
any other netCDF 3 python client, it can transparently compress data
with zlib compression and the HDF5 shuffle filter.

20051222 - version 0.1
======================
First release.  Supports groups, multiple unlimited dimensions, zlib
compression (plus shuffle filter and fletcher32 checksum) and all new
primitive data types.  No support for user-defined data types yet.