Log Message:

	Running test.
	
Skip to content
This repository

    Explore
    Features
    Enterprise
    Blog

25
89

    48

Unidata/netcdf4-python

netcdf4-python/Changelog
@jswhit jswhit 9 days ago reword

5 contributors
@jswhit
@shoyer
@huard
@matthew-brett
@cpaulik
1078 lines (814 sloc) 45.551 kB
 version 1.1.9 (not yet tagged)
 ==============================
 * fix for issue #391 (data is already byte-swapped to native
   endian format by the HDF4 library).
 * fix for issue #415 (copy.deepcopy does not work on
   netcdftime datetime object).
 * fix for issue #420 - len(v) where v is a scalar variable returned
   unexpected IndexError, now returns "TypeError: len() on unsized object"
   (same as numpy does for len() on a scalar array).
 * translate docstrings from epydoc markup to markdown, so
   pdoc can be used (epydoc is dead).

 version 1.1.8 (tag v1.1.8rel)
 =============================
 * v[...] now returns a numpy scalar array (not just a scalar) when
   v is a scalar netcdf variable (issue #413).
 * unix-like paths can now be used in createVariable and createGroup.
   v = nc.createVariable('/path/to/var1',('xdim','ydim'),float)
   will create a Variable named 'var1', while also creating the Groups
   'path' and 'path/to' if they do not already exist.
   Similarly, g = nc.createGroup('/path/to') acts like 'mkdir -p' in unix, creating
   the Groups 'path' and '/path/to', if they don't already exist.
   Users who relied on nc.createGroup(groupname) failing when the
   group already exists will have to modify their code, since nc.createGroup
   will now return the existing group instance.
   Dataset.__getitem__ also added.  nc['/path/to'] returns
   a Group instance, and nc['/path/to/var1'] returns a Variable
   instance.  
 * change minimum required numpy to 1.7.0, fix so all tests pass with 1.7.0.
   Added travis tests for minimum required cython, numpy (issue #404).
 * enable abbreviations to time units specification, as allowed in CF (issue
   #402). Now, instead of just 'seconds' and 'seconds', 'secs', 'sec' and 's'
   are also allowed (similar to minutes, days and hours).
 * install utility scripts in utils directory with setuptools entry points 
   (pull request #392 from @mindw).  Code for utilities moved
   to netCDF4_utils.py - makes utilities more windows-friendly.
 * make sure booleans are treated correctly in setup.cfg. Add
   use_cython (default True) to setup.cfg.  If set to False, then
   cython will not be used to compile netCDF4.pyx (existing netCDF4.c 
   will be used instead).
 * use "from Cython.Build import cythonize" instead of
   "from Cython.Distutils import build_ext" in setup.py (issue #393)
   to conform to new cython build mechanism (CEP 201, described at
   https://github.com/cython/cython/wiki/enhancements-distutils_preprocessing).
 * unicode attributes now written as strings, not bytes (using
   nc_put_att_string instead of nc_put_att_text, issue #388).
 * add __orthogonal_indexing__ attribute to Variable, Dataset and Group (issue \ 
#385) to 
   denote that Variable objects do not follow numpy indexing semantics for \ 
integer and
   boolean array indices. 
 * make sure application of scale_factor and add_offset works correctly when
   scale_factor not given (issue #381).
 * add man pages for nc3tonc4, nc4tonc3, ncinfo in man directory.
   Not installed by setup.py (contributed by Ross Gammon, issue #383).
 * replace tabs with spaces by running reindent.py on all *.py and *.pyx files
   (issue #378).
 * refactor netCDF4_utils and netCDF4 module into netCDF4 package.
   Refactoring effectively removes netCDF4 utils private attributes from
   netCDF4 namespace, so has the potential to break code using private
   attributes (issue #409).

 version 1.1.7 (tag v1.1.7rel)
 =============================
 * check to make sure cython >= 0.19 is available before trying
   to use it (otherwise compilation with fail).  Issue 367.
 * add ipython notebooks from Unidata workshop in examples directory.
 * fix ellipsis variable slicing regression (issue 371).
 * release the Global Interpreter Lock (GIL) when calling the C 
   library for read operations.  Speeds up multi-threaded reads
   (issue 369).  Caution - the HDF5 library may need to be compiled 
   with the threadsafe option to ensure that global data structures
   are not corrupted by simultaneous manipulation by different threads.
 * Make sure USE_NCCONFIG environment variable takes precedence over value
   of use_ncconfig in setup.cfg.  With this change, 'pip install netCDF4'
   with USE_NCCONFIG=1 will use environment variables to find paths to 
   libraries and include files, instead of relying on nc-config (issue #341).

 version 1.1.6 (tag v1.1.6rel)
 =============================

 * fix for issue 353 (num2date can no longer handle units like 'hours since
   2000-01-01 0').
 * fix for issue 354 (num2date no longer supports multi-dimensional arrays).
 * fix for spurious UserWarning about endian-ness mismatch (issue 364).
 * make calendar name keyword for num2date/date2num case insensitive
   (issue 362).
 * make sure units parser returns time-zone naive datetime instance that
   includes UTC offset (issue 357).  UTC offset was applied incorrectly in 
   netcdftime.date2num and num2date.  No longer need to depend on
   python-dateutil.

 version 1.1.5 (tag v1.1.5rel)
 =============================
  
 * add dependency on python-dateutil in setup.py and install docs.
 * use python datetime in num2date and date2num whenever possible. Remove
   duplicate num2date and date2num functions from netcdftime. Addresses issue
   #344. Add microsecond capability to netcdftime.datetime. Roundtrip
   accuracy of num2date/date2num now down to less than a millisecond.
 * use nc-config by default to find dependencies.  setup.py modified to handle \ 
failure
   to find nc-config more gracefully (issue #340). If you wish to use env vars \ 
to point
   to the libs, you must first move the setup.cfg file out of the way
   (rename it to setup.cfg.save), or set USE_NCCONFIG to 0.
 * if endian-ness of variable is specified, adjust datatype to reflect this when \ 
opening
   a file (issue 346).
 * fix for issue #349 (seconds outside the range 0-59 in netcdftime.num2date).

 version 1.1.4 (tag v1.1.4rel)
 =============================

 * speedup conversion of array indices to slices (issue #325).
 * fix for issue #330 (incorrect values for seconds returned by netcdftime).
 * fix reading of scalar vlen variables (issue #333).
 * setting fill_value=False in createVariable for vlen and compound variables 
   now does nothing, instead of causing an error when the Dataset is closed
   (issue #331).
 * cython will regenerate netCDF4.c when install is run, not just build.
   Makes 'pip install' do the right thing when cython is installed (issue #263).

 version 1.1.3 (tag v1.1.3rel)
 =============================

 * checked in _datetime.c to git (resolves issue #315).  Note - _datetime.c
   was *not* included in the 1.1.2 release.
 * Changed __str__ to __repr__ in MFDataset, to be consistent with Dataset
   (issue #317). IPython uses __repr__ to make use-friendly human-readable summaries
   of objects in the terminal.

 version 1.1.2 (tag v1.1.2rel)
 =============================

 * fix for issue 312 (allow slicing with objects that can be cast to ints).
 * indexing netCDF variables with integer sequences and boolean arrays now
   behave the same way (integer sequences are converted to boolean arrays
   internally). Addresses issue #300.  Since indexing using integer sequences
   does not behave exactly as before, some client code may break.  For example,
   previously when integer index arrays had the same length, and that length
   was equal to the number of dimensions of the array being indexed,
   netcdf4-python mirrored the numpy indexing behavior and treated the elements
   of the index arrays as individual sets of integer indices.  This special
   case has been removed. An IndexError is now raised when the new behavior
   would produce a different result than the old, i.e. when the
   indices in an integer sequence are not sorted, or there are duplicate
   indices in the sequence.
 * fix for issue #310 (masked arrays not returned correctly when variable
   has non native endian-ness).
 * fix for issue #306 (slicing variable with "-1" when there is only
   one element along that dimension).
 * Improved speed of num2date and date2num for standard, julian, gregorian
   and proleptic gregorian calendars by vectorizing the functions. See Issue #296
 * Fix for issue #301 ("Datestring parser chokes on years with extra \ 
space").
 * Add name property for Dimension, Variable and Group instances (to access string
   name associated with instance).
 * Allow for null byte attributes (so _FillValue='\x00' can be set manually).
   Issue 273.
 * Added __repr__ (matching __str__) for all types (pull request #291).
   IPython uses __repr__ to make use-friendly human-readable summaries
   of objects in the terminal.