Merge branch 'master' of https://github.com/ARM-DOE/pyart into doc_rt…

…d_enhancement

README.rst

@@ -78,7 +78,9 @@ Install
 =======
 
 The easiest method for installing Py-ART is to use the conda packages from
-the latest release. To do this you must download and install 
+the latest release and use Python 3, as Python 2 support ended January 1st,
+2020 and many packages including Py-ART no longer support Python 2.
+To do this you must download and install 
 `Anaconda <https://www.anaconda.com/download/#>`_ or 
 `Miniconda <https://conda.io/miniconda.html>`_.
 With Anaconda or Miniconda install, it is recommended to create a new conda

continuous_integration/install.sh

@@ -17,7 +17,6 @@ conda update -q conda
 ## Create a testenv with the correct Python version
 conda env create -f continuous_integration/environment-$PYTHON_VERSION.yml
 source activate testenv
-# TODO install cylp and  glpk in Python 2.7
 
 # upgrade pip
 pip install --upgrade pip

guides/CI_SETUP.rst

@@ -13,7 +13,7 @@ Services
 Py-ART's CI setup is hosted on two services, `Travis CI`_ and `AppVeyor`_.
 Both of these are hosted continuous integration services that are free for open
 source projects. Travis CI runs the tests on a Linux platform and AppVeyor
-runs on Windows.  Currently no tests are run on macOS although Travis CI does
+runs on Windows. Currently no tests are run on macOS although Travis CI does
 support this platform.
 
 .. _Travis CI : https://travis-ci.org/
@@ -24,35 +24,35 @@ Travis CI
 ---------
 
 The CI setup on Travis CI is controlled by the ``.travis.yml`` file in the
-repository root.  This file contains multiple sections which define the steps
-taken in a CI run.  The `Travis CI documentation`_ provide details on these
+repository root. This file contains multiple sections which define the steps
+taken in a CI run. The `Travis CI documentation`_ provide details on these
 sections and Travis CI in general.
 
 Briefly the sections in the file include:
 
 - An encrypted environment variable used to push the documentation built
   upon each merge to the master branch to the `pyart-docs-travis`_ repo.
 - A matrix section which defines the various combinations of Python versions
-  and other parameters which will be tested.  Currently Python 2.7, 3.4, 3.5
-  and 3.6 are tested with a second Python 2.7 run also testing that the
+  and other parameters which will be tested. Currently Python  3.6, 3.7
+  and 3.8 are tested with a second Python 3.8 run also testing that the
   package can be built from a conda recipe and that the docs build.
 - An install section which executes the ``install.sh`` script in the
   continuous_integration directory.
-- A script section which runs the tests using ``nosetests``
+- A script section which runs the tests using ``pytest``
 - A after_success section which reports test coverage to `Coveralls`_.
   Currently nothing is done with this information but it may be useful later
   to view how test coverage has changed over time.
 - The after_success section also runs the ``build_docs.sh`` script in the
-  continuous_integration directory for the appropriate matrix element.  On
-  non-merge commits this tests that the documentation can be build.  On merge
+  continuous_integration directory for the appropriate matrix element. On
+  non-merge commits this tests that the documentation can be build. On merge
   commits this builds the docs and uploads them to the `pyart-docs-travis`_
   repository which makes them available on the `Py-ART documentation page`_.
 
 The bulk of the test environment setup is done by the ``install.sh`` script in
-the continuous_integration directory.  This script downloads, installs and
+the continuous_integration directory. This script downloads, installs and
 configures the latest version of Miniconda3, creates a conda environment for
 the Python version being tested with the various requirements specified in the
-environment-XXX.yml file, and builds Py-ART in that environment.  For the
+environment-XXX.yml file, and builds Py-ART in that environment. For the
 matrix element where Py-ART is build from a conda recipe, the if/then block
 installs conda-build, builds Py-ART from the recipe in the repository and
 installs this conda package into the test conda environment.
@@ -73,12 +73,12 @@ enviroment-X.X.yml files in the continuous_integration directory. Over time
 these may need to be updated.
 
 To debug Travis CI issue use the logs provided for the builds themselves to see
-where the issues are occurring.  Un-commenting the ``set -x`` line in the
-``install.sh`` script may be helpful when debugging Travis CI issue.  With this
+where the issues are occurring. Un-commenting the ``set -x`` line in the
+``install.sh`` script may be helpful when debugging Travis CI issue. With this
 line uncommented the commands being executed are echoed in the log.
 
 Additionally, it can be helpful to run the CI builds locally to quickly test
-out changes to the setup and try out specific commands.  Travis CI provides
+out changes to the setup and try out specific commands. Travis CI provides
 the docker `images`_ used by their service and instructions on how to
 `run these locally`_.
 
@@ -89,14 +89,14 @@ the docker `images`_ used by their service and instructions on how to
 AppVeyor
 --------
 
-The second CI service used by Py-ART is `AppVeyor_`.  This setup is controlled
+The second CI service used by Py-ART is `AppVeyor_`. This setup is controlled
 by the ``appveyor.yml`` file in the repository and builds and runs the unit
-tests on a Windows platform.  This services is slower than Travis CI so only
-two Python versions are tested (2.7 and 3.5).  All setup and configuration is
+tests on a Windows platform. This services is slower than Travis CI so only
+two Python versions are tested (3.7 and 3.8). All setup and configuration is
 done in the ``appveyor.yml`` file besides a setup function used by many
 projects which can be found in the ``run_with_env.cmd`` file in the
-continuous_integration/appveyor directory.  This file contains comments which
-explain the steps being performed and some debugging details.  Refer to the
+continuous_integration/appveyor directory. This file contains comments which
+explain the steps being performed and some debugging details. Refer to the
 `AppVeyor documentation`_ for details.
 
 .. _AppVeyor : https://www.appveyor.com/

guides/HOWTO_RELEASE.rst

@@ -9,9 +9,9 @@ Make the tagged git release
 ---------------------------
 
 Change the ``ISRELEASE`` line in the **setup.py** file in the root directory to 
-True.  Check this change into git and create a tagged commit with the version
-string.  The commit message should be "REL: Release X.X.X" with the proper
-version filled in.  This commit and the tag should be pushed directly to the
+True. Check this change into git and create a tagged commit with the version
+string. The commit message should be "REL: Release X.X.X" with the proper
+version filled in. This commit and the tag should be pushed directly to the
 ARM-DOE repository on GitHub. This can be done using the following, assuming
 upstream points to https://github.com/ARM-DOE/pyart.git.
 
@@ -42,7 +42,7 @@ that this file can be used to install a working version of Py-ART.
 
 Once tested the tar.gz source distribution file can be uploaded to PyPI using
 `twine <https://pypi.python.org/pypi/twine>`_, installing if needed, with the 
-command ``twine upload dist/*``.  This command will prompt the user for their
+command ``twine upload dist/*``. This command will prompt the user for their
 PyPI credentials.  After uploading check that this file is available for
 download from 
 `pypi.python.org/pypi/arm_pyart <https://pypi.python.org/pypi/arm_pyart>`_.
@@ -66,7 +66,7 @@ Prepare the release notes
 -------------------------
 
 Prepare a short description of the changes, improvements and deprecations in 
-the new release.  The format used in previous releases makes a good template
+the new release. The format used in previous releases makes a good template
 for this text. The following git commands can be used to list all merge commits
 and the contributors to a particular version which may be helpful when drafting
 these notes.
@@ -91,11 +91,11 @@ Create and upload version specific documentation
 ------------------------------------------------
 
 The Py-ART documentation page contains links to documentation and examples from
-the latest release.  These pages must be updated when a new release is made.
+the latest release. These pages must be updated when a new release is made.
 To build the documentation for a release, go to the doc directory and run the
-``rebuild_full_docs.sh`` shell script.  Note that this should be done while the
+``rebuild_full_docs.sh`` shell script. Note that this should be done while the
 release commit is checked out and on a system with all optional requirements
-for the examples and sphinx install.  Additionally, the files needed for the
+for the examples and sphinx install. Additionally, the files needed for the
 examples must be copied into the correct locations in the examples directory.
 
 Once the documentation has been built, the contents of the doc/build/html
@@ -104,9 +104,9 @@ in the dev directory. Typically this process is best done in a directory where
 the ARM-DOE repository has been cloned and the gh-pages branch checked out.
 
 First, the current dev directory should be moved to a versioned directory using
-``git mv dev v1_7`` and a commit made.  Then create an empty ``dev`` directory 
-and copy all the files from the doc/build/html directory.  Check these new 
-files into git and create a commit.  Finally, push these changes to the ARM-DOE
+``git mv dev v1_7`` and a commit made. Then create an empty ``dev`` directory 
+and copy all the files from the doc/build/html directory. Check these new 
+files into git and create a commit. Finally, push these changes to the ARM-DOE
 repository. After a few minutes check that the new documentation is available
 at `http://arm-doe.github.io/pyart/dev/ <http://arm-doe.github.io/pyart/dev/>`_.
 
@@ -117,8 +117,8 @@ Prepare the repository for the next version
 To begin the development of the next version of Py-ART the ``MINOR`` variable
 should be incremented and ``ISRELEASED`` should be changed back to False.  
 Make this change to the ``setup.py`` file in the root directory and create a 
-commit with the message ``MAINT: start v1.9+dev``.  See the commit for the 
-`start of the 1.8+dev`_ as an example.  Push this commit directly to the
+commit with the message ``MAINT: start v1.9+dev``. See the commit for the 
+`start of the 1.8+dev`_ as an example. Push this commit directly to the
 ARM-DOE repository.
 
 .. _start of the 1.8+dev : https://github.com/ARM-DOE/pyart/commit/e3d01232ac3787da38456152da6b9db3ab46d4d7
@@ -130,17 +130,17 @@ to remove them.
 Create conda packages on conda-forge
 ------------------------------------
 
-Conda package for Py-ART are created using the conda-forge feedstock.  To
+Conda package for Py-ART are created using the conda-forge feedstock. To
 create a new release first fork the `arm_pyart-feedstock`_ repository or update
-your fork.  Then clone the fork and checkout a new branch.  Update the first
+your fork. Then clone the fork and checkout a new branch. Update the first
 few lines of the ``recipe/meta.yaml`` with the new version number and a sha256
-hash of the file on PyPI. Check this change in as a new commit.  Next rerender
+hash of the file on PyPI. Check this change in as a new commit. Next rerender
 the feedstock using ``conda smithy rerender``, installing ``conda-smithy``
 if needed.  
 
 Finally, push the branch up to your fork and create a pull request against the
 conda-forge feedstock. The CI services will test the recipe by building conda
-packages on all three platforms and Python+NumPy combinations.  Once these
+packages on all three platforms and Python+NumPy combinations. Once these
 pass the pull request can be merged and the CI services will again build the
 packages and uploaded them to the conda-forge Anaconda cloud channel.
 
@@ -153,7 +153,7 @@ process works.
 Announce the release on the pyart-user mailing list
 ---------------------------------------------------
 
-Announce the new release on the `Py-ART mailing list`_.  Use a previous
+Announce the new release on the `Py-ART mailing list`_. Use a previous
 announcement as a template and include the release notes written above.
 
 .. _Py-ART mailing list : http://groups.google.com/group/pyart-users/

guides/contributors_guide.rst

@@ -38,15 +38,17 @@ please cite:
     DOI: http://doi.org/10.5334/jors.119
 
 Py-ART implements many published scientific methods which should *also* be
-cited if you make use of them.  Refer to the **References** section in the
+cited if you make use of them. Refer to the **References** section in the
 documentation of the functions used for information on these citations.
 
 
 Install
 -------
 
 The easiest method for installing Py-ART is to use the conda packages from
-the latest release.  To do this you must download and install 
+the latest release and use Python 3, as Python 2 support ended January 1st,
+2020 and many packages including Py-ART no longer support Python 2.
+To do this you must download and install 
 `Anaconda <https://www.anaconda.com/download/#>`_ or 
 `Miniconda <https://conda.io/miniconda.html>`_.  
 Then use the following command in a terminal or command prompt to install

guides/setting_up_an_environment.rst

@@ -73,7 +73,7 @@ if you want Jupyter Notebook to run in that enviroment with those packages::
 while that environment is activated. Another way to create a conda environment
 is by doing it from scratch using the conda create command. An example of this::
 
-        conda create -n pyart_env -c conda-forge python=3.6 arm_pyart netCDF4
+        conda create -n pyart_env -c conda-forge python=3.8 arm_pyart netCDF4
         cartopy scipy numpy matplotlib
 
 This will also create an environment called pyart_env that can be activate the

pyart/__init__.py

@@ -4,8 +4,6 @@
 
 """
 
-from __future__ import print_function
-
 # Detect if we're being called as part of Py-ART's setup procedure
 try:
     __PYART_SETUP__

pyart/_debug_info.py

@@ -6,7 +6,6 @@
 
 """
 
-from __future__ import print_function
 import os
 import sys
 

pyart/aux_io/sinarame_h5.py

@@ -14,8 +14,6 @@
 
 """
 
-from __future__ import print_function
-
 from datetime import datetime
 import glob
 import os

pyart/core/grid.py

@@ -21,7 +21,12 @@
 
 import numpy as np
 from netCDF4 import num2date
-import xarray
+
+try:
+    import xarray
+    _XARRAY_AVAILABLE = True
+except ImportError:
+    _XARRAY_AVAILABLE = False
 
 try:
     import pyproj
@@ -300,6 +305,12 @@ def to_xarray(self):
             in a one dimensional array.
             
         """
+
+        if not _XARRAY_AVAILABLE:
+            raise MissingOptionalDependency(
+                'Xarray is required to use Grid.to_xarray but is not '
+                 + 'installed!')
+
         lon, lat = self.get_point_longitude_latitude()
         z = self.z['data']
         y = self.y['data']

pyart/core/radar.py

@@ -20,7 +20,6 @@
 
 
 """
-from __future__ import print_function
 
 import copy
 import sys

pyart/core/tests/test_grid.py

@@ -1,7 +1,5 @@
 """ Unit Tests for Py-ART's core/grid.py module. """
 
-from __future__ import print_function
-
 import functools
 import pickle
 

pyart/core/tests/test_wind_profile.py

@@ -1,7 +1,5 @@
 """ Unit Tests for Py-ART's core/wind_profile.py module. """
 
-from __future__ import print_function
-
 import numpy as np
 from numpy.testing import assert_almost_equal
 import pytest

pyart/correct/despeckle.py

@@ -21,7 +21,6 @@
 
 """
 
-from __future__ import division
 import numpy as np
 from scipy.ndimage import label
 from scipy.signal import convolve2d

pyart/correct/phase_proc.py

@@ -38,8 +38,6 @@
 
 """
 
-from __future__ import print_function, division
-
 import copy
 from time import time
 

pyart/correct/tests/test_region_dealias.py

@@ -4,8 +4,6 @@
 # to recreate a dealias_plot.png file showing the before and after doppler
 # velocities
 
-from __future__ import print_function
-
 import warnings
 
 import pyart

pyart/correct/tests/test_unwrap.py

@@ -4,8 +4,6 @@
 # to recreate a dealias_plot.png file showing the before and after doppler
 # velocities
 
-from __future__ import print_function
-
 import numpy as np
 from numpy.testing import assert_allclose, assert_almost_equal
 import pytest

pyart/correct/unwrap.py

@@ -19,8 +19,6 @@
 
 """
 
-from __future__ import print_function
-
 import numpy as np
 
 from ..config import get_metadata, get_fillvalue

pyart/graph/cm.py

@@ -60,7 +60,6 @@
 # http://matplotlib.org/.
 # Copyright (c) 2012-2013 Matplotlib Development Team; All Rights Reserved
 
-from __future__ import print_function, division
 import warnings
 
 import matplotlib as mpl

pyart/graph/gridmapdisplay_basemap.py

@@ -12,8 +12,6 @@
 
 """
 
-from __future__ import print_function
-
 import warnings
 
 import numpy as np

pyart/graph/tests/test_radardisplay.py

@@ -5,8 +5,6 @@
 # in baseline_images directory. Current this test only determines if files can
 # be created, not that they are correct.
 
-from __future__ import print_function
-
 import warnings
 import datetime
 

pyart/io/_sigmetfile.pyx

@@ -23,8 +23,6 @@ A class and supporting functions for reading Sigmet (raw format) files.
     _unpack_ingest_header
 
 """
-from __future__ import print_function
-
 import struct
 import datetime
 import warnings