Merge pull request #1 from imagej/xarray

Fix Travis build

Author: mpinkert

.travis/build.sh

@@ -16,9 +16,9 @@ check () {
 }
 
 # -- create a test enviroment --
-conda env create -q -f environment.yml
+conda create -n imagej -y python=$TRAVIS_PYTHON_VERSION
 source activate imagej
-conda install -q -y python=$TRAVIS_PYTHON_VERSION
+conda env update -f environment.yml
 
 # -- ensure supporting tools are available --
 check curl git unzip
@@ -58,6 +58,9 @@ ij_dir=$HOME/Fiji.app
 echo "ij_dir = $ij_dir"
 python setup.py install
 
+# -- unset JAVA_HOME in case it was set --
+unset JAVA_HOME
+
 # -- run tests with local Fiji.app --
 python -O test/test_imagej.py --ij "$ij_dir"
 

README.md

@@ -11,43 +11,65 @@ see "ImageJ Server" below for details.
 
 1. Install [Conda](https://conda.io/):
     * On Windows, install Conda using [Chocolatey](https://chocolatey.org): `choco install miniconda3`
-    * On macOS, install Conda using [Homebrew](https://brew.sh): `brew install miniconda`
+    * On macOS, install Conda using [Homebrew](https://brew.sh): `brew cask install miniconda`
     * On Linux, install Conda using its [RPM or Debian package](https://www.anaconda.com/rpm-and-debian-repositories-for-miniconda/), or [with the Miniconda install script](https://docs.conda.io/projects/conda/en/latest/user-guide/install/linux.html).
 
-2. [Activate the conda-forge channel](https://conda-forge.org/docs/user/introduction.html#how-can-i-install-packages-from-conda-forge):
+2. Configure your shell for use with conda:
+   ```
+   conda init bash
+   ```
+   Where `bash` is the shell you use.
+   Then start a new shell instance.
+
+3. [Activate the conda-forge channel](https://conda-forge.org/docs/user/introduction.html#how-can-i-install-packages-from-conda-forge):
     ```
     conda config --add channels conda-forge
     conda config --set channel_priority strict
     ```
 
-3. Install pyimagej into a new conda environment:
+4. Install pyimagej into a new conda environment:
     ```
     conda create -n pyimagej pyimagej openjdk=8
     ```
 
-4. Whenever you want to use pyimagej, activate its environment:
+5. Whenever you want to use pyimagej, activate its environment:
     ```
     conda activate pyimagej
     ```
 
 ### Installation asides
 
-* If you want to use [scikit-image](https://scikit-image.org/) in conjunction, as demonstrated below, you can install it also via:
-
+* If you want to use [scikit-image](https://scikit-image.org/) in conjunction,
+  as demonstrated below, you can install it also via:
     ```
     conda install scikit-image
     ```
 
-* The above command installs pyimagej with OpenJDK 8; if you leave off the `openjdk=8` it will install OpenJDK 11 by default, which should also work, but is less well tested and may have more rough edges.
+* The above command installs pyimagej with OpenJDK 8; if you leave off the
+  `openjdk=8` it will install OpenJDK 11 by default, which should also work, but
+  is less well tested and may have more rough edges.
 
 * It is possible to dynamically install pyimagej from within a Jupyter notebook:
-
     ```
     import sys
     !conda install --yes --prefix {sys.prefix} -c conda-forge pyimagej openjdk=8
     ```
-
-* If you would prefer to install pyimagej via pip, more legwork is required. See [this thread](https://forum.image.sc/t/how-do-i-install-pyimagej/23189/4) for hints.
+  This approach is useful for [JupyterHub](https://jupyter.org/hub) on the
+  cloud, e.g. [Binder](https://mybinder.org/), to utilize pyimagej in select
+  notebooks without advance installation. This reduces time needed to create
+  and launch the environment, at the expense of a longer startup time the first
+  time a pyimagej-enabled notebook is run. See [this itkwidgets example
+  notebook](https://github.com/InsightSoftwareConsortium/itkwidgets/blob/v0.24.2/examples/ImageJImgLib2.ipynb)
+  for an example.
+
+* It is possible to dynamically install pyimagej on
+  [Google Colab](https://colab.research.google.com/). See
+  [this thread](https://forum.image.sc/t/pyimagej-on-google-colab/32804) for
+  guidance. A major advantage of Google Colab is free GPU in the cloud.
+
+* If you would prefer to install pyimagej via pip, more legwork is required.
+  See [this thread](https://forum.image.sc/t/how-do-i-install-pyimagej/23189/4)
+  for hints.
 
 ## Usage
 

environment.yml

@@ -2,11 +2,11 @@ name: imagej
 channels:
   - conda-forge
 dependencies:
-  - imglyb
+  - imglyb=0.3.5
   - matplotlib
   - numpy
   - openjdk=8
-  - pyjnius
+  - pyjnius=1.2.0
   - scyjava
   - pillow # for server
   - requests # for server