Move requirements-dev.txt dependencies to environment.yml (#812)

* Move dev dependency specification to environment.yml
* Use `conda env update` instead of `conda install`
* Let setup miniconda auto activate pygmt conda environment
* Add section headers to environment.yml file

Co-authored-by: Dongdong Tian <[email protected]>
Co-authored-by: Will Schlitzer <[email protected]>
Co-authored-by: Wei Ji <[email protected]>

Author: 3 people

.github/workflows/ci_tests.yaml

@@ -69,22 +69,15 @@ jobs:
       - name: Setup Miniconda
         uses: conda-incubator/[email protected]
         with:
+          activate-environment: pygmt
           python-version: ${{ matrix.python-version }}
           channels: conda-forge
           miniconda-version: "latest"
 
       # Install GMT and other required dependencies from conda-forge
-      - name: Install GMT and required dependencies
+      - name: Install dependencies
         shell: bash -l {0}
-        run: |
-          requirements_file=full-conda-requirements.txt
-          cat requirements.txt requirements-dev.txt > $requirements_file
-          cat << EOF >> $requirements_file
-          gmt=6.1.1
-          make
-          codecov
-          EOF
-          conda install --yes --file $requirements_file
+        run: conda env update --file environment.yml
 
       # Show installed pkg information for postmortem diagnostic
       - name: List installed packages

MAINTENANCE.md

@@ -42,7 +42,7 @@ The main advantages of this are:
 
 We use GitHub Actions continuous integration (CI) services to
 build and test the project on Linux, macOS and Windows.
-They rely on the `requirements.txt` file to install required dependencies using
+They rely on the `environment.yml` file to install required dependencies using
 conda and the `Makefile` to run the tests and checks.
 
 ### GitHub Actions

environment.yml

@@ -3,22 +3,25 @@ channels:
     - conda-forge
     - defaults
 dependencies:
-    - python=3.8
+    # Required dependencies
     - pip
     - gmt=6.1.1
     - numpy
     - pandas
     - xarray
     - netCDF4
     - packaging
+    # Development dependencies
     - black
     - blackdoc
+    - codecov
     - coverage[toml]
     - docformatter
     - flake8
     - ipython
     - isort>=5
     - jupyter
+    - make
     - matplotlib
     - nbsphinx
     - pylint

requirements-dev.txt

@@ -1,4 +1,4 @@
-# Requirements for installing with conda
+# Required packages
 numpy
 pandas
 xarray