Package

.gitignore

@@ -1,14 +1,35 @@
-# Logs and databases #
-######################
-slurm-*.out
-output*.out
-ERA5_createForcing_CONUS*.sh
-core.*
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*/__pycache__/
+*.py[cod]
+*$py.class
 
-# notebook stuff
-.ipynb_checkpoints
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+*.pyc
 
-# rtd
-rtd/build
+# Sphinx documentation
+docs/_build/
 
-rtd/source/*.md
+# test suite
+tests/*/results
+
+# code editors
+.vscode

README.rst

@@ -0,0 +1,49 @@
+CWARHM
+=======
+
+**CWARHM** is a Python library to organize workflows to build hydrological models.
+The Package branch is to develop the original SUMMA-CWARHM (currently on main, in this branch moved under dependencies/summa-cwarhm) into a python package where the code exectuting workflow steps are callable functions, and the actual workflows (here a workflow are all the steps involved in setting up an hydrological model) are scripts calling the available functions in a certain order.
+
+Installation
+--------------
+The dependencies of the CWARHM package (or CWARHM-Assembler - CWARHMA) can be installed using conda:
+
+  ``cd /path/to/summaWorkflow_public``
+
+  ``conda env create -f environment.yml``
+
+  ``conda activate cwarhm-env``
+
+
+Additionally the package can be installed in the environment in developer mode (not needed for the test):
+
+  ``cd /path/to/summaWorkflow_public``
+
+  ``conda activate cwarhm-env``
+
+  ``pip install -e .``
+
+Test Bow at Banff SUMMA
+----------
+As test case workflows/summa_bowatbanff/ is available. Before running the test script test_bow_at_banff.py, some path modifications have to be made:
+
+  1.) In workflows/summa_bowatbanff/control_Bow_at_Banff_test.txt, change the **root_path** modeling domain setting to a local folder where the test results should be saved.
+
+  2.) In workflows/summa_bowatbanff/test_bow_at_banff.py, change the paths: 
+    - cwarhm_summa_folder; to point to the `./summaWorkflow_public/dependencies/cwarhm-summa` folder
+    - results_folder_path , to match the root_path in the control file
+    - test_data_path , path to the test data, if you want to skip the data download workflow steps (by default)
+    - reset_test to True (default), this is a flag that starts a new run (and deletes all data from any results folder), and restarts by copying the test data to the results folder or False: continue with existing data in the results folder.
+
+  3.) The test data is not part of this repo due to its size (30+GB). For those in the comphyd group it can be found here:
+
+**copernicus** /project/gwf/gwf_cmt/cwarhm_test_data
+**GRAHAM** /project/6008034/CompHydCore/cwarhm_test_data
+
+If this is not accessible, you can download the data with the CWARHM functions (wrapped from the original CWARHM). You can also use a results directory from an earlier test run.
+
+Test Bow at Banff MESH
+---------
+The Bow at Banff MESH test, performs a part of a complete workflow (with data specific parts and most of the model agnostic parts processed by the above workflow). to have a look at the relevant functions have a look at tests/test_mesh_bowatbanff.py
+Note that for it to run some input data is needed that is now included in the test folder and will be extracted automatically (21MB).
+Only one path has to adjusted, but in two(!) places (see test/test_mesh_bowatbanff.py).