Prepare 0.3 release (#90)

* Only run compliance on PR, merge and release

* Allow PySpark 3.2 in setup.cfg for trac-runtime (py), add to readme and tests

* Try to build with Python 3.10 in CI

* Update supported versions in the runtime README

* Update main README for 0.3 release

.github/workflows/build.yml

@@ -74,9 +74,9 @@ jobs:
       matrix:
         enviroment:
 
-          - { ENV_NAME: "Latest: Python 3.9, PySpark 3.1.x, Pandas >= 1.3",
-              PYTHON_VERSION: "3.9",
-              PYSPARK_VERSION: ">= 3.1.0, < 3.2.0",
+          - { ENV_NAME: "Latest: Python 3.10, PySpark 3.2.x, Pandas >= 1.3",
+              PYTHON_VERSION: "3.10",
+              PYSPARK_VERSION: ">= 3.2.0, < 3.3.0",
               PANDAS_VERSION: ">= 1.3.0, < 2.0.0",
               PYPANDOC: false }
 

.github/workflows/compliance.yml

@@ -8,7 +8,7 @@ on:
   # Re-run compliance jobs in main, to make sure there are no issues from the merge
   push:
     branches:
-      - '**'
+      - main
 
   # Use the release:publish event to generate compliance reports for releases
   # This lines up with the packaging workflow, so reports will be published when packages are published

README.md

@@ -39,10 +39,11 @@ https://github.com/Accenture/trac/actions/workflows/integration.yml)
 [![Packaging](https://github.com/Accenture/trac/actions/workflows/packaging.yml/badge.svg)](
 https://github.com/Accenture/trac/actions/workflows/packaging.yml)
 
-The current release series (0.2.x) is intended for reference and experimentation.
-It includes the metadata service, runtime engine for Python and a partial implementation
-of the platform Gateway. This release can be used to prototype client applications (e.g.
-web UIs) and to build and run models in a development sandbox.
+The current release series (0.3.x) is intended for reference and experimentation.
+It includes the metadata service, data service and core data engine, runtime engine
+for Python and a partial implementation of the platform Gateway. This release can
+be used to prototype client applications (e.g. web UIs) and to build and run models
+in a development sandbox.
 
 At the moment TRAC's API calls and metadata structures are subject to change between 
 versions, including between point versions. As the platform evolves these APIs will

trac-runtime/python/README.md

@@ -18,7 +18,7 @@ The TRAC runtime for Python has these requirements:
 
 * Python: 3.7 or later
 * Pandas: 1.0 or later
-* PySpark 2.4.x, 3.0.x or 3.1.x
+* PySpark 2.4.x or 3.x
 
 Not every combination of versions will work, e.g. PySpark 3 requires Python 3.8.
 

trac-runtime/python/setup.cfg

@@ -53,5 +53,5 @@ python_requires = >=3.7
 
 install_requires =
     pandas >= 1.0.0, < 2.0.0
-    pyspark >= 2.4.0, < 3.2.0
+    pyspark >= 2.4.0, < 3.3.0
     pyyaml >= 5.3.0, < 6.0.0