add script to check whether JSON data for available software has been updated (ignoring timestamp)

.github/workflows/update_available_software.yml

@@ -21,25 +21,29 @@ jobs:
       - name: update overview of available software
         id: update_available_software
         run: |
-          cd scripts/available_software
-
           # install required Python packages in virtual environment
           python -m venv venv
           . venv/bin/activate
           pip install -r requirements.txt
 
-          python available_software.py
+          # copy current JSON data, so we can compare after updating it
+          cp docs/available_software/data/json_data.json docs/available_software/data/json_data.json.orig
+          cp docs/available_software/data/json_data_detail.json docs/available_software/data/json_data_detail.json.orig
+
+          python scripts/available_software/available_software.py
           git status
 
           # determine whether pull request should be opened:
           # if JSON files in docs/available_software/data have been updated, then a PR should be opened
-          json_data_changed=$((git diff --exit-code ../../docs/available_software/data > /dev/null && echo 'no') || echo 'yes')
-          if [[ ${json_data_changed} == "yes" ]]; then
-            echo "JSON files in docs/available_software/data have been changed, PR should be opened"
-          else
+          json_data_changed=$(python scripts/available_software/check_json_updates.py docs/available_software/data/json_data.json.orig docs/available_software/data/json_data.json)
+          json_data_detail_changed=$(python scripts/available_software/check_json_updates.py docs/available_software/data/json_data_detail.json.orig docs/available_software/data/json_data_detail.json)
+          if [[ ${json_data_changed} == "no changes" ]] && [[ ${json_data_detail_changed} == "no changes" ]]; then
             echo "JSON files in docs/available_software/data have not been changed, no need to open PR"
+            echo "json_data_changed=no" >> $GITHUB_OUTPUT
+          else
+            echo "JSON files in docs/available_software/data have been changed, PR should be opened"
+            echo "json_data_changed=yes" >> $GITHUB_OUTPUT
           fi
-          echo "json_data_changed=${json_data_changed}" >> $GITHUB_OUTPUT
 
       - name: create pull request
         uses: peter-evans/create-pull-request@6d6857d36972b65feb161a90e484f2984215f83e # v6.0.5

scripts/available_software/available_software.py

@@ -1,3 +1,4 @@
+#!/usr/bin/env python
 #
 # Copyright 2023-2023 Ghent University
 #

scripts/available_software/check_json_updates.py

@@ -0,0 +1,34 @@
+#!/usr/bin/env python
+#
+# Copyright 2023-2023 Ghent University
+#
+# SPDX license identifier: GPL-3.0-or-later
+#
+"""
+Python script to check for differences between provided JSON files
+that contain data on available software.
+
+@author: Kenneth Hoste (Ghent University)
+"""
+import json
+import sys
+
+if len(sys.argv) != 3:
+    sys.stderr.write(f"Usage: {sys.argv[0]} <one.json> <two.json>\n")
+    sys.exit(1)
+
+json1_path = sys.argv[1]
+json2_path = sys.argv[2]
+
+with open(json1_path) as json1:
+    json1_data = json.load(json1)
+with open(json2_path) as json2:
+    json2_data = json.load(json2)
+
+json1_data_filtered = {k: v for (k, v) in json1_data.items() if k != 'time_generated'}
+json2_data_filtered = {k: v for (k, v) in json2_data.items() if k != 'time_generated'}
+
+if json1_data_filtered == json2_data_filtered:
+    print("no changes")
+else:
+    print("differences found")