OS2mo SDTool+

OS2mo integration for SD.

Usage

podman compose up -d

Configuration is done through environment variables. Available options can be seen in sdtoolplus/config.py. Complex variables such as dict or lists can be given as JSON strings, as specified by Pydantic's settings parser.

Testing

After starting the project, tests can be run using pytest, for example:

podman compose stop sdtool-plus
podman compose run --rm sdtool-plus pytest tests/integration/test_engagement_timeline.py

All tests should be runnable both locally and in CI -- it is considered a bug if not.

APOS

When running in REGION mode, engagements are placed in the units related to the SD payroll units according to the strategy implemented in the function engagement_ou_strategy_region in engagements.py. In order for this to work, the extension_5 attributes of the MO engagements (originating from APOS) need to be populated with the UUID of the organization unit where the engagement is placed in SD, before the engagements are imported to MO for production. This is done by generating a JSON file with these data to be consumed by the APOS importer. Generating the JSON file is a two-step process as described below.

Step 1

This step is very time-consuming and needs to be performed in multiple iterations due to the opening ours of the SD API.

1. Extract all engagements directly from the MO database (to save time). This can be done by using the SQL query found in all-engagements.sql. The query will generate a CSV file to be used by the Python script in 2).
2. Run the Python script sd_engagement_json.py from the SD pod (see more optional parameters in the file):

   $ python -m scripts.sd_engagement_json --username <SD_USERNAME> --password <SD_PASSWORD>
   

   This will generate a JSON file in /tmp/engagements.json containing the extension_5 data for all the engagements in MO.

Step 2

Since step 1 is very time-consuming, it is preferable to only do this once (e.g. a few weeks before the system goes live). Immediately before the final APOS import is run, the JSON file from step 1 must be updated with the latest changes from SD. This is done using the script update_engagement_placement_cache.py (see more optional parameters in the file):

$ python -m scripts.update_engagement_placement_cache --username <SD_USERNAME> \
  --password <SD_PASSWORD> \
  --file /tmp/engagements.json \
  --since <DATETIME-WHEN-ENGAGEMENT-JSON-WAS-GENERATED>

This will result in the file /tmp/engagements-patched.json containing the latest changed from SD.