Add physics method dependency description in docs#865
Conversation
sunt05
temporarily deployed
to
github-pages-preview
GitHub Actions
Inactive
sunt05
changed the title
🔍 Schema Preview DeployedPreview URLs:
Production URLs (unchanged):
|
|
🤖 I've automatically formatted the code in this PR using:
Please pull the latest changes before making further edits. |
sunt05
temporarily deployed
to
github-pages-preview
GitHub Actions
Inactive
…ndling Add comprehensive validation for ModelPhysics configuration to enforce critical compatibility rules between physics methods: - EHC storage heat requires SPARTACUS net radiation (>= 1000) - SPARTACUS net radiation requires EHC storage heat - STEBBS storage heat requires active stebbsmethod - RC method requires active STEBBS - OhmIncQf inclusion requires OHM-based storage heat methods Improvements: - Add type hints to _unwrap_method_value() helper function - Add explicit None checks to prevent TypeErrors - Format multiple errors as numbered list for clarity - Add comprehensive test coverage for all validation rules - Enhance documentation with dependency metadata (depends_on, provides_to) - Add clarifying comments for physics_anchor_fields in RST generator
Add blank lines after depends_on and provides_to in RST generation to ensure they appear on separate rows in the rendered documentation instead of on the same line. This improves readability of method interaction metadata.
Co-authored-by: sunt05 <1802656+sunt05@users.noreply.github.com>
Replace complex sys.path manipulation and mock module creation with standard import pattern used by other tests in test/data_model/. Pytest handles import paths correctly without this boilerplate.
sunt05
temporarily deployed
to
github-pages-preview
GitHub Actions
Inactive
| json_schema_extra={"unit": "dimensionless"}, | ||
| json_schema_extra={ | ||
| "unit": "dimensionless", | ||
| "depends_on": ["ohmincqf", "snowuse"], |
There was a problem hiding this comment.
Hi @dayantur - this shows the idea I mentioned about using data structure to illustrate the module dependency.
There was a problem hiding this comment.
Hi @sunt05, I had a quick look and it seems like a good improvement.. definitely a faster way to implement this kind of rule. I think we just need to make sure we clean up the existing dependency rules in phase_b.py (there are already a couple or so in there), and that we place this new model validator in SUEWSConfig (since we've moved all validators into config.py - unless you think this specific case requires otherwise?).
I’m about to push a couple of updated CSV files in PR #851, where I’ve collected all the rules currently in the pipeline. That should give us a clearer picture of what would need to be deduplicated if we move forward with this approach.
P.S.: I think we can agree - though correct me if you think I'm wrong! - that we'll still need some rules in phase_b.py for physics-option dependencies when a particular combination of model physics settings imposes checks or constraints on other parameters, rather than just encoding the basic physics-option dependency itself.
2 participants
Add comprehensive validation for physics method configurations to enforce critical compatibility rules between methods. Includes:
Validation includes explicit None checks, improved error formatting with numbered lists for multiple errors, and full test coverage. Documentation enhanced with depends_on/provides_to metadata and proper cross-reference links.