Greptile Summary

This PR fixes the bug in validate_host_metrics where data_quality_issues were never surfaced at the root level for list-based payloads (previously collected by iterating validated_data after _validate_flat_metrics, which never writes that key to individual points).

• P1 — structural regression in list items: Switching to validator.validate_metrics(data_point.copy()) for each flat data point triggers _validate_cpu_metric and _validate_disk_metric on scalar values, wrapping cpu and disk with {"raw": ..., "validated": False} dicts instead of preserving the original integers. A simpler fix would be to keep _validate_flat_metrics per item and collect issues from validator.issues after each call, avoiding the double-dispatch into the nested-structure branches.
• P1 — test gap: test_wrapper_list_structure_payload only asserts "data_quality_issues" in result and does not assert the preserved scalar types of cpu/disk inside result["data"], so the regression passes undetected.

Confidence Score: 3/5

Not safe to merge as-is — the fix introduces a structural regression that changes the shape of every validated list item.

Two P1 findings: the switch to validate_metrics on individual flat items silently replaces scalar cpu/disk values with {"raw": ..., "validated": False} dicts, and the new test only checks for the presence of data_quality_issues rather than verifying the unchanged shape of the data items, leaving the regression undetected.

app/tools/utils/data_validation.py (line 398) and tests/tools/utils/test_data_validation.py (lines 129–138) both need attention before merging.

Important Files Changed

Flowchart

%%{init: {'theme': 'neutral'}}%%
flowchart TD
    A["validate_host_metrics(metrics)"] --> B{Is metrics a dict?}
    B -- No --> C["Return error with data_quality_issues"]
    B -- Yes --> D{"'data' key present and is a list?"}
    D -- No --> E["validator.validate_metrics(metrics) — flat/nested path"]
    D -- Yes --> F["Loop over data points"]
    F --> G{Is data_point a dict?}
    G -- No --> H["Append as-is"]
    G -- Yes --> I["validator.validate_metrics(data_point) ⚠️ wraps scalar cpu/disk"]
    I --> J{"data_quality_issues in validated_point?"}
    J -- Yes --> K["Extend all_issues; del from validated_point"]
    J -- No --> L["Skip"]
    K --> M["Append validated_point"]
    L --> M
    H --> M
    M --> F
    F --> N["result = metrics.copy(); result['data'] = validated_data"]
    N --> O{any all_issues?}
    O -- Yes --> P["result['data_quality_issues'] = all_issues"]
    O -- No --> Q["Return result"]
    P --> Q

_{Reviews (1): Last reviewed commit: "Fix: attach data_quality_issues for list..." | Re-trigger Greptile}

@emefienem review points, CI checks

😤 @emefienem said "I will fix this" and then actually fixed it. Legendary behavior.

👋 Join us on Discord - OpenSRE : hang out, contribute, or hunt for features and issues. Everyone's welcome.

awesome @emefienem merging this for velocity!

🫡