Home/pyrite/JI: Enum validation and code quality fixes

JI: Enum validation and code quality fixesbacklog_item

qualityjirefactor
1 min read · Edit on Pyrite

Problem

Code review identified several quality issues in the journalism-investigation plugin:

1. Missing enum validation: The validator checks claim_status and confidence against constants but does not validate asset_type, account_type, evidence_type, transaction_type, case_type, case_status, mechanism, or reliability against their defined enum tuples. The constants are defined but unused.

2. Duplicated code: JSON metadata parsing (5-line block repeated 7 times in MCP handlers), EventStatus parsing (repeated 3 times in from_frontmatter methods), import json repeated per-handler instead of module-level.

3. str(None) safety: Multiple from_frontmatter methods use str(meta.get("field", "")) which produces "None" instead of "" when YAML has explicit null values.

4. summary field inconsistency: Entity types conditionally add summary in to_frontmatter(), event types do not. Should be consistent.

Scope

Enum Validation (TDD)

  • Add validation for all enum fields in _validate_investigation_entry
  • Validation should warn on invalid values, not block (enums are advisory, new values may be valid)
  • Add tests for invalid enum values across all entry types

    • Code Dedup

  • Extract _parse_meta(entry_dict) helper for JSON metadata parsing
  • Extract _parse_event_status(meta) helper for EventStatus parsing
  • Move import json to module level in plugin.py

    • Safety Fixes

  • Add _str_or_empty(value) helper that returns "" for None
  • Apply consistently across all from_frontmatter methods

    • Acceptance Criteria

  • All enum fields validated with warnings for invalid values
  • No duplicated 5-line metadata parsing blocks
  • str(None) returns "" not "None" in all from_frontmatter methods
  • All existing tests still pass
  • New tests for enum validation edge cases