[code-simplifier] simplify: dedup validateAllowedIssueFields and extract parseUnknownModelAICreditsFromAuditEntry#40725
Conversation
…delAICreditsFromAuditEntry - allowed_issue_fields.cjs: validateAllowedIssueFields now delegates to validateAllowedIssueFieldName per field, removing duplicated Set construction and wildcard/empty-list guard logic. - ai_credits_context.cjs: extract inline DFS object traversal from parseUnknownModelAICreditsFromAuditLog's accumulate callback into a named parseUnknownModelAICreditsFromAuditEntry function, consistent with the parseMaxAICreditsFromAuditEntry / parseAICreditsErrorInfoFromAuditEntry pattern. Behavior is unchanged in both files. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
|
✅ Test Quality Sentinel completed test quality analysis. No test files were added or modified in this PR. The changes only affect production files: actions/setup/js/ai_credits_context.cjs and actions/setup/js/allowed_issue_fields.cjs. Test Quality Sentinel skipped. |
|
✅ Design Decision Gate 🏗️ completed the design decision gate check. No ADR enforcement needed: PR #40725 does not have the 'implementation' label and has 0 new lines of code in business logic directories (well under the 100-line threshold). |
|
✅ PR Code Quality Reviewer completed the code quality review. |
|
🧠 Matt Pocock Skills Reviewer has completed the skills-based review. ✅ |
There was a problem hiding this comment.
Pull request overview
This PR applies two small refactors in the actions/setup/js runtime helpers to improve readability and consistency while preserving existing logic for allowed issue field validation and firewall audit-log parsing.
Changes:
- Simplifies
validateAllowedIssueFieldsby delegating per-field checks tovalidateAllowedIssueFieldName. - Extracts the
unknown_model_ai_creditsaudit-entry traversal intoparseUnknownModelAICreditsFromAuditEntryto match the pattern used by other audit-entry parsers.
Show a summary per file
| File | Description |
|---|---|
| actions/setup/js/allowed_issue_fields.cjs | Refactors allowed issue field list validation to reuse the single-field validator. |
| actions/setup/js/ai_credits_context.cjs | Extracts per-entry unknown-model detection into a dedicated helper for consistency with other parsers. |
Copilot's findings
Tip
Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
- Files reviewed: 2/2 changed files
- Comments generated: 1
![github-actions[bot]](https://avatars.githubusercontent.com/in/15368?s=60&v=4){kind=small}
There was a problem hiding this comment.
Skills-Based Review 🧠
Applied /improve-codebase-architecture and /zoom-out — approving with two minor observations.
📋 Key Themes & Highlights
Key Themes
- Delegation trade-off (
allowed_issue_fields.cjs):validateAllowedIssueFieldNamenow buildsallowedFieldSetonce per field instead of once pervalidateAllowedIssueFieldscall. Negligible for small arrays; noted for awareness. - Pattern surfacing (
ai_credits_context.cjs): the newacc || fn(entry)form is consistent with line 253 and better-behaved than thefn(entry) || accform still used at line 189 — a latent inconsistency this PR makes visible.
Positive Highlights
- ✅ Clean extraction of
parseUnknownModelAICreditsFromAuditEntry— creates parity with the two existing per-entry helpers (parseMaxAICreditsFromAuditEntry,parseMaxAICreditsExceededFromAuditEntry) - ✅ Net −6 lines with no behavior change; existing test suites cover all guard conditions
- ✅
acc || parseUnknownModelAICreditsFromAuditEntry(entry)has correct short-circuit semantics
🧠 Reviewed using Matt Pocock's skills by Matt Pocock Skills Reviewer · 61 AIC · ⌖ 7.63 AIC · ⊞ 6.5K
| if (!allowedFieldSet.has(field.name.toLowerCase())) { | ||
| throw new Error(`${ERR_VALIDATION}: issue field "${field.name}" is not in the allowed-fields list: ${allowedFields.join(", ")}`); | ||
| } | ||
| validateAllowedIssueFieldName(field.name, allowedFields); |
There was a problem hiding this comment.
[/improve-codebase-architecture] The refactor moves allowedFieldSet construction (and the empty/wildcard guards) from a single pre-loop site into each validateAllowedIssueFieldName call — O(N) Set builds instead of O(1). For current usage (small issueFields arrays) this is negligible, but good to be aware of if this path ever becomes hot.
There was a problem hiding this comment.
Addressed in ee75179. The bulk validator now restores single-pass set construction, so this path no longer rebuilds the set for each field.
| } | ||
| return false; | ||
| } | ||
| (acc, entry) => acc || parseUnknownModelAICreditsFromAuditEntry(entry) |
There was a problem hiding this comment.
[/zoom-out] The acc || parseUnknownModelAICreditsFromAuditEntry(entry) form matches the pattern at line 253 (parseMaxAICreditsExceededFromAuditLog's reducer) and is the right call: it short-circuits the entry-parse when acc is already true. By contrast, line 189 (parseMaxAICreditsFromAuditLog) still uses the entry-first form fn(entry) || acc, which evaluates fn even after the accumulator has been set. This PR inadvertently exposes that inconsistency — a follow-up to align line 189 would complete the pattern unification.
There was a problem hiding this comment.
No change here. I kept the ai_credits_context.cjs extraction as-is and left the line 189 reducer ordering for a separate follow-up so this PR stays scoped to the review feedback on allowed_issue_fields.cjs.
There was a problem hiding this comment.
COMMENT — one medium behavioral-discrepancy finding; ai_credits_context.cjs extraction is clean.
Review summary
ai_credits_context.cjs — clean refactor ✓
Extracting parseUnknownModelAICreditsFromAuditEntry is correct:
- Short-circuit semantics preserved:
acc || fn(entry)is equivalent to the originalif (acc) return true; /* inline DFS */. - Entry guard (
!entry || typeof entry !== "object") and DFS traversal are identical. - Function is not exported, consistent with module design.
allowed_issue_fields.cjs — one finding
See inline comment: the delegation to validateAllowedIssueFieldName silently passes falsy field.name values (including ""), whereas the original code would have thrown a validation error for those inputs. The PR's claim of "behavior is unchanged" is technically incorrect for this edge case.
Performance regression (O(n·m) Set reconstruction per field) is already covered by the existing review comment.
🔎 Code quality review by PR Code Quality Reviewer · 123.2 AIC · ⌖ 7.53 AIC · ⊞ 5.1K
| if (allowedFieldSet.has("*")) { | ||
| return; | ||
| } | ||
| if (!Array.isArray(issueFields) || issueFields.length === 0) return; |
There was a problem hiding this comment.
Behavioral change: falsy field.name now silently passes validation — the refactor unintentionally weakens the field-name guard, contradicting the PR claim that "behavior is unchanged."
💡 Details
The original validateAllowedIssueFields checked each field name directly against the allowedFieldSet:
// old
if (!allowedFieldSet.has(field.name.toLowerCase())) {
throw new Error(...);
}For field.name === "" this evaluates "".toLowerCase() === "", which is not in the set → throws a validation error.
The new delegate path starts with:
// inside validateAllowedIssueFieldName
if (!fieldName) return; // empty string is falsy → silently passesSo a field with name: "" now passes validation with no error instead of being rejected. While the JSDoc type says name: string, the runtime has no such enforcement guarantee. Any caller passing { name: "", value: ... } will see silently accepted invalid input instead of the previous validation error.
Suggested fix — add an explicit guard before delegating, or document that empty-named fields are now intentionally permitted:
function validateAllowedIssueFields(issueFields, allowedFields) {
if (!Array.isArray(issueFields) || issueFields.length === 0) return;
for (const field of issueFields) {
validateAllowedIssueFieldName(field.name, allowedFields);
}
}If the callers are known to always produce non-empty field.name, update the JSDoc @param to @param {Array<{name: NonEmptyString, ...}>} or add an assertion.
There was a problem hiding this comment.
Fixed in ee75179. validateAllowedIssueFields now rejects an empty field.name again, and I added a regression test covering that case.
|
`` Please fix the falsy field.name regression and re-request review.
|
|
``
|
|
@copilot run pr-finisher skill |
|
Hey One thing to address before merge:
If you'd like a hand:
|
Co-authored-by: pelikhan <4175913+pelikhan@users.noreply.github.com>
pelikhan
deleted the
simplify/dedup-validation-and-audit-traversal-808e6d21e286f88c
branch
Summary
Refactoring and cleanup pass over
allowed_issue_fieldsandai_credits_contextin the setup action JS layer. No logic changes are introduced; the commits address review feedback and reduce duplication.Changes
actions/setup/js/ai_credits_context.cjsparseUnknownModelAICreditsFromAuditLogreducer into a new named helper:parseUnknownModelAICreditsFromAuditEntry.actions/setup/js/allowed_issue_fields.cjs.map()callback parameter fromf→fieldfor clarity.actions/setup/js/allowed_issue_fields.test.cjsvalidateAllowedIssueFieldsthrowsERR_VALIDATIONwhen a field name is an empty string.Type of change
Breaking changes
None.
Testing
validateAllowedIssueFields.