Skip to content

fix(review): canonicalize calibration login lookup #5206

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
JSONbored merged 1 commit into from
Jul 12, 2026
Merged

fix(review): canonicalize calibration login lookup #5206

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 3 additions & 2 deletions src/review/predicted-gate-calibration-ledger.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -141,13 +141,14 @@ export async function recordPredictedGateCalibration(
export async function computeContributorCalibration(env: PredictedGateCalibrationEnv, login: string | null | undefined): Promise<ContributorCalibrationSignal | null> {
const trimmed = login?.trim();
if (!trimmed) return null;
const normalizedLogin = trimmed.toLowerCase();
try {
const row = await env.DB.prepare(
`SELECT COUNT(*) AS sampleSize, COALESCE(AVG(agreed), 0) AS agreementRate
FROM predicted_gate_calibration_ledger
WHERE login = ?`,
WHERE lower(login) = ?`,
)
.bind(trimmed)
.bind(normalizedLogin)
.first<{ sampleSize: number; agreementRate: number }>();
// COUNT(*)/AVG(...) with no GROUP BY always returns exactly one row, even over zero matches (COUNT: 0,
// AVG: NULL -> COALESCE: 0) -- .first()'s nullable return type is a TypeScript-level formality here, not
Expand Down
14 changes: 14 additions & 0 deletions test/unit/predicted-gate-calibration-ledger.test.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -220,6 +220,20 @@ describe("computeContributorCalibration — per-login calibration read (#2349)",
expect(await computeContributorCalibration(env, "someone-else")).toEqual({ sampleSize: 1, agreementRate: 1 });
});

it("canonicalizes GitHub login casing so case variants cannot bypass or infer calibration", async () => {
const env = createTestEnv();
await seedLedgerRow(env, { login: "OctoCat", pullNumber: 1, agreed: false });
await seedLedgerRow(env, { login: "octocat", pullNumber: 2, agreed: false });
await seedLedgerRow(env, { login: "OCTOCAT", pullNumber: 3, agreed: true });
await seedLedgerRow(env, { login: "someone-else", pullNumber: 1, agreed: true });

const expected = { sampleSize: 3, agreementRate: 1 / 3 };
expect(await computeContributorCalibration(env, "OctoCat")).toEqual(expected);
expect(await computeContributorCalibration(env, " octocat ")).toEqual(expected);
expect(await computeContributorCalibration(env, "OCTOCAT")).toEqual(expected);
expect(await computeContributorCalibration(env, "someone-else")).toEqual({ sampleSize: 1, agreementRate: 1 });
});

it("aggregates across ALL of a login's history regardless of which repo each pairing came from", async () => {
const env = createTestEnv();
await seedLedgerRow(env, { login: "octocat", project: "owner/repo-a", pullNumber: 1, agreed: true });
Expand Down