Skip to content

#1 Persistent memory for AI coding agents based on real-world benchmarks

test(status): keep the vector rows test to vector state only

Rohit Ghumaread9e88992911
diff --git a/test/status.test.ts b/test/status.test.ts
index 4950f73e..4751c488 100644
--- a/test/status.test.ts
+++ b/test/status.test.ts
@@ -510,7 +510,7 @@ describe("markLlmFunctions", () => {
});
describe("vector durability rows", () => {
- it("shows the pending vector log, a paused backlog and skipped repeats in plain language", () => {
+ it("shows the pending vector log and a paused backlog in plain language", () => {
const report = evaluateStatus(
inputs({
index: { ...inputs().index, pendingVectorBackfill: 7, vectorBackfillState: "paused" },
@@ -524,16 +524,13 @@ describe("vector durability rows", () => {
pendingLog: 3,
pendingLogError: "state write timed out",
},
- observeDedup: { skippedSinceStart: 4, windowSeconds: 300 },
}),
);
expect(codes(report)).toContain("index-pending-log-failing");
const backlog = report.problems.find((p) => p.code === "index-vector-backfill-pending");
expect(backlog?.fix).toContain("retried on the next start");
- expect(report.observeDedup).toEqual({ skippedSinceStart: 4, windowSeconds: 300 });
const html = renderStatusHtml(report, "n");
expect(html).toContain("3 vector changes");
expect(html).toContain("paused, retried on the next start");
- expect(html).toContain("4 since start");
});
});