From 325326ba3b5bb29be5c1511fc04c81160ab90b53 Mon Sep 17 00:00:00 2001 From: Andrei Radu Date: Sat, 29 Aug 2026 13:46:37 +0200 Subject: [PATCH] test: retarget DecodeBench assertions at the lab structured protocol The 'DecodeBench uses Showcase structural prompts' test still asserts pickShowcasePrompts("structural") and withFillToMaxInstruction, both of which 1.8.3 removed when DecodeBench moved onto the lab structured protocol (count 1 -> 200), and 1.8.4 replaced with the output-type picker. The test therefore fails on a clean main. Retargeted at the current API (pickDecodeBenchPrompts / decodeBenchPromptForType / normalizeDecodeBenchType) and added top_p: 1 plus a negative assertion on withFillToMaxInstruction, so the protocol is pinned in the direction it actually moved. The remaining assertions are unchanged and still pass. --- server/collectors/__tests__/showcasePrompts.test.js | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/server/collectors/__tests__/showcasePrompts.test.js b/server/collectors/__tests__/showcasePrompts.test.js index 57ce6ea2..5910d12e 100644 --- a/server/collectors/__tests__/showcasePrompts.test.js +++ b/server/collectors/__tests__/showcasePrompts.test.js @@ -91,12 +91,18 @@ test("catalog prompts exist for text, structural, and mixed pickers", () => { assert.ok(textCount >= 2); }); -test("DecodeBench uses Showcase structural prompts at temperature 0, thinking off", () => { - assert.match(benchSrc, /pickShowcasePrompts\("structural"/); - assert.match(benchSrc, /withFillToMaxInstruction/); +test("DecodeBench uses the lab structured protocol at temperature 0, thinking off", () => { + // 1.8.3 moved DecodeBench off the Showcase structural catalog + fill-to-max + // and onto the lab structured protocol (count 1 -> 200); 1.8.4 added the + // output-type picker. These assertions still referenced the removed API. + assert.match(benchSrc, /pickDecodeBenchPrompts\(/); + assert.match(benchSrc, /decodeBenchPromptForType\(/); + assert.match(benchSrc, /normalizeDecodeBenchType\(/); assert.match(benchSrc, /temperature:\s*0/); + assert.match(benchSrc, /top_p:\s*1/); assert.match(benchSrc, /applyThinkingFlags\(body,\s*modelId,\s*false\)/); assert.match(benchSrc, /min_tokens:\s*maxTokens/); + assert.doesNotMatch(benchSrc, /withFillToMaxInstruction/); assert.doesNotMatch(benchSrc, /uniquePrefillPrefix/); assert.doesNotMatch(benchSrc, /BENCH_PROMPTS/); });