/** * Unit tests — the UI parse seam (src/core/tool-parts.ts). * * The rule born of the 2026-09-12 live crashes: model-authored tool input * is consumed only through these parsers, each running the SAME zod schema * the model faced at the boundary. These pins hold the two exact live * crash shapes (the model's double-encoding habit) and the boundary-parity * property: anything a boundary schema accepts, the display parse accepts. */ import assert from "node:assert/strict"; import { describe, it } from "node:test"; import { quizInputProblems } from "../src/core/question-core.ts"; import { Value } from "typebox/value"; import { restoreMathEscapes } from "../src/core/math-delims.ts"; import { parseAskInput, parseProbeDoneInput, parseQuizInput, parseExplainInput, } from "../src/core/tool-parts.ts"; import { quizInputSchema, probeQuizInputSchema } from "../src/core/question-core.ts"; import { probeDoneInputSchema } from "../src/core/contracts/probe.ts"; const TEACHER_QUIZ = { question: "What does Tarski's convention-T define?", options: [ { label: "Truth for a language", value: "truth" }, { label: "Provability", value: "provability" }, ], correctAnswer: "truth", explanation: "Convention-T schematizes truth for the object language.", type: "check", // teach-only role field — the display parse KEEPS it (the dock gates by it) nodeId: "convention-t", // teacher-only field — the display parse must accept and strip it }; const PROBER_QUIZ = { question: "Which planet is closest to the sun?", options: [{ label: "Venus" }, { label: "Mercury" }], correctAnswer: "Mercury", explanation: "Mercury is the innermost planet.", }; const VALID_ASSESSMENT = { assessment: { strands: [ { strand: "modal logic basics", assessment: "Floor: solid intuition; ceiling: notation." }, { strand: "indexical 'actual'", assessment: "Genuine floor — prior exposure likely." }, ], summary: "Bracketed: teach the symbols on the intuition he has.", }, }; describe("parseQuizInput", () => { it("parses a boundary-valid TEACHER quiz and strips the teacher-only nodeId", () => { const parsed = parseQuizInput(TEACHER_QUIZ); assert.ok(parsed); assert.equal(parsed.question, TEACHER_QUIZ.question); assert.equal((parsed as Record).nodeId, undefined, "unknown keys are stripped, not trusted"); assert.equal((parsed as { type?: unknown }).type, "check", "the role field survives the display parse (owner 2026-10-03)"); }); it("parses a boundary-valid PROBER quiz (no nodeId at all)", () => { assert.ok(parseQuizInput(PROBER_QUIZ)); }); it("boundary parity: anything the TEACH schema accepts, the display parse accepts", () => { assert.ok(Value.Check(quizInputSchema, TEACHER_QUIZ), "fixture must be boundary-valid for this pin to mean anything"); assert.ok(parseQuizInput(TEACHER_QUIZ)); }); it("boundary parity: anything the PROBE schema accepts, the display parse accepts", () => { assert.ok(Value.Check(probeQuizInputSchema, PROBER_QUIZ)); assert.ok(parseQuizInput(PROBER_QUIZ)); }); it("boundary parity: the 2026-09-26 string-options shape heals on read (universal-heal mirror, 2026-10-09)", () => { // The boundary now heals a JSON-encoded options string at EVERY // registry (healToolRecord, owner 2026-10-09 — the bare-quiz sets // carried no prepareArguments at all), and the record keeps the RAW // args — so the display parse heals the same shape, per this file's // own parity law: anything the boundary accepts, the display parse // accepts, or the posed quiz renders nowhere while the turn has // ended = learner lockout. const stringified = { ...PROBER_QUIZ, options: JSON.stringify(PROBER_QUIZ.options) }; const parsed = parseQuizInput(stringified); assert.ok(parsed, "the healed-at-boundary record must parse on read"); assert.deepEqual(parsed.options, PROBER_QUIZ.options); }); it("boundary parity under Value.Convert: a NUMERIC correctAnswer parses (live-smoke lockout fix, 2026-09-26)", () => { // The live smoke's model emitted correctAnswer: 17 (number). pi's // boundary runs Value.Convert before Value.Check, so the quiz posed // — but the record keeps the RAW args. A parse seam without Convert // would reject exactly what the boundary accepted: the pending quiz // renders nowhere while the turn has ended = learner lockout. // Dock-visible parity case: the smoke's real emission — options whose // values ARE the numbers, correctAnswer numeric. Convert → "17", // refinement passes (it matches an option), the quiz parses. const numeric = { question: "What is 5 + 3 × 4?", options: [ { label: "17", value: "17" }, { label: "27", value: "27" }, { label: "9", value: "9" }, ], explanation: "Multiplication first.", correctAnswer: 17, }; const parsed = parseQuizInput(numeric); assert.ok(parsed); assert.equal(parsed.correctAnswer, "17", "converted to the string the boundary saw"); const numericArray = { ...numeric, multiSelect: true, correctAnswer: [17, "27"] }; const parsedArray = parseQuizInput(numericArray); assert.ok(parsedArray); assert.deepEqual(parsedArray.correctAnswer, ["17", "27"]); // Non-convertible garbage still never parses (structure owns it)… assert.equal(parseQuizInput({ ...PROBER_QUIZ, correctAnswer: { bad: true } }), undefined); // …and Convert's widening (undefined → "") is caught by the refinement // chain the boundary itself uses — garbage renders as a quiet fallback. const widened = { ...PROBER_QUIZ, correctAnswer: undefined }; assert.equal(parseQuizInput(widened), undefined, "the refinement owns the reject"); }); it("garbage never parses: non-object, missing question, missing correctAnswer", () => { assert.equal(parseQuizInput("a string"), undefined); assert.equal(parseQuizInput({ options: [] }), undefined); assert.equal(parseQuizInput({ ...PROBER_QUIZ, correctAnswer: undefined }), undefined); }); }); describe("parseAskInput", () => { it("parses free-text mode (no options) and option mode", () => { assert.ok(parseAskInput({ question: "What do you want to learn?" })); assert.ok(parseAskInput({ question: "Which topics?", options: [{ label: "Physics" }, { label: "Logic" }], multiSelect: true })); }); it("the live crash shape — stringified options — does not parse", () => { assert.equal(parseAskInput({ question: "q", options: "[]" }), undefined); }); }); describe("parseProbeDoneInput", () => { it("parses the valid assessment", () => { const parsed = parseProbeDoneInput(VALID_ASSESSMENT); assert.ok(parsed); assert.equal(parsed.assessment.strands.length, 2); assert.equal(parsed.assessment.summary, "Bracketed: teach the symbols on the intuition he has."); }); it("boundary parity: anything the tool schema accepts, the display parse accepts", () => { assert.ok(Value.Check(probeDoneInputSchema, VALID_ASSESSMENT)); assert.ok(parseProbeDoneInput(VALID_ASSESSMENT)); }); it("the live crash shape — assessment STRINGIFIED — does not parse", () => { // The exact double-encoding from the modal-realism run: strands died // on `.map` because the card cast the shape. The parse kills the class. const stringified = { assessment: JSON.stringify(VALID_ASSESSMENT.assessment) }; assert.equal(parseProbeDoneInput(stringified), undefined); const garbage = JSON.stringify(VALID_ASSESSMENT.assessment).slice(0, 30); assert.equal(parseProbeDoneInput(garbage), undefined); }); it("a mid-stream partial (no strands yet) does not parse", () => { assert.equal(parseProbeDoneInput({ assessment: { summary: "…" } }), undefined); }); }); describe("LaTeX-escape restoration (the JSON-debris fix, owner ruling 2026-09-26)", () => { // The exact live shapes from the modal-logic session: single-backslash // LaTeX inside JSON args — JSON escape resolution ate the command's // first letter (\f → FF ate 'f' of \forall, \v → VT ate 'v' of \varphi). it("parseQuizInput restores escape debris inside math spans (question + options + explanation)", () => { const parsed = parseQuizInput({ question: "What does \\(\\forall x\\, \\varphi(x)\\) mean?", options: [ { label: "\\(\x0barphi\\) is true when every element satisfies it", value: "tarski" }, { label: "\\(\x0corall\\) and \\(\\exists\\) are just shorthand", value: "shorthand" }, ], correctAnswer: "tarski", explanation: "\\(\x0corall\\) quantifies over the domain.", }); assert.ok(parsed); assert.equal(parsed.question, "What does \\(\\forall x\\, \\varphi(x)\\) mean?"); assert.equal(parsed.options[0]!.label, "\\(\\varphi\\) is true when every element satisfies it"); assert.equal(parsed.options[1]!.label, "\\(\\forall\\) and \\(\\exists\\) are just shorthand"); assert.equal(parsed.explanation, "\\(\\forall\\) quantifies over the domain."); }); it("parseAskInput restores debris in ask option labels too", () => { const parsed = parseAskInput({ question: "Goal?", options: [{ label: "\\(\x0corall\\) rigor" }, { label: "Intuition first" }], }); assert.ok(parsed); assert.equal(parsed.options![0]!.label, "\\(\\forall\\) rigor"); }); it("restores TAB debris (the \\to case) inside spans only", () => { const parsed = parseQuizInput({ question: "Is \\(p\to q\\) valid?", options: [{ label: "yes" }, { label: "no" }], correctAnswer: "no", explanation: "", }); assert.ok(parsed); assert.equal(parsed.question, "Is \\(p\\to q\\) valid?"); }); it("control chars OUTSIDE math spans are left alone (real prose/pasted text)", () => { const parsed = parseQuizInput({ question: "Pick:\ta\x0bb", options: [{ label: "x" }, { label: "y" }], correctAnswer: "x", explanation: "", }); assert.ok(parsed); assert.equal(parsed.question, "Pick:\ta\x0bb"); }); it("unbalanced delimiters pass through untouched (mid-stream safety)", () => { const parsed = parseQuizInput({ question: "\\(\\x0corall", options: [{ label: "x" }, { label: "y" }], correctAnswer: "x", explanation: "", }); assert.ok(parsed); assert.equal(parsed.question, "\\(\\x0corall"); }); }); describe("display-math newline rule (Option A, owner ruling 2026-09-27)", () => { // TS literal below: \\[ is \[, \\forall is intact \forall, \n is a REAL newline, \\\\ is the LaTeX \\ line break. const MULTILINE = "Compare:\n\\[\n\\forall x, & f(x) = x^2 \\\\\n& g(x) = x^3\n\\]"; it("real newlines inside display math are kept — multi-line equations pass through untouched", () => { const parsed = parseQuizInput({ question: MULTILINE, options: [{ label: "f grows slower" }, { label: "g grows slower" }], correctAnswer: "f grows slower", explanation: "", }); assert.ok(parsed); assert.equal(parsed.question, MULTILINE); }); it("non-newline debris inside display math still restores (the \\right CR case)", () => { const parsed = parseQuizInput({ question: "Evaluate \\[\\left( a \right)\\] carefully.", options: [{ label: "a" }, { label: "not a" }], correctAnswer: "a", explanation: "", }); assert.ok(parsed); assert.equal(parsed.question, "Evaluate \\[\\left( a \\right)\\] carefully."); }); it("sloppy \\neq in display math stays broken — the accepted cost (newline kept, byte-identical to a real line break)", () => { const debris = "Is \\[p \neq q\\] valid?"; const parsed = parseQuizInput({ question: debris, options: [{ label: "yes" }, { label: "no" }], correctAnswer: "no", explanation: "", }); assert.ok(parsed); assert.equal(parsed.question, debris); }); it("inline spans still restore newline debris (inline math is single-line by nature)", () => { const parsed = parseQuizInput({ question: "Is \\(p \neq q\\) valid?", options: [{ label: "yes" }, { label: "no" }], correctAnswer: "no", explanation: "", }); assert.ok(parsed); assert.equal(parsed.question, "Is \\(p \\neq q\\) valid?"); }); }); describe("dollar-delimiter recognition — existing escape restoration rules", () => { it("restores question, option, and explanation debris in dollar math", () => { const parsed = parseQuizInput({ question: "What does $\x0corall x$ mean?", options: [{ label: "$\x0barphi$ holds", value: "yes" }, { label: "No", value: "no" }], correctAnswer: "yes", explanation: "Here $p\to q$ is an implication.", }); assert.ok(parsed); assert.equal(parsed.question, "What does $\\forall x$ mean?"); assert.equal(parsed.options[0]!.label, "$\\varphi$ holds"); assert.equal(parsed.explanation, "Here $p\\to q$ is an implication."); }); it("keeps the same six-character inline recovery map", () => { for (const [debris, command] of [ ["\beta", "\\beta"], ["\to", "\\to"], ["\neq", "\\neq"], ["\x0barphi", "\\varphi"], ["\x0corall", "\\forall"], ["\rho", "\\rho"], ]) { assert.equal(restoreMathEscapes(`$${debris}$`), `$${command}$`); } }); it("preserves display newlines and real LaTeX row breaks", () => { const math = "Compare:\n$$\n\\begin{aligned}\nx &= 1 \\\\\ny &= 2\n\\end{aligned}\n$$"; assert.equal(restoreMathEscapes(math), math); }); it("restores five display debris characters but keeps newline ambiguity unchanged", () => { for (const [debris, command] of [ ["\beta", "\\beta"], ["\to", "\\to"], ["\x0barphi", "\\varphi"], ["\x0corall", "\\forall"], ["\\left(x \right)", "\\left(x \\right)"], ]) { assert.equal( restoreMathEscapes(`Compare:\n$$\n${debris}, p \neq q\n$$`), `Compare:\n$$\n${command}, p \neq q\n$$`, ); } }); it("preserves legacy replacement frequency when a span has multiple debris characters", () => { const body = "p\to q, \rho"; const legacy = restoreMathEscapes(`\\(${body}\\)`).slice(2, -2); assert.equal(restoreMathEscapes(`$${body}$`), `$${legacy}$`); assert.equal(legacy, "p\\to q, \rho", "the existing first-match-only behavior is unchanged"); }); it("shares dollar recovery with ask and free-recall fields", () => { const ask = parseAskInput({ question: "Goal?", options: [{ label: "$\x0corall$ rigor" }, { label: "Intuition" }] }); assert.ok(ask); assert.equal(ask.options![0]!.label, "$\\forall$ rigor"); const explain = parseExplainInput({ question: "Explain $p\to q$", correctAnswer: "$p\neq q$ is different." }); assert.ok(explain); assert.equal(explain.question, "Explain $p\\to q$"); assert.equal(explain.correctAnswer, "$p\\neq q$ is different."); }); it("leaves non-math debris, escaped opening dollars, and code examples untouched", () => { const prose = "Pasted:\t\x0cb. \\$\x0corall x$. A good $x$."; assert.equal(restoreMathEscapes(prose), prose); const code = "Literal `$\x0corall x$`.\n\n```text\n$\x0barphi$\n```\n\nReal $\x0corall x$."; assert.equal(restoreMathEscapes(code), code.replace("Real $\x0corall x$", "Real $\\forall x$")); }); it("leaves unbalanced and mismatched delimiter runs untouched", () => { for (const text of ["$\x0corall", "$$\n\x0corall x\n", "$\x0corall$$", "$$\x0corall$"]) { assert.equal(restoreMathEscapes(text), text); } }); it("recognizes native inline runs and longer display closing fences", () => { assert.equal(restoreMathEscapes("Inline $$p\neq q$$."), "Inline $$p\\neq q$$."); assert.equal(restoreMathEscapes("$$\n\x0corall x\n$$$"), "$$\n\\forall x\n$$$"); }); it("does not add an over-escaping repair or reprocess legacy spans", () => { const overescaped = "$$\n\\\\Omega\n$$"; assert.equal(restoreMathEscapes(overescaped), overescaped); const legacy = "\\[\np $a\neq b$\n\\]"; assert.equal(restoreMathEscapes(legacy), legacy); }); }); // ─── parseExplainInput (free-recall display parse, owner 2026-09-27) ─────── describe("parseExplainInput: the dock's display seam for explain calls", () => { const good = { question: "Explain why Mercury is hard to observe.", correctAnswer: "Its small angular separation from the Sun — it never strays far from the Sun in our sky.", nodeId: "planet-order", // the teacher's field — STRIPPED by Clean, like quiz details: "Think about where it appears at dusk.", }; it("parses a valid call and strips the teacher-only nodeId", () => { const parsed = parseExplainInput(good)!; assert.equal(parsed.question, "Explain why Mercury is hard to observe."); assert.equal(parsed.correctAnswer, "Its small angular separation from the Sun — it never strays far from the Sun in our sky."); assert.equal(parsed.details, "Think about where it appears at dusk."); assert.ok(!("nodeId" in parsed), "the teacher-only key is cleaned away"); }); it("an empty commitment never arms the dock", () => { assert.equal(parseExplainInput({ ...good, correctAnswer: " " }), undefined); }); it("restores LaTeX escape debris inside math spans (the pi-ai streaming wobble)", () => { // \f of \forall was eaten by the lenient args parse: U+000C + "orall". const dirty = "Because \( orall x\) holds."; const parsed = parseExplainInput({ ...good, correctAnswer: dirty })!; assert.equal(parsed.correctAnswer, "Because \(\forall x\) holds."); }); it("wrong-shape input never arms the dock (numeric strings DO coerce — the seam's leniency)", () => { assert.equal(parseExplainInput("just a string"), undefined); assert.equal(parseExplainInput({ explanation: "e", correctAnswer: "a" }), undefined, "missing question"); assert.equal(parseExplainInput({ question: "q", explanation: "e", correctAnswer: { nested: true } }), undefined); // 42 → "42": the same numeric coercion the quiz seam pins. assert.equal(parseExplainInput({ question: "q", explanation: "e", correctAnswer: 42 })?.correctAnswer, "42"); }); }); describe("over-escaped command heal at the parse seam (owner ruling 2026-10-15)", () => { // xiaomi/mimo-v2.6-pro and deepseek-v4.1-flash stochastically DOUBLE // LaTeX command backslashes inside JSON tool arguments — per-call and // per-field (gradient-descent msg 3: question doubled, explanation and // options in the SAME call clean). restoreMathEscapesDeep now rides // healOverescapedCommands after the debris restore, so doubled spans // heal at every display parse: live dock, settled cards, old records. const CLEAN = "Let $f : \\mathbb{R}^n \\to \\mathbb{R}$ be differentiable at $x$, with $\\nabla f(x) \\neq 0$. Among all unit directions $\\|u\\|_2 = 1$, which makes $D_u f(x)$ most negative?"; const DOUBLE = (s: string): string => s.replace(/\\/g, "\\\\"); const BASE = { options: [ { label: "$u^\\star = -\\nabla f(x)$", value: "neg" }, { label: "$u^\\star = \\nabla f(x)$", value: "pos" }, ], correctAnswer: "neg", explanation: "By Cauchy\u2013Schwarz: $$u^\\star = -\\frac{\\nabla f(x)}{\\|\\nabla f(x)\\|_2}$$", }; it("parseQuizInput heals a doubled question field (the gradient-descent msg-3 shape)", () => { const parsed = parseQuizInput({ ...BASE, question: DOUBLE(CLEAN) }); assert.ok(parsed, "parses"); assert.equal(parsed.question, CLEAN); // The CLEAN sibling fields of the same call stay byte-identical. assert.equal(parsed.options[0]!.label, BASE.options[0]!.label); assert.equal(parsed.explanation, BASE.explanation); }); it("parseQuizInput heals a quadrupled field (the agent's over-correction, msg 6)", () => { const parsed = parseQuizInput({ ...BASE, question: DOUBLE(DOUBLE(CLEAN)) }); assert.ok(parsed, "parses"); assert.equal(parsed.question, CLEAN); }); it("doubled option labels and explanations heal too (the inference-runtime msg-7 shape)", () => { const parsed = parseQuizInput({ question: "Let $A \\in \\mathbb{R}^{m \\times k}$ and $C = AB$.", options: [ { label: DOUBLE("$mkn$"), value: "mkn" }, { label: DOUBLE("$2mkn$"), value: "2" }, ], correctAnswer: "mkn", explanation: DOUBLE("$$c_{ij} = \\sum_{\\ell=1}^{k} a_{i\\ell}\\, b_{\\ell j}$$"), }); assert.ok(parsed, "parses"); assert.equal(parsed.options[0]!.label, "$mkn$"); assert.equal(parsed.options[1]!.label, "$2mkn$"); assert.equal(parsed.explanation, "$$c_{ij} = \\sum_{\\ell=1}^{k} a_{i\\ell}\\, b_{\\ell j}$$"); }); it("parseAskInput heals doubled ask-option descriptions the same way", () => { const parsed = parseAskInput({ question: "Depth?", options: [ { label: "Rigorous", description: DOUBLE("why $-\\nabla f$ is steepest descent") }, { label: "Intuition" }, ], }); assert.ok(parsed, "parses"); assert.equal(parsed.options![0]!.description, "why $-\\nabla f$ is steepest descent"); }); it("clean records pass through byte-identical (no over-heal)", () => { const parsed = parseQuizInput({ ...BASE, question: CLEAN }); assert.ok(parsed, "parses"); assert.equal(parsed.question, CLEAN); assert.equal(parsed.explanation, BASE.explanation); }); it("legitimate row breaks inside matrices survive the seam", () => { const matrix = "Which matrix matches? $$\\begin{pmatrix} 1 \\ 2 \\\\ 3 \\ 4\\end{pmatrix}$$"; const parsed = parseQuizInput({ ...BASE, question: matrix }); assert.ok(parsed, "parses"); assert.equal(parsed.question, matrix); }); });