Merge pull request '扩展 AI 公司研究并修复收入统计表布局' (#31) from codex/ai-research-cn-sme into main

Reviewed-on: #31

apps/docs/scripts/auth-cli.ts

@@ -182,10 +182,10 @@ async function main() {
   assert.equal(researchResponse.status, 200, "authenticated research API");
   assert.ok(researchResponse.headers.get("cache-control")?.includes("no-store"));
   const research = await researchResponse.json();
-  assert.equal(research.companies.length, 1029);
+  assert.equal(research.companies.length, 1051);
   assert.equal(research.categories.length, 11);
-  assert.equal(research.evidence.length, 1412);
-  assert.equal(research.observations.length, 192);
+  assert.equal(research.evidence.length, 1858);
+  assert.equal(research.observations.length, 237);
   assert.equal((await request("/api/research/ai-companies", { method: "POST", headers })).status, 405);
   const researchPage = await request("/docs/other-research/ai-company-research/overview", { headers });
   assert.equal(researchPage.status, 200);

apps/docs/src/components/IncomeDistribution.tsx

@@ -75,7 +75,7 @@ export default function IncomeDistribution({ data, onSelect }: {
         </svg></div>
         <figcaption>圆点：产品观测　竖线：样本均值　菱形：样本中位数。业务类别标题下各行按支付平台{splitModels && "及收费与交付模式"}分别统计；n为该组产品数。点击点或下表产品名可看原始证据。</figcaption>
       </figure>
-      <div className="research-table-wrap"><table><caption>{metric}样本统计；每行是类别下的一个统计组，CV使用样本标准差</caption><thead><tr><th scope="col">业务类别</th><th scope="col">支付平台 / 收费模式</th><th scope="col">n</th><th scope="col">均值</th><th scope="col">中位数</th><th scope="col">均值/中位数</th><th scope="col">CV</th><th scope="col">去除最大值后的均值 / 下降幅度</th><th scope="col">产品与限制</th></tr></thead>{categoryGroups.map(({ category, rows }) => <tbody key={category}>{rows.map((group, index) => <tr key={group.key}>
+      <div className="research-table-wrap"><table className="income-statistics"><caption>{metric}样本统计；每行是类别下的一个统计组，CV使用样本标准差</caption><thead><tr><th scope="col">业务类别</th><th scope="col">支付平台 / 收费模式</th><th scope="col">n</th><th scope="col">均值</th><th scope="col">中位数</th><th scope="col">均值/中位数</th><th scope="col">CV</th><th scope="col">去除最大值后的均值 / 下降幅度</th><th scope="col">产品与限制</th></tr></thead>{categoryGroups.map(({ category, rows }) => <tbody key={category}>{rows.map((group, index) => <tr key={group.key}>
         {index === 0 && <th scope="rowgroup" rowSpan={rows.length}>{categories.get(category)}<small>{rows.length}个统计组</small></th>}
         <td>{group.provider}<small>{group.model}</small></td><td>{group.stats.n}</td><td>{usd(group.stats.mean)}</td><td>{usd(group.stats.median)}</td><td>{group.stats.meanMedianRatio?.toFixed(2) ?? "—"}</td><td>{group.stats.cv?.toFixed(2) ?? "—"}</td><td>{usd(group.stats.meanWithoutLargest)}<small>{percent(group.stats.largestRemovalDrop)}</small></td>
         <td>{group.points.map((point) => <span key={point.company.id}><button className="research-company-name" onClick={() => onSelect(point.company.id)}>{point.company.name}</button>{" "}</span>)}<small>{group.stats.n < 3 ? "样本少于3，不能据此判断领域典型收入。" : "自选的小样本，须结合产品年龄、成本与留存评估。"}</small></td>

apps/docs/src/styles/globals.css

@@ -1473,6 +1473,13 @@ svg { flex-shrink: 0; }
 
 .research-distribution { margin: 24px 0 38px; }
 .research-distribution figure { margin: 20px 0; }
+.ai-research .income-statistics { min-width: 1200px; }
+/* Row-spanning category headings shift cell indexes between rows. */
+.ai-research .income-statistics td { min-width: 0; }
+.ai-research .income-statistics th[scope="rowgroup"] { min-width: 150px; white-space: nowrap; }
+.ai-research .income-statistics th[scope="rowgroup"] small { display: block; margin-top: 5px; }
+.ai-research .income-statistics td:not(:last-child) { white-space: nowrap; }
+.ai-research .income-statistics td:last-child { min-width: 220px; }
 .income-distribution-chart { display: block; width: 100%; min-width: 900px; height: auto; }
 .income-distribution-chart [role="button"] { cursor: pointer; outline: none; }
 .income-distribution-chart [role="button"]:focus circle:last-child { stroke: #b17725; stroke-width: 3; }

apps/docs/tests/research.test.ts

@@ -23,9 +23,9 @@ test("research revisions share the authentication database and preserve replay l
     for (let i = 0; i < 20; i++) consumeCode(config, "invalid");
     assert.equal(consumeCode(config, "invalid"), "limited");
     const initial = readResearchSnapshot();
-    assert.equal(initial.companies.length, 1029);
-    assert.equal(initial.evidence.length, 1412);
-    assert.equal(initial.observations.length, 192);
+    assert.equal(initial.companies.length, 1051);
+    assert.equal(initial.evidence.length, 1858);
+    assert.equal(initial.observations.length, 237);
     assert.deepEqual(initial.candidateBatches, seed.candidateBatches, "candidate gaps persist in the same database metadata");
     assert.deepEqual(initial.directoryRechecks, seed.directoryRechecks, "follow-up classifications persist without rewriting the discovery audit");
     assert.deepEqual(readResearchSnapshot(), initial, "reopening does not duplicate records");
@@ -64,9 +64,9 @@ test("research validation rejects broken evidence and unsafe source links", () =
 
 test("reviews link explicit companies without turning global or excluded cases into companies", () => {
   const snapshot = structuredClone(seed) as ResearchSnapshot;
-  assert.equal(snapshot.reviews.filter((r) => r.scope === "companies").length, 74);
-  assert.equal(snapshot.reviews.filter((r) => r.scope === "dataset").length, 10);
-  assert.equal(snapshot.reviews.filter((r) => r.scope === "excluded-candidate").length, 37);
+  assert.equal(snapshot.reviews.filter((r) => r.scope === "companies").length, 115);
+  assert.equal(snapshot.reviews.filter((r) => r.scope === "dataset").length, 12);
+  assert.equal(snapshot.reviews.filter((r) => r.scope === "excluded-candidate").length, 44);
   const review = snapshot.reviews.find((r) => r.id === "R4")!;
   assert.deepEqual(reviewCompanyIds(snapshot, review), ["AI007", "AI013", "AI021", "AI034"]);
   assert.deepEqual(reviewCompanyIds(snapshot, snapshot.reviews.find((r) => r.id === "CN-R-PSI")!), []);
@@ -85,20 +85,24 @@ test("reviews link explicit companies without turning global or excluded cases i
 test("disclosure counts exclude client income, biographies, targets and duplicate milestones", () => {
   const snapshot = seed as ResearchSnapshot;
   const summary = categorySummary(snapshot);
-  assert.equal(summary.reduce((n, c) => n + c.count, 0), 1029);
-  assert.equal(summary.reduce((n, c) => n + c.self, 0), 28);
+  assert.equal(summary.reduce((n, c) => n + c.count, 0), 1051);
+  assert.equal(summary.reduce((n, c) => n + c.self, 0), 43);
   assert.equal(summary.reduce((n, c) => n + c.estimate, 0), 2);
-  assert.equal(summary.reduce((n, c) => n + c.platform, 0), 16);
-  assert.equal(summary.reduce((n, c) => n + c.unknown, 0), 983);
+  assert.equal(summary.reduce((n, c) => n + c.platform, 0), 19);
+  assert.equal(summary.reduce((n, c) => n + c.unknown, 0), 987);
   assert.equal(snapshot.companies.find((c) => c.name === "Rork")?.revenueStatus, "unknown");
   assert.equal(snapshot.companies.find((c) => c.name === "Result")?.revenueStatus, "unknown");
   assert.ok(snapshot.observations.filter((o) => o.selection === "signal").every((o) => o.scope === "公司整体"));
+  const arena = snapshot.observations.find((o) => o.id === "COARENA-O-CUMULATIVE")!;
+  assert.equal(arena.selection, "historical", "cumulative launch revenue is not a recurring company-wide signal");
+  assert.equal(arena.comparisonGroup, null);
+  assert.ok(arena.metric !== "mrr" && arena.metric !== "arr", "a two-week cumulative figure cannot become MRR or ARR");
 });
 
 test("technical reviews keep accessible materials separate from tested capabilities", () => {
   const snapshot = structuredClone(seed) as ResearchSnapshot;
-  assert.equal(snapshot.companies.filter((c) => c.technicalReview).length, 28);
-  assert.equal(snapshot.companies.filter((c) => c.technicalReview?.level === "public-artifact").length, 7);
+  assert.equal(snapshot.companies.filter((c) => c.technicalReview).length, 59);
+  assert.equal(snapshot.companies.filter((c) => c.technicalReview?.level === "public-artifact").length, 9);
   assert.equal(snapshot.companies.filter((c) => c.technicalReview?.level === "independent-test").length, 1);
   assert.equal(snapshot.companies.find((c) => c.id === "AI051")?.revenueStatus, "unknown");
   assert.match(snapshot.companies.find((c) => c.id === "AI051")!.technicalReview!.testScope!, /Sep 2026/);
@@ -137,6 +141,43 @@ test("ambiguous ARR claims retain currency, precision and attribution limits", (
   assert.equal(snapshot.companies.find((c) => c.id === "AI070")?.revenueStatus, "unknown");
   assert.equal(snapshot.observations.filter((o) => o.company === "AI070").length, 0,
     "unreadable announcement summaries do not become numeric ARR");
+  const waitlist = snapshot.observations.find((o) => o.id === "SR25-O-OMI-WAITLIST")!;
+  assert.equal(waitlist.selection, "excluded", "waitlist conversions are not paid recurring revenue");
+  assert.equal(snapshot.companies.find((c) => c.id === waitlist.company)?.revenueStatus, "unknown");
+  const bluma = snapshot.observations.find((o) => o.id === "BLUMA-O-HISTORICAL-MRR")!;
+  assert.equal(bluma.selection, "historical");
+  assert.ok(bluma.unit !== "USD" && !bluma.observedAt && bluma.comparisonGroup === null);
+  const native = snapshot.observations.find((o) => o.id === "NATIVE-O-ARR-CLAIM")!;
+  assert.equal(native.selection, "crosscheck");
+  assert.ok(!native.metric && !native.observedAt && native.comparisonGroup === null);
+  const marketVolume = snapshot.observations.find((o) => o.id === "PALISADE-O-GMV")!;
+  assert.equal(marketVolume.selection, "excluded", "customer GMV must not establish company revenue or ARR");
+  assert.equal(snapshot.companies.find((c) => c.id === marketVolume.company)?.revenueStatus, "unknown");
+  const pingo = snapshot.observations.find((o) => o.id === "PINGO-O-MRR-JUL25")!;
+  assert.equal(pingo.selection, "historical", "a July 2025 self-report is not a current payment-platform sample");
+  assert.ok(pingo.unit !== "USD" && !pingo.observedAt && pingo.comparisonGroup === null);
+  const designService = snapshot.observations.find((o) => o.id === "SEMIOTIC-O-LAUNCH-REVENUE")!;
+  assert.equal(designService.selection, "historical");
+  assert.ok(!designService.metric && designService.comparisonGroup === null, "four-week service revenue is not MRR or ARR");
+  const restaurantAdoption = snapshot.observations.filter((o) => o.company === "YC32815");
+  assert.ok(restaurantAdoption.length > 0 && restaurantAdoption.every((o) => o.selection === "adoption" && !o.metric),
+    "restaurant count and list price must not produce an invented MRR observation");
+  const fishPilot = snapshot.observations.find((o) => o.id === "OCTAPULSE-O-CONTRACT")!;
+  assert.equal(fishPilot.selection, "excluded", "a paid pilot or annual contract does not establish recognized recurring revenue");
+  assert.equal(snapshot.companies.find((c) => c.id === fishPilot.company)?.revenueStatus, "unknown");
+  assert.equal(fishPilot.comparisonGroup, null);
+  assert.notEqual(fishPilot.unit, "USD", "six-figure wording does not establish the currency");
+  assert.ok(!fishPilot.metric && !fishPilot.observedAt, "the publication date is not the financial reporting date");
+  const cedar = snapshot.observations.filter((o) => o.company === "AI102");
+  assert.ok(cedar.every((o) => o.selection !== "signal" && o.comparisonGroup === null),
+    "closed and contracted amounts without recognition or period do not establish ARR");
+  assert.equal(snapshot.observations.find((o) => o.id === "SR25-O-LOOPS-CUSTOMER-SALES")?.selection, "excluded");
+  const portfolio = snapshot.observations.filter((o) => o.company === "AI100");
+  assert.deepEqual(portfolio.filter((o) => o.selection === "signal").map((o) => o.id), ["SR25-O-DRS-PORTFOLIO"],
+    "a product milestone must not be added to its company portfolio as another company");
+  assert.ok(snapshot.observations.filter((o) => o.id.startsWith("SR25-O-")).every((o) =>
+    o.unit !== "USD" && !o.metric && !o.observedAt && o.comparisonGroup === null),
+  "access dates and dollar symbols must not invent dated USD income samples");
 });
 
 // Retractions and forecasts must stay visible for review without becoming revenue signals.
@@ -179,8 +220,18 @@ test("directory coverage reconciles every candidate and keeps profile-only incom
     assert.ok(company.profileOnly || company.directoryOrigin === "yc-2025-2026");
     assert.equal(company.url, row.source);
     if (company.profileOnly) assert.equal(company.revenueStatus, "unknown");
-    assert.equal(company.year, row.foundedYear);
-    assert.ok([2025, 2026].includes(company.year!));
+    if (company.id === "YC31370") {
+      assert.equal(row.foundedYear, 2025, "original discovery remains unchanged");
+      assert.equal(company.year, null, "a pivot year cannot settle the company founding year");
+      assert.ok(snapshot.reviews.some((review) => review.id === "DAYJOB-R-COHORT" && review.companyIds?.includes(company.id)));
+    } else if (company.id === "YC36279") {
+      assert.equal(row.foundedYear, 2026, "original Native discovery remains unchanged");
+      assert.equal(company.year, null, "a model launch and YC date cannot settle the older business founding year");
+      assert.ok(snapshot.reviews.some((review) => review.id === "NATIVE-R-COHORT" && review.companyIds?.includes(company.id)));
+    } else {
+      assert.equal(company.year, row.foundedYear);
+      assert.ok([2025, 2026].includes(company.year!));
+    }
   }
   const invalid = structuredClone(snapshot);
   invalid.coverage!.newProfiles += 1;
@@ -194,6 +245,20 @@ test("directory coverage reconciles every candidate and keeps profile-only incom
   assert.throws(() => validateResearch(lostOrigin), /coverage accounting/);
 });
 
+test("business pivots and procurement volume do not become comparable startup ARR", () => {
+  const snapshot = seed as ResearchSnapshot;
+  const dayjob = snapshot.companies.find((company) => company.id === "YC31370")!;
+  assert.equal(dayjob.year, null);
+  assert.equal(dayjob.directoryOrigin, "yc-2025-2026");
+  const milestones = snapshot.observations.filter((observation) => observation.company === dayjob.id);
+  assert.deepEqual(milestones.filter((observation) => observation.selection === "crosscheck").map((observation) => observation.value).sort(), [496000, 653000]);
+  assert.ok(milestones.every((observation) => observation.comparisonGroup === null && !observation.metric && !observation.observedAt));
+  assert.ok(milestones.filter((observation) => [150000, 800000].includes(observation.value)).every((observation) => observation.selection === "excluded"));
+  const procurement = snapshot.observations.find((observation) => observation.id === "SAUDARA-O-SOURCING")!;
+  assert.equal(procurement.selection, "excluded");
+  assert.equal(procurement.comparisonGroup, null);
+});
+
 test("income descriptions report sample dispersion and maximum-value sensitivity", () => {
   const stats = describeIncome([100, 200, 300])!;
   assert.equal(stats.mean, 200);
@@ -215,14 +280,21 @@ test("income descriptions report sample dispersion and maximum-value sensitivity
 test("platform samples exclude stale data and keep ARR, estimates and collections separate", () => {
   const snapshot = seed as ResearchSnapshot;
   const recent = platformIncomeSample(snapshot, { windowDays: 7 });
-  assert.equal(recent.points.length, 14);
-  assert.equal(recent.excluded.length, 3);
+  assert.equal(recent.points.length, 16);
+  assert.equal(recent.excluded.length, 4);
   assert.ok(recent.points.every((p) => p.observation.metric === "mrr" && p.observation.selection === "platform"));
-  assert.deepEqual(new Set(recent.excluded.map((p) => p.observation.company)), new Set(["AI016", "MR013", "MR014"]));
+  assert.deepEqual(new Set(recent.excluded.map((p) => p.observation.company)), new Set(["AI016", "MR013", "MR014", "MR017"]));
   assert.ok(recent.points.some((p) => p.company.name === "TrackAI" && p.observation.connectionExpired), "dated recent value is not a claim of active connection");
   const older = platformIncomeSample(snapshot, { windowDays: 90 });
-  assert.equal(older.points.length, 15);
+  assert.equal(older.points.length, 18);
   assert.ok(older.points.some((p) => p.company.id === "AI016"));
+  assert.ok(!recent.points.some((p) => p.company.id === "MR017"));
+  assert.ok(older.points.some((p) => p.company.id === "MR017" && p.observation.connectionExpired),
+    "an expired connection retains its historical observation date in the wider window");
+  assert.equal(recent.points.find((p) => p.company.id === "MR018")?.value, 67864,
+    "an older Markdown cache must not replace or duplicate the later web snapshot");
+  assert.equal(recent.points.find((p) => p.company.id === "MR019")?.value, 65431,
+    "period-table revenue and revenue snapshots must not replace MRR");
   const scaled = platformIncomeSample(snapshot, { windowDays: 7, annualize: true });
   const monthlyGroups = groupIncome(recent.points);
   const annualGroups = groupIncome(scaled.points);
@@ -307,7 +379,7 @@ test("unreconciled directory financials remain source reviews instead of income
   assert.ok(review.evidenceIds.includes("PH-TANA-DR"));
   assert.equal(reviewCompanyIds(snapshot, review).length, 0);
   assert.ok(!snapshot.observations.some((o) => o.evidence === "PH-TANA-DR"));
-  assert.equal(snapshot.candidateBatches!.flatMap((b) => b.records).filter((r) => r.status === "pending-year").length, 10);
+  assert.equal(snapshot.candidateBatches!.flatMap((b) => b.records).filter((r) => r.status === "pending-year").length, 18);
 });
 
 test("product pivots and founder histories do not become new company years or current revenue", () => {
@@ -320,6 +392,15 @@ test("product pivots and founder histories do not become new company years or cu
   assert.equal(candidate.companyId, pivot.id);
   assert.ok(!platformIncomeSample(snapshot, { windowDays: null }).points.some((point) => point.company.id === pivot.id));
   assert.ok(!snapshot.observations.some((o) => ["PH-FUNDRAISLY-ABOUT", "LF-PIVOT", "LF-HISTORY"].includes(o.evidence)), "other ventures' ARR, users and funding must not transfer");
+  const candidates = snapshot.candidateBatches!.flatMap((b) => b.records);
+  assert.equal(candidates.find((r) => r.id === "PH26-12")?.status, "outside-cohort",
+    "a brand iteration with an earlier business chain is not a new company");
+  const happycapy = candidates.find((r) => r.id === "PH26-04")!;
+  assert.ok(happycapy.evidenceIds?.includes("FOLLOWUP-CAPY-ACRA"));
+  assert.equal(happycapy.status, "pending-year", "a seller's registration does not establish a new global business");
+  assert.equal(happycapy.companyId, undefined);
+  assert.ok(!snapshot.observations.some((o) => o.evidence.startsWith("FOLLOWUP-")),
+    "registration, funding, app-store prices and client amounts are not company income");
 });
 
 test("follow-up evidence can admit an uncertain directory candidate without rewriting its original status", () => {