feat: add website enrichment crawler

tests/lead-discovery-source.test.ts

@@ -0,0 +1,84 @@
+import assert from "node:assert/strict";
+import { readFileSync } from "node:fs";
+import path from "node:path";
+import test from "node:test";
+
+const leadDiscoveryPath = path.join(process.cwd(), "convex", "leadDiscovery.ts");
+const leadDiscoverySource = readFileSync(leadDiscoveryPath, "utf8");
+
+function hasPattern(source: string, pattern: RegExp) {
+  return pattern.test(source);
+}
+
+function extractExportSource(name: string) {
+  const marker = `export const ${name} = `;
+  const declarationIndex = leadDiscoverySource.indexOf(marker);
+
+  assert.notEqual(declarationIndex, -1, `Expected declaration for ${name}`);
+
+  const openBraceIndex = leadDiscoverySource.indexOf("{", declarationIndex);
+  let depth = 0;
+  let end = -1;
+
+  for (let index = openBraceIndex; index < leadDiscoverySource.length; index++) {
+    const char = leadDiscoverySource[index];
+
+    if (char === "{") {
+      depth += 1;
+    } else if (char === "}") {
+      depth -= 1;
+      if (depth === 0) {
+        end = index;
+        break;
+      }
+    }
+  }
+
+  assert.notEqual(end, -1, `Expected balanced braces for ${name}`);
+  return leadDiscoverySource.slice(openBraceIndex, end + 1);
+}
+
+test("startCampaignRun checks active campaign runs via by_type_and_status", () => {
+  const source = extractExportSource("startCampaignRun");
+
+  assert.equal(
+    hasPattern(
+      source,
+      /withIndex\(\s*"by_type_and_status"\s*,\s*\(q\)\s*=>[\s\S]*?q\.eq\("type",\s*"campaign"\)\.eq\("status",\s*"running"\),?[\s\S]*?\)/,
+    ),
+    true,
+    "Campaign starts should only consider running campaign-type runs as blockers",
+  );
+});
+
+test("persistDiscoveredLeads does not schedule website enrichment jobs directly", () => {
+  const source = extractExportSource("persistDiscoveredLeads");
+
+  assert.equal(
+    source.includes("ctx.scheduler.runAfter"),
+    false,
+    "Lead persistence must not call runAfter",
+  );
+});
+
+test("processCampaignRun schedules website enrichment after lead persistence", () => {
+  const source = extractExportSource("processCampaignRun");
+
+  const persistIndex = source.indexOf(
+    "internal.leadDiscovery.persistDiscoveredLeads",
+  );
+  const queueCall = source.indexOf("internal.websiteEnrichment.queueLeadEnrichment");
+  const eventMessageIndex = source.indexOf("Website-Kontaktanreicherung geplant.");
+
+  assert.notEqual(persistIndex, -1, "processCampaignRun should persist discovered leads");
+  assert.notEqual(queueCall, -1, "processCampaignRun should schedule website enrichment");
+  assert.notEqual(eventMessageIndex, -1, "processCampaignRun should append enrichment schedule events");
+  assert.ok(
+    persistIndex < queueCall,
+    "processCampaignRun should schedule enrichment after persistence succeeds",
+  );
+  assert.ok(
+    queueCall < eventMessageIndex,
+    "processCampaignRun should append enrichment event after scheduling",
+  );
+});