feat(qajobsboard): fetch job detail pages for concrete location text

The jobs.json feed often labels roles Remote/Worldwide while the public job page JSON-LD and description include the real city (e.g. Mumbai/Nagpur). Enrich vague rows by reading each QAJobsBoard detail URL before import. Co-authored-by: Cursor <cursoragent@cursor.com>
2026-05-16 17:20:53 -04:00
parent 0a63316100
commit 03d293699a
3 changed files with 263 additions and 34 deletions
@@ -0,0 +1,33 @@
+import { describe, expect, it } from "vitest";
+import {
+  extractJobLocationFromText,
+  parseQaJobDetailPage,
+} from "../src/detail-page.js";
+
+const SAMPLE_JSON_LD = `<!DOCTYPE html><html><body>
+<script type="application/ld+json">
+{
+  "@context": "https://schema.org/",
+  "@type": "JobPosting",
+  "title": "Sr. QA Automation Engineer",
+  "description": "<p><strong>Job Location: Mumbai/Nagpur</strong></p><p>India | UK | USA</p>",
+  "jobLocationType": "TELECOMMUTE"
+}
+</script>
+</body></html>`;
+
+describe("parseQaJobDetailPage", () => {
+  it("extracts job location from JSON-LD description on the detail page", () => {
+    const result = parseQaJobDetailPage(SAMPLE_JSON_LD);
+    expect(result?.location).toBe("Mumbai/Nagpur");
+    expect(result?.jobDescription).toContain("Mumbai/Nagpur");
+  });
+
+  it("extracts job location lines from plain text", () => {
+    expect(
+      extractJobLocationFromText(
+        "Job Location: Vancouver, British Columbia, Canada",
+      ),
+    ).toBe("Vancouver, British Columbia, Canada");
+  });
+});