fix(jobindex-search): decode hex HTML entities in CLI output (#56)

decodeHtmlEntities (duplicated in src/helpers.ts and src/commands/detail.ts) only handled decimal numeric character references (é); the equally valid hexadecimal form (é) fell through undecoded and surfaced as raw text in titles, companies, locations and descriptions. This bites Danish content especially (ae/o/aa often arrive as entities). It also used String.fromCharCode, which corrupts supplementary-plane code points (e.g. emoji, U+1F600). Add a hexadecimal numeric-entity rule and route both decimal and hex through a fromCodePoint-based helper with a valid-range guard, in both copies. Add network-free unit tests via the exported parseJobCards.
2026-09-17 00:26:26 +00:00 · 2026-07-07 19:41:23 +02:00
parent b27a3b5e81
commit 844b245583
3 changed files with 70 additions and 2 deletions
@@ -0,0 +1,46 @@
+import { describe, test, expect } from "bun:test";
+import { parseJobCards } from "../src/helpers";
+
+// Minimal jobad-wrapper markup: parseJobCards splits on `jobad-wrapper-<id>`
+// and reads the title from the <h4><a href>…</a></h4> and the company from the
+// jix-toolbar-top__company link. We inject HTML entities to exercise decoding.
+function card(id: string, title: string, company = "Acme A/S"): string {
+  return `<div id="jobad-wrapper-${id}" class="PaidJob">
+    <h4><a href="https://www.jobindex.dk/jobannonce/${id}">${title}</a></h4>
+    <div class="jix-toolbar-top__company">
+      <a href="https://www.jobindex.dk/virksomhed/acme">${company}</a>
+    </div>
+  </div>`;
+}
+
+describe("decodeHtmlEntities (via parseJobCards)", () => {
+  test("decodes hexadecimal numeric entities (&#xF8; -> ø)", () => {
+    const [c] = parseJobCards(card("h1", "Sm&#xF8;rrebro Chef"));
+    expect(c.title).toBe("Smørrebro Chef");
+  });
+
+  test("decodes uppercase-X hexadecimal entities (&#XE6; -> æ)", () => {
+    const [c] = parseJobCards(card("h2", "K&#XE6;re Kollega"));
+    expect(c.title).toBe("Kære Kollega");
+  });
+
+  test("still decodes decimal numeric entities (&#229; -> å) — regression", () => {
+    const [c] = parseJobCards(card("h3", "&#229;rhus Lead"));
+    expect(c.title).toBe("århus Lead");
+  });
+
+  test("decodes supplementary-plane code points with fromCodePoint (&#128512;)", () => {
+    const [c] = parseJobCards(card("h4", "Growth &#128512;"));
+    expect(c.title).toBe("Growth 😀");
+  });
+
+  test("decodes hex supplementary-plane code points (&#x1F600;)", () => {
+    const [c] = parseJobCards(card("h5", "Growth &#x1F600;"));
+    expect(c.title).toBe("Growth 😀");
+  });
+
+  test("decodes hex entities in the company name too", () => {
+    const [c] = parseJobCards(card("h6", "Engineer", "N&#xF8;rrebro ApS"));
+    expect(c.company).toBe("Nørrebro ApS");
+  });
+});