chore: update search scripts, extra sources and post drafts

Adds the one-off scripts accumulated across recent sessions, removes temp scripts whose searches now live in config.ts as role-family sweeps, and updates fetch-other.ts plus the LinkedIn post drafts.
2026-08-27 09:03:05 +02:00
parent 3f5360f44c
commit b4957025b1
31 changed files with 2008 additions and 283 deletions
@@ -0,0 +1,82 @@
+#!/usr/bin/env bun
+import type { SearchConfig } from "../config.ts";
+import { fetchJobs, isEnglishContent, JUNIOR_TITLE_RE, JAVA_TITLE_RE } from "../fetch.ts";
+
+const CSV = "./results/jobs.csv";
+
+// Titles the engineer-centric KEYWORDS list never searches for.
+const BROAD_KEYWORDS = [
+  "Implementation Manager",
+  "Implementation Consultant",
+  "Implementation Specialist",
+  "Solutions Consultant",
+  "Solution Architect",
+  "Technical Consultant",
+  "Technical Account Manager",
+  "Technical Program Manager",
+  "Technical Project Manager",
+  "Delivery Manager",
+  "Integration Engineer",
+  "Integration Specialist",
+  "Customer Success Engineer",
+  "Sales Engineer",
+  "Presales Engineer",
+  "Digital Transformation",
+  "Product Owner",
+  "Scientific Software",
+  "Research Software",
+  "Automation Engineer",
+  "Systems Integration",
+].join(" OR ");
+
+// Microsoft stack = hard no (see cv/polo.md).
+const MICROSOFT_RE = /\b(\.net|dotnet|c#|azure|sharepoint|dynamics|power\s*(platform|bi|apps|automate)|sccm|intune|windows\s*server|m365|office\s*365|uipath)\b/i;
+const SAP_RE = /\bsap\b|\bs\/4hana\b|\bsalesforce\b|\bservicenow\b/i;
+
+const AREAS: SearchConfig[] = [
+  { label: "Basel (city)", keywords: BROAD_KEYWORDS, geoId: "103383283", location: "Basel", workType: ["onsite", "hybrid"], timePosted: "week" },
+  { label: "Basel Metropolitan Area", keywords: BROAD_KEYWORDS, geoId: "90009885", location: "Basel Metropolitan Area", workType: ["onsite", "hybrid"], timePosted: "week" },
+  { label: "Basel-Country", keywords: BROAD_KEYWORDS, geoId: "107117530", location: "Basel-Country", workType: ["onsite", "hybrid"], timePosted: "week" },
+];
+
+const existingIds = new Set<string>();
+try {
+  const text = await Bun.file(CSV).text();
+  for (const line of text.split("\n").slice(1)) {
+    const id = line.match(/\/jobs\/view\/(\d+)\//)?.[1];
+    if (id) existingIds.add(id);
+  }
+} catch { /* first run */ }
+
+type Row = { id: string; title: string; company: string; location: string; link: string; posted: string; isNew: boolean };
+const seen = new Set<string>();
+const rows: Row[] = [];
+
+for (const config of AREAS) {
+  // raw: true — bypass SOFTWARE_TITLE_RE so non-engineer titles survive
+  const jobs = await fetchJobs(config, { raw: true });
+  console.error(`[${config.label}] ${jobs.length} jobs`);
+  for (const job of jobs) {
+    if (seen.has(job.id)) continue;
+    seen.add(job.id);
+    const t = job.title;
+    if (JUNIOR_TITLE_RE.test(t)) { console.error(`  junior: ${t}`); continue; }
+    if (JAVA_TITLE_RE.test(t)) { console.error(`  java: ${t}`); continue; }
+    if (MICROSOFT_RE.test(t)) { console.error(`  MS stack: ${t}`); continue; }
+    if (SAP_RE.test(t)) { console.error(`  SAP/SF/SNOW: ${t}`); continue; }
+    rows.push({
+      id: job.id, title: t, company: job.company,
+      location: (job as any).location ?? "", link: job.link,
+      posted: (job as any).postedAt ?? "", isNew: !existingIds.has(job.id),
+    });
+  }
+}
+
+const survivors: Row[] = [];
+for (const r of rows) {
+  if (await isEnglishContent(r.id)) survivors.push(r);
+  else console.error(`  DE/FR content dropped: ${r.title} — ${r.company}`);
+}
+
+console.error(`\n${survivors.length} survivors (${survivors.filter((r) => r.isNew).length} not in CSV)`);
+console.log(JSON.stringify(survivors, null, 2));