Add MojeDelo scraper

MojeDelo has no public API; use the internal JSON API its search page
calls. Tenant headers are read from the site's public config each run.
Takes a search URL copied from the site, paginates, and checks results
against the requested category/region filters.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

scrapper/README.md

@@ -11,6 +11,7 @@ No runtime dependencies. It uses Bun's built-in SQLite, HTTP server and HTMLRewr
 | slo-tech.com (Ljubljana) | HTML scrape | ~20 newest jobs |
 | startup.si | HTML scrape (no public API or feed) | Site lists every posting since 2019 on one page, so only the last 90 days are kept (`startupSi(days)` in `src/scrapers/index.ts`) |
 | adecco.com (Slovenia, IT) | Internal JSON API the site's job search uses (no public API) | Filtered by category; add more with `adecco("<CategoryFilter value>", "<label>")` |
+| mojedelo.com (IT, Osrednjeslovenska) | Internal JSON API the site's search uses (no public API) | Takes a search URL copied from the site; add more with `mojeDelo("<label>", "<search URL>")` (don't rename the label later, it's part of the source ID) |
 
 ## Run
 

scrapper/src/scrapers/index.ts

@@ -1,4 +1,5 @@
 import { adecco } from "./adecco";
+import { mojeDelo } from "./mojedelo";
 import { sloTech } from "./slotech";
 import { startupSi } from "./startupsi";
 import type { Scraper } from "./types";
@@ -12,6 +13,10 @@ export const scrapers: Scraper[] = [
   sloTech("Ljubljana"),
   startupSi(),
   adecco("ADCSICAT014", "IT"),
+  mojeDelo(
+    "IT, Osrednjeslovenska",
+    "https://www.mojedelo.com/iskanje-zaposlitve?jobCategoryIds=64f003ff-6d8b-4be0-b58c-4580e4eeeb8a&regionIds=d1dce9b1-9fa4-438b-b582-10d371d442e6",
+  ),
 ];
 
 export type { Scraper, ScrapedJob } from "./types";

scrapper/src/scrapers/mojedelo.ts

@@ -0,0 +1,110 @@
+import { USER_AGENT } from "./http";
+import type { ScrapedJob, Scraper } from "./types";
+
+// MojeDelo is a single-page app with no public API; its search page calls this internal JSON API.
+// The API requires tenant headers, which the site itself reads from a public config file.
+const API = "https://api.mojedelo.com";
+const CONFIG_URL = `${API}/uploaded-files/config/www.mojedelo.com/jb.globals.js`;
+const PAGE_SIZE = 20;
+const MAX_PAGES = 25;
+
+interface Named {
+  id: string;
+  translation?: string;
+  name?: string;
+}
+
+interface MojeDeloJob {
+  id: string;
+  title: string;
+  startDate: string | null;
+  repostingDate: string | null;
+  company: { name?: string } | null;
+  anonymizedCompany: { name?: string } | null;
+  town: Named | null;
+  regions: Named[] | null;
+  jobCategories: Named[] | null;
+  workModes: Named[] | null;
+}
+
+interface SearchResponse {
+  data?: { items?: MojeDeloJob[]; total?: number };
+}
+
+async function tenantHeaders(): Promise<Record<string, string>> {
+  const res = await fetch(CONFIG_URL, { headers: { "User-Agent": USER_AGENT } });
+  if (!res.ok) throw new Error(`GET ${CONFIG_URL} -> HTTP ${res.status}`);
+  const config = await res.text();
+  const get = (key: string) => {
+    const value = new RegExp(`["']?${key}["']?\\s*:\\s*["']([^"']+)["']`).exec(config)?.[1];
+    if (!value) throw new Error(`${key} not found in site config (site changed?)`);
+    return value;
+  };
+  return { tenantId: get("tenantId"), channelId: get("jbChannelId"), languageId: get("defaultLanguageId") };
+}
+
+/** Same slug format as the site's own /oglas/<slug>/<id> links (only the ID matters). */
+function slugify(s: string): string {
+  return s
+    .normalize("NFD")
+    .replace(/[̀-ͯ]/g, "")
+    .replace(/[^A-Za-z0-9]+/g, "-")
+    .replace(/^-|-$/g, "")
+    .toLowerCase();
+}
+
+/**
+ * Scraper for mojedelo.com. Pass a search URL copied from the site (filters in the query string,
+ * e.g. jobCategoryIds / regionIds). `label` is shown in the UI and also forms the stable source ID,
+ * so don't change it later.
+ */
+export function mojeDelo(label: string, searchUrl: string): Scraper {
+  const filters = new URL(searchUrl).searchParams;
+  filters.delete("page");
+
+  return {
+    id: `mojedelo-${slugify(label)}`,
+    name: `MojeDelo (${label})`,
+    async scrape() {
+      const headers = { "User-Agent": USER_AGENT, ...(await tenantHeaders()) };
+      const raw: MojeDeloJob[] = [];
+      for (let page = 0; page < MAX_PAGES; page++) {
+        const params = new URLSearchParams(filters);
+        params.set("pageSize", String(PAGE_SIZE));
+        params.set("startFrom", String(page * PAGE_SIZE));
+        const url = `${API}/job-ads-search?${params}`;
+        const res = await fetch(url, { headers });
+        if (!res.ok) throw new Error(`GET ${url} -> HTTP ${res.status}`);
+        const body = (await res.json()) as SearchResponse;
+        const items = body.data?.items;
+        if (!Array.isArray(items)) throw new Error("unexpected API response (no `data.items` array)");
+        raw.push(...items);
+        if (items.length < PAGE_SIZE || raw.length >= (body.data?.total ?? 0)) break;
+      }
+
+      // The API may silently ignore filters it doesn't recognise, so check results actually match.
+      const wanted = (param: string) => new Set(filters.getAll(param).flatMap((v) => v.split(",")));
+      const categories = wanted("jobCategoryIds");
+      const regions = wanted("regionIds");
+      const matches = (list: Named[] | null, ids: Set<string>) => ids.size === 0 || !!list?.some((x) => ids.has(x.id));
+      const jobs = raw.filter((j) => matches(j.jobCategories, categories) && matches(j.regions, regions));
+      if (raw.length > 0 && jobs.length === 0) {
+        throw new Error("search filters were ignored by the API (query format changed?)");
+      }
+
+      return jobs.map(
+        (j): ScrapedJob => ({
+          externalId: j.id,
+          title: j.title.trim(),
+          url: `https://www.mojedelo.com/oglas/${slugify(j.title)}/${j.id}`,
+          company: j.anonymizedCompany?.name ?? j.company?.name ?? undefined,
+          location: j.town?.translation ?? j.town?.name ?? undefined,
+          tags: [...(j.workModes ?? []), ...(j.jobCategories ?? [])]
+            .map((x) => x.translation?.trim())
+            .filter((t): t is string => !!t),
+          postedAt: j.repostingDate ?? j.startDate ?? undefined,
+        }),
+      );
+    },
+  };
+}