init

providers/trudvsem.mjs

@@ -0,0 +1,144 @@
+// @ts-check
+/** @typedef {import('./_types.js').Provider} Provider */
+
+// "Работа России" (trudvsem.ru) provider — hits the Russian government's
+// public open-data vacancy API (opendata.trudvsem.ru). Aggregates postings
+// submitted to the state employment service; no auth/API key required.
+//
+// Docs: https://trudvsem.ru/opendata/api
+//
+// Portal entry fields (all optional except `provider`):
+//   searchKeywords — `text` query param, e.g. "Frontend Vue" (default: "")
+//   region         — region code for the `region` path filter (default: none — all regions)
+//   pageSize       — `limit` per page, API max is 100 (default: 100)
+//   maxPages       — maximum pages to fetch via `offset` (default: 3)
+
+const API_HOST = 'opendata.trudvsem.ru';
+const API_BASE = 'https://opendata.trudvsem.ru/api/v1/vacancies';
+
+const DEFAULT_PAGE_SIZE = 100;
+const DEFAULT_MAX_PAGES = 3;
+
+/** @param {string} url */
+function assertTrudvsemUrl(url) {
+  let parsed;
+  try {
+    parsed = new URL(url);
+  } catch {
+    throw new Error(`trudvsem: invalid URL: ${url}`);
+  }
+  if (parsed.protocol !== 'https:') throw new Error(`trudvsem: URL must use HTTPS: ${url}`);
+  if (parsed.hostname !== API_HOST)
+    throw new Error(`trudvsem: untrusted hostname "${parsed.hostname}" — must be ${API_HOST}`);
+  return url;
+}
+
+function toEpochMs(value) {
+  if (!value) return undefined;
+  const parsed = Date.parse(value);
+  return Number.isNaN(parsed) ? undefined : parsed;
+}
+
+/**
+ * Parse a single trudvsem API result item into the canonical Job shape.
+ *
+ * The API returns results shaped like:
+ *   { results: { vacancies: [ { vacancy: {
+ *       "job-name": "...", vac_url: "https://trudvsem.ru/vacancy/card/...",
+ *       company: { name: "..." }, region: { name: "..." },
+ *       "creation-date": "2026-07-01", date_modify: "2026-07-02T14:32:05+0300"
+ *   } } ] } }
+ *
+ * @param {any} item — raw `{ vacancy: {...} }` wrapper from the results array
+ * @param {string} fallbackCompany — company name fallback from the portal entry
+ * @returns {{title: string, url: string, company: string, location: string, postedAt: number|undefined}|null}
+ */
+export function parseTrudvsemItem(item, fallbackCompany) {
+  const vacancy = item?.vacancy;
+  if (!vacancy || typeof vacancy !== 'object') return null;
+
+  const title = (vacancy['job-name'] || '').trim();
+  if (!title) return null;
+
+  const rawUrl = (vacancy.vac_url || '').trim();
+  if (!rawUrl) return null;
+  let url;
+  try {
+    const parsed = new URL(rawUrl);
+    if (parsed.protocol !== 'https:') return null;
+    url = parsed.href;
+  } catch {
+    return null;
+  }
+
+  const company = (vacancy.company?.name || fallbackCompany || '').trim();
+  const location = (vacancy.region?.name || '').trim();
+  const postedAt = toEpochMs(vacancy.date_modify || vacancy['creation-date']);
+
+  return { title, url, company, location, ...(postedAt != null ? { postedAt } : {}) };
+}
+
+/**
+ * Build the search URL with query parameters.
+ * @param {{keywords?: string, region?: string, limit?: number, offset?: number}} params
+ * @returns {string}
+ */
+function buildSearchUrl({ keywords, region, limit, offset }) {
+  const base = region ? `${API_BASE}/region/${encodeURIComponent(region)}` : API_BASE;
+  const url = new URL(base);
+  if (keywords) url.searchParams.set('text', keywords);
+  url.searchParams.set('limit', String(limit || DEFAULT_PAGE_SIZE));
+  url.searchParams.set('offset', String(offset || 0));
+  return url.href;
+}
+
+/** @type {Provider} */
+export default {
+  id: 'trudvsem',
+
+  detect(_entry) {
+    // trudvsem is a job board aggregator, not a company ATS.
+    // Use `provider: trudvsem` explicitly in portals.yml.
+    return null;
+  },
+
+  async fetch(entry, ctx) {
+    const keywords = entry.searchKeywords || '';
+    const region = entry.region ? String(entry.region) : '';
+    const pageSize = Number(entry.pageSize) || DEFAULT_PAGE_SIZE;
+    const maxPages = Number(entry.maxPages) || DEFAULT_MAX_PAGES;
+    const fallbackCompany = entry.name || '';
+
+    const allJobs = [];
+
+    for (let page = 0; page < maxPages; page++) {
+      const offset = page * pageSize;
+      const searchUrl = buildSearchUrl({ keywords, region, limit: pageSize, offset });
+      assertTrudvsemUrl(searchUrl);
+
+      let json;
+      try {
+        json = /** @type {any} */ (await ctx.fetchJson(searchUrl, { redirect: 'error' }));
+      } catch (err) {
+        if (page === 0) throw err;
+        console.error(`trudvsem: page ${page + 1} fetch failed — ${err.message}`);
+        break;
+      }
+
+      const vacancies = Array.isArray(json?.results?.vacancies) ? json.results.vacancies : [];
+      if (vacancies.length === 0) break;
+
+      for (const item of vacancies) {
+        const job = parseTrudvsemItem(item, fallbackCompany);
+        if (job) allJobs.push(job);
+      }
+
+      if (vacancies.length < pageSize) break;
+
+      // Be a polite citizen of a free government API
+      await new Promise(resolve => setTimeout(resolve, 200));
+    }
+
+    return allJobs;
+  },
+};