From ad2fa2b7753aedc814f5de602aa1e4ef1f33584a Mon Sep 17 00:00:00 2001
From: Nikola Petrov <nikola@petrovv.com>
Date: Mon, 12 Oct 2026 21:29:24 +0200
Subject: [PATCH] Add Optius scraper via its official RSS feed

Optius publishes an RSS feed with region/field filters, so no HTML
scraping is needed. Also move the duplicated slug helper into http.ts
(output unchanged, so existing source IDs and URLs are unaffected).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---
 scrapper/README.md                |  1 +
 scrapper/src/scrapers/adecco.ts   |  9 ++-----
 scrapper/src/scrapers/http.ts     |  9 +++++++
 scrapper/src/scrapers/index.ts    |  2 ++
 scrapper/src/scrapers/mojedelo.ts | 12 +--------
 scrapper/src/scrapers/optius.ts   | 42 +++++++++++++++++++++++++++++++
 6 files changed, 57 insertions(+), 18 deletions(-)
 create mode 100644 scrapper/src/scrapers/optius.ts

diff --git a/scrapper/README.md b/scrapper/README.md
index ca175f1..841060c 100644
--- a/scrapper/README.md
+++ b/scrapper/README.md
@@ -12,6 +12,7 @@ No runtime dependencies. It uses Bun's built-in SQLite, HTTP server and HTMLRewr
 | 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) |
+| optius.com (IT, Osrednjeslovenska) | Official RSS feed | Build a feed URL at https://www.optius.com/rss/ (its IDs differ from the search page's); add more with `optius("<label>", "<feed URL>")` (same label rule as MojeDelo) |
 
 ## Run
 
diff --git a/scrapper/src/scrapers/adecco.ts b/scrapper/src/scrapers/adecco.ts
index 547172f..2f90114 100644
--- a/scrapper/src/scrapers/adecco.ts
+++ b/scrapper/src/scrapers/adecco.ts
@@ -1,4 +1,4 @@
-import { USER_AGENT } from "./http";
+import { slugify, USER_AGENT } from "./http";
 import type { ScrapedJob, Scraper } from "./types";
 
 // Adecco has no public API, but its job search page loads results from this internal JSON endpoint
@@ -28,12 +28,7 @@ interface AdeccoResponse {
 
 /** The site builds "<title>-<location>/<jobid>" URLs; only the job ID matters, the slug is cosmetic. */
 function jobUrl(j: AdeccoJob): string {
-  const slug = `${j.jobTitle}-${j.jobLocation ?? ""}`
-    .normalize("NFD")
-    .replace(/[̀-ͯ]/g, "")
-    .replace(/[^A-Za-z0-9]+/g, "-")
-    .replace(/^-|-$/g, "")
-    .toLowerCase();
+  const slug = slugify(`${j.jobTitle}-${j.jobLocation ?? ""}`);
   return `${JOB_PAGE}/${slug}/${j.jobId.toLowerCase()}`;
 }
 
diff --git a/scrapper/src/scrapers/http.ts b/scrapper/src/scrapers/http.ts
index 5456fa1..c3ed5ff 100644
--- a/scrapper/src/scrapers/http.ts
+++ b/scrapper/src/scrapers/http.ts
@@ -45,3 +45,12 @@ export function decodeEntities(s: string): string {
 }
 
 export const clean = (s: string) => decodeEntities(s).replace(/\s+/g, " ").trim();
+
+/** ASCII URL/ID slug: "Inženir (m/ž)" -> "inzenir-m-z". */
+export const slugify = (s: string) =>
+  s
+    .normalize("NFD")
+    .replace(/[̀-ͯ]/g, "")
+    .replace(/[^A-Za-z0-9]+/g, "-")
+    .replace(/^-|-$/g, "")
+    .toLowerCase();
diff --git a/scrapper/src/scrapers/index.ts b/scrapper/src/scrapers/index.ts
index eed4037..64438d3 100644
--- a/scrapper/src/scrapers/index.ts
+++ b/scrapper/src/scrapers/index.ts
@@ -1,5 +1,6 @@
 import { adecco } from "./adecco";
 import { mojeDelo } from "./mojedelo";
+import { optius } from "./optius";
 import { sloTech } from "./slotech";
 import { startupSi } from "./startupsi";
 import type { Scraper } from "./types";
@@ -17,6 +18,7 @@ export const scrapers: Scraper[] = [
     "IT, Osrednjeslovenska",
     "https://www.mojedelo.com/iskanje-zaposlitve?jobCategoryIds=64f003ff-6d8b-4be0-b58c-4580e4eeeb8a&regionIds=d1dce9b1-9fa4-438b-b582-10d371d442e6",
   ),
+  optius("IT, Osrednjeslovenska", "https://www.optius.com/rss/?regija=7&podrocje=7"),
 ];
 
 export type { Scraper, ScrapedJob } from "./types";
diff --git a/scrapper/src/scrapers/mojedelo.ts b/scrapper/src/scrapers/mojedelo.ts
index 626ded0..205ac23 100644
--- a/scrapper/src/scrapers/mojedelo.ts
+++ b/scrapper/src/scrapers/mojedelo.ts
@@ -1,4 +1,4 @@
-import { USER_AGENT } from "./http";
+import { slugify, 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.
@@ -43,16 +43,6 @@ async function tenantHeaders(): Promise<Record<string, string>> {
   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(/^-|-$/g, "")
-    .toLowerCase();
-}
-
 /**
  * Scraper for mojedelo.com. Pass a search URL copied from the site (filters in the query string,
  * e.g. jobCategoryIds / regionIds). `label` is shown in the UI and also forms the stable source ID,
diff --git a/scrapper/src/scrapers/optius.ts b/scrapper/src/scrapers/optius.ts
new file mode 100644
index 0000000..260be68
--- /dev/null
+++ b/scrapper/src/scrapers/optius.ts
@@ -0,0 +1,42 @@
+import { clean, fetchHtml, slugify } from "./http";
+import type { ScrapedJob, Scraper } from "./types";
+
+// Optius publishes an official RSS feed with region/field filters, so no HTML scraping is needed.
+// Build a feed URL at https://www.optius.com/rss/ (its IDs differ from the search page's IDs).
+
+const tag = (xml: string, name: string) => new RegExp(`<${name}[^>]*>([\\s\\S]*?)</${name}>`).exec(xml)?.[1];
+const text = (s: string | undefined) => clean((s ?? "").replace(/^<!\[CDATA\[([\s\S]*)\]\]>$/, "$1"));
+
+export function parseOptiusFeed(xml: string): ScrapedJob[] {
+  return [...xml.matchAll(/<item>([\s\S]*?)<\/item>/g)]
+    .map((m): ScrapedJob | null => {
+      const item = m[1]!;
+      const url = text(tag(item, "link") ?? tag(item, "guid"));
+      const id = /-(\d+)-\1\/?$/.exec(url)?.[1] ?? /-(\d+)\/?$/.exec(url)?.[1];
+      const title = text(tag(item, "title"));
+      if (!id || !title) return null;
+      const pubDate = tag(item, "pubDate");
+      return {
+        externalId: id,
+        title,
+        url,
+        company: text(tag(item, "employer")) || undefined,
+        tags: [...item.matchAll(/<field(?:\s[^>]*)?>([\s\S]*?)<\/field>/g)].map((f) => text(f[1])).filter(Boolean),
+        postedAt: pubDate ? new Date(pubDate).toISOString() : undefined,
+      };
+    })
+    .filter((j): j is ScrapedJob => j !== null);
+}
+
+/** Scraper for an optius.com RSS feed, e.g. "https://www.optius.com/rss/?regija=7&podrocje=7". */
+export function optius(label: string, feedUrl: string): Scraper {
+  return {
+    id: `optius-${label.normalize("NFD").replace(/[̀-ͯ]/g, "").replace(/[^A-Za-z0-9]+/g, "-").replace(/^-|-$/g, "").toLowerCase()}`,
+    name: `Optius (${label})`,
+    async scrape() {
+      const xml = await fetchHtml(feedUrl);
+      if (!xml.includes("<rss")) throw new Error("response is not an RSS feed (site changed?)");
+      return parseOptiusFeed(xml);
+    },
+  };
+}