Sportski objekti: API + Leaflet map page + address enrichment

DB: pgz_sport.sportski_objekti (103 objekti, 103 s geo, 60 s adresom, 31 tip) API: - /api/v2/sportski-objekti (filter: tip, grad, sport, q) - /api/v2/sportski-objekti/meta (tipovi, gradovi, sportovi, ukupno) Frontend: - /static/objekti.html — Leaflet (OpenStreetMap) interactive map - 3 dropdown filter (tip, grad, sport) + search - Side panel s listom + map markers s ikonama (🏟️⚽🏊⛵🎿🎳⛸️🎯🥌🏃) - Popup: naziv, tip, kapacitet, adresa, upravitelj, izgradeno, sportovi, web link, Google Maps link - /objekti, /sport/objekti, /sport/api/v2/sportski-objekti routes Sidebar app.html: +Sportski objekti link Background: scripts/objekti_enrich_address.py (Nominatim reverse-geocode 60 objekata bez adrese)
2026-05-05 18:35:04 +02:00
parent 6e5ada8517
commit ae9c4e2bfd
15 changed files with 767 additions and 4 deletions
@@ -0,0 +1,64 @@
+#!/usr/bin/env python3
+"""UNIRI akademski repozitorij + znanstveni radovi."""
+import sys, json, time
+sys.path.insert(0, "/opt/pgz-sport/scrapers/harvesters")
+from _common import (fetch, extract_text, extract_title, chunk_text,
+                      upsert_facts, find_internal_links, DSN)
+from urllib.parse import urlparse
+import psycopg2
+
+ACADEMIC = {
+    "uniri_repozitorij":   ["https://repozitorij.uniri.hr/"],
+    "portal_znanstveni":   ["https://portal.uniri.hr/"],
+    "hrčak_uniri":        ["https://hrcak.srce.hr/"],
+    "pfri_radovi":        ["https://repository.pfri.uniri.hr/"],
+    "medri_radovi":       ["https://medri.uniri.hr/znanstveni-radovi/"],
+    "tfr_radovi":         ["https://www.riteh.uniri.hr/"],
+    "ffri_radovi":        ["https://www.ffri.uniri.hr/znanstveni-radovi/"],
+}
+
+
+def crawl(name, urls, max_pages=15):
+    conn = psycopg2.connect(DSN); conn.autocommit = True
+    visited = set(); queue = list(urls); facts = 0
+    while queue and len(visited) < max_pages:
+        url = queue.pop(0)
+        if url in visited: continue
+        visited.add(url)
+        html, status = fetch(url, timeout=20)
+        if not html or status != 200: continue
+        title = extract_title(html); text = extract_text(html)
+        if not text or len(text) < 300: continue
+        ff = []
+        if title and len(title) > 15:
+            ff.append({"fact": f"[Academic] {name} - {title}", "url": url, "title": title})
+        for c in chunk_text(text, 900):
+            if len(c) > 150:
+                ff.append({"fact": c, "url": url, "title": title})
+        facts += upsert_facts(conn, ff, source_name=name,
+                               category="akademski_pgz", confidence=0.90)
+        base = urlparse(url).hostname
+        for link in find_internal_links(html, url):
+            if link not in visited and (urlparse(link).hostname or "") == base and len(queue) < 40:
+                queue.append(link)
+        time.sleep(0.7)
+    conn.close()
+    return {"name": name, "visited": len(visited), "facts": facts}
+
+
+def main():
+    results = []
+    for name, urls in ACADEMIC.items():
+        try:
+            r = crawl(name, urls, max_pages=12)
+            print(f"  {name:25} {r['visited']:>3}p  {r['facts']:>5}f")
+            results.append(r)
+        except Exception as e:
+            print(f"  {name:25} FAIL: {str(e)[:60]}")
+    total = sum(r.get("facts", 0) for r in results)
+    print(f"=== TOTAL: {total} ===")
+    print(json.dumps({"academic_count": len(results), "total_facts": total}))
+
+
+if __name__ == "__main__":
+    main()