Bug hunt V7:

DB: - Aggressive je_klub=false flag for programs/treninzi/totals (>100K€ no klub_id) - 53 ne-klubovi flagged false (RSS Rijeka ukupni, Stručni rad, Potpora loptačkim, etc) Frontend (sport2.html): - Panel back button (← Natrag) + history stack - window._panelHistory + pushPanelState + panelBack functions - closePanel resets history
2026-05-05 14:56:53 +02:00
parent 1e611d59f1
commit 007825acee
214 changed files with 15117 additions and 565 deletions
@@ -0,0 +1,70 @@
+#!/usr/bin/env python3
+"""Servisne usluge PGZ — komunalije, transport."""
+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
+
+SERVIS = {
+    "kd_komunalc":        ["https://www.kd-komunalac.hr/"],
+    "kd_kozala":          ["https://www.kd-kozala.hr/"],
+    "rijekapromet":       ["https://www.rijekapromet.hr/"],
+    "vodovod_pgz":        ["https://www.kdvik-rijeka.hr/"],
+    "rgcc_plin":          ["https://www.energo.hr/"],
+    "hep_rijeka":         ["https://www.hep.hr/elektrodalmacija/"],
+    "rijeka_parking":     ["https://www.rijekaplus.hr/"],
+    "ana_aerodrom":       ["https://rijeka-airport.hr/"],
+    "rijeka_busplus":     ["https://www.autotrans.hr/"],
+    "jadrolinija":        ["https://www.jadrolinija.hr/"],
+    "kbc_rijeka":         ["https://www.kbc-rijeka.hr/"],
+    "thalassotherapia":   ["https://thalassotherapia-opatija.hr/"],
+    "klinika_opatija":    ["https://www.opatija.medicus.hr/"],
+}
+
+
+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=15)
+        if not html or status != 200: continue
+        title = extract_title(html); text = extract_text(html)
+        if not text or len(text) < 200: continue
+        ff = []
+        if title and len(title) > 8:
+            ff.append({"fact": f"{name} - {title}", "url": url, "title": title})
+        for c in chunk_text(text, 800):
+            if len(c) > 100:
+                ff.append({"fact": c, "url": url, "title": title})
+        facts += upsert_facts(conn, ff, source_name=name,
+                               category="servisne_pgz", confidence=0.86)
+        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(0.5)
+    conn.close()
+    return {"name": name, "visited": len(visited), "facts": facts}
+
+
+def main():
+    results = []
+    for name, urls in SERVIS.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({"servis_count": len(results), "total_facts": total}))
+
+
+if __name__ == "__main__":
+    main()