#!/usr/bin/env python3
# sukob_sport_scraper.py — Povjerenstvo za sukob interesa, filter za sport funkcionere
import os, time, hashlib, logging, re, json
from urllib.parse import urljoin
import urllib.request
import psycopg2
from html import unescape

logging.basicConfig(level=logging.INFO, format='%(asctime)s [sukob] %(message)s')
log = logging.getLogger("sukob")
DSN = "host=10.10.0.2 port=6432 dbname=rinet_v3 user=rinet password=R1net2026!SecureDB#v7"
UA = "Mozilla/5.0 (Ri.NET Bot 1.0)"

def fetch(url):
    try:
        req = urllib.request.Request(url, headers={"User-Agent": UA})
        with urllib.request.urlopen(req, timeout=20) as r:
            return r.read().decode('utf-8', errors='replace'), r.status
    except: return None, 0

def harvest():
    conn = psycopg2.connect(DSN); conn.autocommit = True
    cur = conn.cursor()
    
    # Get OIB-ovi predsjednika i tajnika klubova/saveza za cross-check
    cur.execute("""
        SELECT DISTINCT ime FROM (
            SELECT predsjednik AS ime FROM pgz_sport.klubovi WHERE predsjednik IS NOT NULL AND length(predsjednik)>5
            UNION
            SELECT tajnik FROM pgz_sport.klubovi WHERE tajnik IS NOT NULL AND length(tajnik)>5
            UNION
            SELECT predsjednik FROM pgz_sport.savezi WHERE predsjednik IS NOT NULL AND length(predsjednik)>5
        ) t LIMIT 100
    """)
    sport_imena = [r[0].strip() for r in cur.fetchall() if r[0]]
    log.info(f"Sport imena za cross-check: {len(sport_imena)}")
    
    # Search sukobinteresa.hr za neka imena
    facts = 0
    for ime in sport_imena[:20]:
        # Pretraga po imenu
        from urllib.parse import quote_plus
        url = f"https://www.sukobinteresa.hr/hr/imovinsko-stanje/imovinske-kartice?search={quote_plus(ime)}"
        html, status = fetch(url)
        time.sleep(2)
        if not html or status != 200: continue
        
        # Ako ima rezultata, izvuci
        if ime.lower() in html.lower():
            # cleanup html
            text = re.sub(r'<script[^>]*>.*?</script>', '', html, flags=re.S|re.I)
            text = re.sub(r'<[^>]+>', ' ', text)
            text = re.sub(r'\s+', ' ', unescape(text)).strip()
            # Find context around ime
            idx = text.lower().find(ime.lower())
            if idx > 0:
                ctx = text[max(0, idx-300):idx+500]
                fact = f"FORENSIČKI FLAG: {ime} se nalazi u registru imovinskih kartica Povjerenstva za sukob interesa. Kontekst: {ctx[:600]}"
                fh = hashlib.sha256(f"sukob:{ime}".encode()).hexdigest()[:32]
                try:
                    cur.execute("""INSERT INTO dabi.knowledge
                        (fact, category, source, source_refs, confidence, data_hash, created_at)
                        VALUES (%s,'sukob_interesa_sport','sukob_scraper',%s::jsonb,0.90,%s,now())
                        ON CONFLICT (data_hash) DO NOTHING""",
                        (fact[:2000], json.dumps([{"url":url, "ime":ime}]), fh))
                    facts += cur.rowcount
                    log.info(f"✓ Match: {ime}")
                except: pass
    
    log.info(f"FINAL: {facts} sukob facts")
    cur.close(); conn.close()

if __name__ == "__main__":
    harvest()