feat: gnula_grabber 2-part pipeline (crawl ES + CDP stream download)

crawl.py

@@ -0,0 +1,100 @@
+#!/usr/bin/env python3
+"""PARTE 1 — Crawler gnula: detecta pelis en ESPAÑOL (bandera es.png) y las cataloga.
+
+Usa FlareSolverr (Chrome NordVPN+CF, headless) para fetchear las páginas listado
+(gnula está ISP-bloqueado + Cloudflare). Parsea los <article class="item movies">,
+filtra los que tienen flag es.png (Español/castellano), dedup vs catálogo + Radarr,
+y guarda en SQLite. NO descarga nada (eso es Parte 2: grab_stream.py).
+
+Uso: python gnula_crawl.py [base_path] [max_pages]
+  base_path: ej 'peliculas/estrenos' (default), 'release/2025', 'peliculas/accion'...
+"""
+import json, re, sqlite3, sys, time, urllib.request
+
+FLARE = "http://localhost:8191/v1"
+SITE = "https://www.gnularetro.cc"
+DB = "/home/lucas/.config/popelis/gnula_catalog.db"
+RADARR = ("http://localhost:7878", "63fb51c8c95746e2a327740baac02f5e")
+
+def flare_get(url, timeout=150, retries=2):
+  for _a in range(retries+1):
+    try:
+      return _flare_get(url, timeout)
+    except Exception as e:
+      if _a==retries: raise
+      time.sleep(5)
+def _flare_get(url, timeout=150):
+    body = json.dumps({"cmd": "request.get", "url": url, "maxTimeout": 60000}).encode()
+    req = urllib.request.Request(FLARE, data=body, headers={"Content-Type": "application/json"})
+    resp = json.load(urllib.request.urlopen(req, timeout=timeout))
+    sol = resp.get("solution", {})
+    return sol.get("status"), sol.get("response", "")
+
+def parse_cards(html):
+    out = []
+    for blk in re.findall(r'<article[^>]*class="[^"]*item movies[^"]*".*?</article>', html, re.S):
+        href = re.search(r'href="([^"]*ver-pelicula[^"]*)"', blk)
+        title = re.search(r'<h3[^>]*>.*?<a[^>]*>([^<]+)', blk, re.S) or re.search(r'title="([^"]+)"', blk)
+        flags = re.findall(r'/flags/(\w+\.png)', blk)
+        yr = re.search(r'<span>(\d{4})</span>', blk) or re.search(r'\b(20\d{2})\b', blk)
+        if href and title:
+            out.append({
+                "href": href.group(1),
+                "title": title.group(1).strip(),
+                "year": int(yr.group(1)) if yr else None,
+                "flags": sorted(set(flags)),
+                "lang_es": 1 if any("es.png" == f for f in flags) else 0,
+            })
+    return out
+
+def db_init():
+    import os; os.makedirs("/home/lucas/.config/popelis", exist_ok=True)
+    c = sqlite3.connect(DB)
+    c.execute("""CREATE TABLE IF NOT EXISTS movies(
+        href TEXT PRIMARY KEY, title TEXT, year INTEGER, flags TEXT,
+        lang_es INTEGER, status TEXT DEFAULT 'pending', in_library INTEGER DEFAULT 0,
+        detected_at TEXT, downloaded_at TEXT)""")
+    c.commit(); return c
+
+def radarr_titles():
+    try:
+        req = urllib.request.Request(f"{RADARR[0]}/api/v3/movie", headers={"X-Api-Key": RADARR[1]})
+        return {m["title"].lower() for m in json.load(urllib.request.urlopen(req, timeout=20))}
+    except Exception:
+        return set()
+
+def main():
+    base = sys.argv[1] if len(sys.argv) > 1 else "peliculas/estrenos"
+    max_pages = int(sys.argv[2]) if len(sys.argv) > 2 else 8
+    c = db_init(); have = radarr_titles()
+    now = time.strftime("%Y-%m-%dT%H:%M:%S")
+    total_new = 0; total_es = 0
+    for p in range(1, max_pages + 1):
+        url = f"{SITE}/{base}/" if p == 1 else f"{SITE}/{base}/page/{p}/"
+        st, html = flare_get(url)
+        if st != 200:
+            print(f"[crawl] page {p}: HTTP {st} -> stop"); break
+        cards = parse_cards(html)
+        if not cards:
+            print(f"[crawl] page {p}: 0 cards -> stop"); break
+        es = [x for x in cards if x["lang_es"]]
+        for x in es:
+            inlib = 1 if any(t in x["title"].lower() or x["title"].lower() in t for t in have) else 0
+            c.execute("""INSERT INTO movies(href,title,year,flags,lang_es,status,in_library,detected_at)
+                VALUES(?,?,?,?,?,?,?,?)
+                ON CONFLICT(href) DO UPDATE SET flags=excluded.flags, in_library=excluded.in_library""",
+                (x["href"], x["title"], x["year"], ",".join(x["flags"]), 1,
+                 "have" if inlib else "pending", inlib, now))
+        c.commit()
+        total_es += len(es); total_new += len(cards)
+        print(f"[crawl] page {p}: {len(cards)} pelis, {len(es)} en español")
+        time.sleep(1)
+    # resumen
+    cur = c.execute("SELECT COUNT(*),SUM(in_library),SUM(status='pending') FROM movies WHERE lang_es=1")
+    tot, inlib, pend = cur.fetchone()
+    print(json.dumps({"crawled_pages_upto": max_pages, "es_seen": total_es,
+                      "catalog_total_es": tot, "in_library": inlib or 0,
+                      "pending_download": pend or 0, "db": DB}))
+
+if __name__ == "__main__":
+    main()