feat(bizin): scraper final com bypass Cloudflare + monitor de auto-reinício

- bizin_scraper_final.py: scraper híbrido curl_cffi + undetected-chromedriver com suporte a distritos e categorias, escrita segura (fsync) e enriquecimento externo - monitor_scraper.sh: watchdog que reinicia o processo automaticamente em crash - IMPLEMENTADO.md + README.md: actualizados para reflectir estado Abril 2026 - GEMINI.md: instruções técnicas de automação - test_curl.py, test_curl_clean.py, test_playwright.py: scripts de teste/diagnóstico Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-28 17:16:48 +01:00
parent 865a9459a6
commit 8e0dbbeca0
8 changed files with 337 additions and 220 deletions
@@ -0,0 +1,43 @@
+
+import asyncio
+from playwright.async_api import async_playwright
+from playwright_stealth import Stealth
+
+async def test_bizin():
+    async with async_playwright() as p:
+        # Tentar usar o Chrome do sistema
+        try:
+            browser = await p.chromium.launch(headless=True, channel="chrome")
+        except:
+            browser = await p.chromium.launch(headless=True)
+
+        context = await browser.new_context(
+            user_agent="Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36"
+        )
+        await Stealth().apply_stealth_async(context)
+        page = await context.new_page()
+        
+        print("Acedendo a https://pt.bizin.eu/por/ ...")
+        try:
+            await page.goto("https://pt.bizin.eu/por/", wait_until="domcontentloaded", timeout=30000)
+        except Exception as e:
+            print(f"Timeout ou erro na carga inicial: {e}")
+        
+        # Esperar um pouco para o desafio resolver
+        print("Aguardando 45 segundos por possíveis desafios...")
+        await asyncio.sleep(45)
+        
+        content = await page.content()
+        if "Just a moment..." in content or "Um momento…" in content:
+            print("Bloqueado pelo Cloudflare.")
+        else:
+            print("Sucesso! Página carregada.")
+            print(f"Título: {await page.title()}")
+            # Salvar sucesso para conferir
+            with open("logs/success_playwright.html", "w", encoding="utf-8") as f:
+                f.write(content)
+        
+        await browser.close()
+
+if __name__ == "__main__":
+    asyncio.run(test_bizin())