scraper_afc.py

from selenium import webdriver
from selenium.webdriver.common.by import By
from selenium.webdriver.chrome.options import Options
import pandas as pd
import time
import requests
import fitz  

# Configurar navegador
options = Options()
# options.add_argument("--headless")  # Opcional, para correr sin ventana
options.add_argument("--window-size=1920,1080")
driver = webdriver.Chrome(options=options)

# Acceder a la página
driver.get("https://www.previred.com/indicadores-previsionales/")
time.sleep(3)

# Expandir todos los años disponibles
headers = driver.find_elements(By.CLASS_NAME, "accordion__header")
print(f"Encontrados {len(headers)} años.")

for header in headers:
    try:
        driver.execute_script("arguments[0].scrollIntoView(true);", header)
        time.sleep(1)
        header.click()
        time.sleep(2)
    except Exception as e:
        print("Error al hacer clic en el año:", e)

# Obtener todos los links a PDF visibles tras expandir
links = driver.find_elements(By.XPATH, "//a[contains(@href, '.pdf')]")

data = []

for link in links:
    url = link.get_attribute("href")
    texto = link.text.strip()
    print(f"Descargando: {texto} | URL: {url}")
    
    try:
        response = requests.get(url)
        response.raise_for_status()
        doc = fitz.open(stream=response.content, filetype="pdf")
        
        pdf_text = ""
        for page in doc:
            pdf_text += page.get_text()
        doc.close()
        
        data.append({
            "nombre": texto,
            "url": url,
            "contenido": pdf_text
        })
    except Exception as e:
        print(f"Error con {url}: {e}")

driver.quit()

# Convertir a DataFrame y guardar
df = pd.DataFrame(data)
df.to_csv("previred_indicadores.csv", index=False)

print("Datos guardados en previred_indicadores.csv")