generador_logs.py

import random
import datetime
import argparse

# --- Configuración de los Mensajes de Log ---
LOG_MESSAGES = {
    "INFO": [
        "User {user_id} logged in successfully from IP {ip_address}.",
        "Request received for endpoint /api/v1/data.",
        "Processing batch job #{job_id} started.",
        "Data synchronization with external service completed.",
        "Configuration reloaded successfully.",
        "Cache cleared for key: '{cache_key}'.",
    ],
    "WARNING": [
        "API response time is high: {response_time}ms for endpoint /api/v1/status.",
        "Disk space is running low on /dev/sda1 ({disk_space}% used).",
        "A deprecated function 'old_function()' was called.",
        "Could not resolve hostname 'temp-service.local', using fallback.",
        "User session for {user_id} is about to expire.",
    ],
    "ERROR": [
        "Database connection failed: Timeout expired.",
        "NullPointerException in module 'PaymentProcessor' at line {line_number}.",
        "Failed to write to file /var/logs/app.log: Permission denied.",
        "Critical component 'MessageQueue' is not responding.",
        "Authentication failed for user '{user_name}'. Invalid credentials provided.",
    ],
    "DEBUG": [
        "Executing SQL query: SELECT * FROM users WHERE id={user_id}.",
        "Variable 'x' has value: {random_value}.",
        "Entering function 'calculate_metrics()'.",
        "HTTP request headers: {header_info}.",
    ]
}

def generate_random_ip():
    """Genera una dirección IP aleatoria."""
    return ".".join(str(random.randint(1, 255)) for _ in range(4))

def generate_varied_log_line(current_time):
    """Genera una única línea de log variada (comportamiento original)."""
    log_level = random.choices(["INFO", "DEBUG", "WARNING", "ERROR"], [0.65, 0.20, 0.10, 0.05], k=1)[0]
    message_template = random.choice(LOG_MESSAGES[log_level])
    message = message_template.format(
        user_id=random.randint(100, 999), ip_address=generate_random_ip(),
        job_id=random.randint(1000, 9999), cache_key=f"user_data_{random.randint(1, 100)}",
        response_time=random.randint(200, 2000), disk_space=random.randint(85, 99),
        line_number=random.randint(50, 500), user_name=random.choice(["admin", "guest"]),
        random_value=round(random.uniform(0, 1), 4), header_info="{...}"
    )
    timestamp = current_time.strftime('%Y-%m-%d %H:%M:%S') + f",{current_time.microsecond // 1000:03d}"
    return f"[{timestamp}] {log_level}: {message}"

def generate_uniform_log_line(current_time):
    """Genera una línea de log que siempre sigue la misma plantilla."""
    # Usamos siempre la misma plantilla para que el modelo de ML no encuentre anomalías.
    template = "INFO: Data processing task {task_id} completed in {duration} seconds."
    message = template.format(
        task_id=random.randint(10000, 99999),
        duration=round(random.uniform(0.5, 15.2), 2)
    )
    timestamp = current_time.strftime('%Y-%m-%d %H:%M:%S') + f",{current_time.microsecond // 1000:03d}"
    return f"[{timestamp}] {message}"


def create_log_file(filename, num_lines, mode):
    """Crea un archivo de log con un número específico de líneas y en un modo determinado."""
    print(f"Generando {num_lines} líneas de log en modo '{mode}' para el archivo '{filename}'...")
    
    current_time = datetime.datetime.now() - datetime.timedelta(hours=3)

    try:
        with open(filename, 'w', encoding='utf-8') as f:
            for i in range(num_lines):
                time_increment = random.uniform(0.05, 3.0)
                current_time += datetime.timedelta(seconds=time_increment)
                
                if mode == 'uniform':
                    log_line = generate_uniform_log_line(current_time)
                else:  # El modo por defecto es 'varied'
                    log_line = generate_varied_log_line(current_time)
                
                f.write(log_line + "\n")
                
                if (i + 1) % 1000 == 0:
                    print(f"  ... {i + 1} líneas generadas")

        print(f"¡Éxito! Archivo '{filename}' creado con {num_lines} líneas.")

    except IOError as e:
        print(f"Error al escribir en el archivo '{filename}': {e}")


if __name__ == "__main__":
    parser = argparse.ArgumentParser(
        description="Generador de Archivos de Log de Muestra.",
        formatter_class=argparse.RawTextHelpFormatter
    )
    
    parser.add_argument(
        "num_lines", type=int, help="El número de líneas de log a generar."
    )
    
    parser.add_argument(
        "-o", "--output", type=str, default="sample.log",
        help="El nombre del archivo de log de salida.\n(default: sample.log)"
    )

    # --- NUEVO ARGUMENTO PARA ELEGIR EL MODO ---
    parser.add_argument(
        "--mode", type=str, choices=['varied', 'uniform'], default='varied',
        help="Elige el tipo de log a generar:\n"
             "'varied': mezcla realista de logs (comportamiento por defecto).\n"
             "'uniform': logs muy repetitivos para no generar anomalías de ML."
    )

    args = parser.parse_args()
    
    create_log_file(args.output, args.num_lines, args.mode)