gh-christianlouis-docuelevate/app/tasks/upload_to_ftp.py

#!/usr/bin/env python3

# Security Warning: FTP is an insecure protocol. FTPS (FTP_TLS) is strongly recommended.
# This module attempts to use FTPS by default and falls back to plaintext FTP only if configured.
import ftplib  # nosec B402 - FTP usage is intentional for legacy server support
import logging
import os

from app.celery_app import celery
from app.config import settings
from app.tasks.retry_config import UploadTaskWithRetry
from app.utils import log_task_progress

logger = logging.getLogger(__name__)


@celery.task(base=UploadTaskWithRetry, bind=True)
def upload_to_ftp(self, file_path: str, file_id: int = None, folder_override: str = None):
    """
    Uploads a file to an FTP server in the configured folder.

    Security Note: This function prefers FTPS (FTP with TLS) for secure connections.
    Plaintext FTP is only used if FTPS fails and ftp_allow_plaintext=True (default).
    For security-critical environments, set ftp_allow_plaintext=False and ftp_use_tls=True.

    Args:
        file_path: Path to the file to upload
        file_id: Optional file ID to associate with logs
    """
    task_id = self.request.id
    logger.info(f"[{task_id}] Starting FTP upload: {file_path}")
    log_task_progress(
        task_id, "upload_to_ftp", "in_progress", f"Uploading to FTP: {os.path.basename(file_path)}", file_id=file_id
    )

    if not os.path.exists(file_path):
        error_msg = f"File not found: {file_path}"
        logger.error(f"[{task_id}] {error_msg}")
        log_task_progress(task_id, "upload_to_ftp", "failure", error_msg, file_id=file_id)
        raise FileNotFoundError(error_msg)

    # Extract filename
    filename = os.path.basename(file_path)

    # Check if FTP settings are configured
    if not settings.ftp_host:
        error_msg = "FTP host is not configured"
        logger.error(f"[{task_id}] {error_msg}")
        log_task_progress(task_id, "upload_to_ftp", "failure", error_msg, file_id=file_id)
        raise ValueError(error_msg)

    try:
        # First attempt FTPS (FTP with TLS)
        use_tls = getattr(settings, "ftp_use_tls", True)  # Default to try TLS
        allow_plaintext = getattr(settings, "ftp_allow_plaintext", True)  # Default to allow plaintext fallback
        used_tls = False  # Track whether we successfully used TLS

        if use_tls:
            try:
                logger.info(f"Attempting FTPS connection to {settings.ftp_host}")
                ftp = ftplib.FTP_TLS()  # noqa: S321
                ftp.connect(host=settings.ftp_host, port=settings.ftp_port or 21)

                # Login with credentials
                ftp.login(user=settings.ftp_username, passwd=settings.ftp_password)

                # Enable data protection - encrypt the data channel
                ftp.prot_p()
                logger.info("Successfully established FTPS connection with TLS")
                used_tls = True
            except Exception as e:
                if not allow_plaintext:
                    error_msg = f"FTPS connection failed and plaintext FTP is forbidden: {str(e)}"
                    logger.error(error_msg)
                    raise Exception(error_msg)
                else:
                    logger.warning(f"FTPS connection failed, falling back to regular FTP: {str(e)}")
                    # Fall back to regular FTP - only if explicitly allowed by configuration
                    ftp = ftplib.FTP()  # nosec B321 - Fallback to FTP intentional when configured  # noqa: S321
                    ftp.connect(host=settings.ftp_host, port=settings.ftp_port or 21)

                    # Login with credentials
                    ftp.login(user=settings.ftp_username, passwd=settings.ftp_password)
        else:
            # Check if plaintext is allowed when TLS is explicitly disabled
            if not allow_plaintext:
                error_msg = "Plaintext FTP is forbidden by configuration"
                logger.error(error_msg)
                raise Exception(error_msg)

            # Directly use regular FTP if TLS is explicitly disabled
            logger.warning("Using plaintext FTP - connection is NOT encrypted!")
            ftp = ftplib.FTP()  # nosec B321 - Plaintext FTP intentional when explicitly configured  # noqa: S321
            ftp.connect(host=settings.ftp_host, port=settings.ftp_port or 21)

            # Login with credentials
            ftp.login(user=settings.ftp_username, passwd=settings.ftp_password)

        # Change to target directory if specified
        ftp_folder_setting = folder_override if folder_override is not None else settings.ftp_folder
        if ftp_folder_setting:
            try:
                # Try to navigate to the directory, create if it doesn't exist
                ftp_folder = ftp_folder_setting
                # Remove leading slash if present
                if ftp_folder.startswith("/"):
                    ftp_folder = ftp_folder[1:]

                # Try to change to the directory
                try:
                    ftp.cwd(ftp_folder)
                except ftplib.error_perm:
                    # Create directory structure if it doesn't exist
                    folders = ftp_folder.split("/")
                    current_dir = ""
                    for folder in folders:
                        if folder:
                            current_dir += f"/{folder}"
                            try:
                                ftp.cwd(current_dir)
                            except ftplib.error_perm:
                                ftp.mkd(current_dir)
                                ftp.cwd(current_dir)
            except ftplib.Error as e:
                error_msg = f"Failed to change/create directory on FTP server: {str(e)}"
                logger.error(error_msg)
                raise Exception(error_msg)

        # Upload the file
        with open(file_path, "rb") as file_data:
            ftp.storbinary(f"STOR {filename}", file_data)

        # Close FTP connection
        ftp.quit()

        logger.info(f"[{task_id}] Successfully uploaded {filename} to FTP server at {settings.ftp_host}")
        log_task_progress(task_id, "upload_to_ftp", "success", f"Uploaded to FTP: {filename}", file_id=file_id)
        return {
            "status": "Completed",
            "file": file_path,
            "ftp_host": settings.ftp_host,
            "ftp_path": f"{ftp_folder_setting}/{filename}" if ftp_folder_setting else filename,
            "used_tls": used_tls,
        }

    except Exception as e:
        error_msg = f"Failed to upload {filename} to FTP server: {str(e)}"
        logger.error(f"[{task_id}] {error_msg}")
        log_task_progress(task_id, "upload_to_ftp", "failure", error_msg, file_id=file_id)
        raise Exception(error_msg)