Kai-Wei Chiang

# python s3_uploader.py /path/to/files repository_name --notes "Optional notes"
import hashlib
import multiprocessing
from multiprocessing import Pool
import sys
import logging
from logging.handlers import QueueHandler
from datetime import datetime
import os
import time
import json
from enum import Enum
import tempfile
import magic
from typing import Optional, Dict, Any
from dotenv import load_dotenv
from datetime import timezone

# File handling
import py7zr
import pyzipper
from magika import Magika

# S3 and database
from minio import Minio
import clickhouse_connect

load_dotenv(override=True)

class UploadResult(Enum):
    CORRECTLY = 0
    FAILED = 1
    SKIPPED = 2  # For duplicates

class FileNameFormatter(logging.Formatter):
    def format(self, record):
        record.filenameinfo = getattr(record, "filenameinfo", "unknown")
        return super().format(record)

def setup_logger(log_file, filename):
    """Set up a logger for a specific file"""
    logger = logging.getLogger(filename)
    if not logger.handlers:
        if os.getenv("SERVER_ENV") == "prod":
            logger.setLevel(logging.INFO)
        else:
            logger.setLevel(logging.DEBUG)
            
        # Create file handler
        handler = logging.FileHandler(log_file)
        formatter = FileNameFormatter(
            "%(asctime)s - %(filenameinfo)s - %(levelname)s - %(message)s"
        )
        handler.setFormatter(formatter)
        logger.addHandler(handler)
        logger.propagate = False
    return logger

def logger_thread(log_queue, log_file):
    handler = logging.FileHandler(log_file)
    formatter = FileNameFormatter(
        "%(asctime)s - %(filenameinfo)s - %(levelname)s - %(message)s"
    )
    handler.setFormatter(formatter)

    while True:
        try:
            record = log_queue.get()
            if record is None:
                break
            message = formatter.format(record)
            handler.stream.write(message + "\n")
            handler.stream.flush()
        except Exception:
            import traceback
            print("[ERR] Error in logger thread:", file=sys.stderr)
            traceback.print_exc(file=sys.stderr)

def get_s3_key(sha256: str, is_archived: bool, original_ext: str = None) -> str:
    """
    Generate S3 key using sharding pattern with appropriate extension
    
    Args:
        sha256: The file's SHA256 hash
        is_archived: Whether the file was already archived (zip/7z)
        original_ext: The original file extension (if any)
    """
    # If file was already archived, keep its extension
    if is_archived and original_ext and original_ext.lower() in ['.zip', '.7z']:
        extension = original_ext
    else:
        # For files we archived ourselves, use .zip
        extension = '.zip'
        
    return f"{sha256[:2]}/{sha256[2:4]}/{sha256}{extension}"

def process_single_file(args):
    """Independent worker function for processing a single file"""
    filepath, file_number, total_files, config, log_file = args
    filename = os.path.basename(filepath)
    base_logger = setup_logger(log_file, filename)
    extra = {"filenameinfo": filename}
    logger = logging.LoggerAdapter(base_logger, extra)
    s3_client = None
    ch_client = None

    try:
        logger.info(f"Processing file {file_number}/{total_files}: {filepath}")

        # Create connections for this process
        s3_client = Minio(
            config["s3_endpoint"],
            access_key=config["s3_access_key"],
            secret_key=config["s3_secret_key"],
            secure=True
        )
        
        ch_client = clickhouse_connect.get_client(
            host=config['clickhouse_host'],
            port=config['clickhouse_port'],
            username=config['clickhouse_user'],
            password=config['clickhouse_password'],
            database=config['clickhouse_database'],
            verify=config['clickhouse_verify']
        )

        # Process the file and get contents
        ext = os.path.splitext(filepath.lower())[1]
        is_archived = ext in ['.zip', '.7z']

        with tempfile.TemporaryDirectory() as temp_dir:
            try:
                if filename.startswith("."):
                    logger.debug(f"Skipping hidden file: {filename}")
                    return UploadResult.SKIPPED

                elif ext == ".zip":
                    logger.debug(f"Processing ZIP file: {filepath}")
                    with pyzipper.AESZipFile(filepath) as zf:
                        zf.pwd = b"infected"
                        filename = zf.namelist()[0]
                        zf.extractall(temp_dir)
                        extracted_path = os.path.join(temp_dir, filename)
                        
                        with open(extracted_path, "rb") as f:
                            file_content = f.read()
                        with open(filepath, "rb") as f:
                            archive_content = f.read()
                            
                elif ext == ".7z":
                    logger.debug(f"Processing 7z file: {filepath}")
                    with py7zr.SevenZipFile(filepath, mode="r", password="infected") as z:
                        z.extractall(path=temp_dir)
                        # Get first file in the archive
                        for root, _, files in os.walk(temp_dir):
                            if files:
                                filename = files[0]
                                extracted_path = os.path.join(root, filename)
                                break
                                
                        with open(extracted_path, "rb") as f:
                            file_content = f.read()
                        with open(filepath, "rb") as f:
                            archive_content = f.read()
                else:
                    logger.debug(f"Processing non-archived file: {filepath}")
                    # For non-archived files, create password protected zip
                    with open(filepath, "rb") as f:
                        file_content = f.read()
                    filename = os.path.basename(filepath)
                    
                    # Create password protected zip
                    with tempfile.NamedTemporaryFile(delete=False) as temp_zip:
                        with pyzipper.AESZipFile(
                            temp_zip.name,
                            'w',
                            compression=pyzipper.ZIP_LZMA,
                            encryption=pyzipper.WZ_AES
                        ) as zf:
                            zf.pwd = b'infected'
                            zf.writestr(filename, file_content)
                        
                        with open(temp_zip.name, 'rb') as f:
                            archive_content = f.read()
                    
                    os.unlink(temp_zip.name)

                # Calculate hash
                sha256 = hashlib.sha256(file_content).hexdigest()
                # s3_key = f"{sha256[:2]}/{sha256[2:4]}/{sha256}"
                s3_key = get_s3_key(sha256, is_archived, ext if is_archived else None)
                logger.debug(f"Calculated SHA256: {sha256}")

                # Check if exists in S3
                try:
                    s3_client.stat_object(config["s3_bucket"], s3_key)
                    exists = True
                    logger.debug(f"File exists in S3: {s3_key}")
                except:
                    exists = False
                    logger.debug(f"File does not exist in S3: {s3_key}")

                # Prepare metadata
                max_timestamp = int(datetime(9999, 12, 31, 23, 59, 59, tzinfo=timezone.utc).timestamp() * 1000)
                now = datetime.now(timezone.utc)
                now_timestamp = int(now.timestamp() * 1000)
                inverted_timestamp = datetime.fromtimestamp((max_timestamp - now_timestamp)/1000, tz=timezone.utc)


                data = [[
                    sha256,                     # String
                    filename,                   # String
                    config['repository'],       # String
                    len(file_content),         # UInt64
                    magic.from_buffer(file_content),        # String
                    magic.from_buffer(file_content, mime=True),  # String
                    Magika().identify_bytes(file_content).output.ct_label,  # String
                    now,                       # upload_date
                    now,                       # first_seen
                    inverted_timestamp,        # version_date for replacing
                    config['s3_bucket'],       # String
                    s3_key,                    # String
                    config['notes'] if config['notes'] else None  # Nullable(String)
                ]]

                column_names = [
                    'sha256', 'filename', 'repository', 'file_size',
                    'filetype', 'filetype_mime', 'filetype_magika',
                    'upload_date', 'first_seen', 'version_date', 's3_bucket', 's3_key', 'notes'
                ]

                column_type_names = [
                    'String', 'String', 'LowCardinality(String)', 'UInt64',
                    'String', 'String', 'String',
                    'DateTime64(3, \'UTC\')', 'DateTime64(3, \'UTC\')', 
                    'DateTime64(3, \'UTC\')', 'String', 'String', 'Nullable(String)'
                ]
                
                if exists:
                    # Just update database
                    try:
                        ch_client.insert(
                            'samples_catalog',
                            data,
                            column_names=column_names,
                            column_type_names=column_type_names,
                            settings={'input_format_values_interpret_expressions': 0}
                        )
                        logger.info(f"Updated metadata for existing file: {sha256}")
                    except Exception as e:
                        logger.error(f"Error inserting metadata: {str(e)}")
                        import traceback
                        logger.error(f"Full traceback: {traceback.format_exc()}")
                    return UploadResult.SKIPPED
                else:
                    try:    
                        # Upload to S3 and insert metadata
                        import io
                        file_data = io.BytesIO(archive_content)
                        s3_client.put_object(
                            bucket_name=config['s3_bucket'],
                                object_name=s3_key,
                                data=file_data,
                                length=len(archive_content)
                            )
                        logger.info(f"Uploaded to S3: {s3_key}")
                    except Exception as e:
                        logger.error(f"Error uploading to S3: {str(e)}")
                        import traceback
                        logger.error(f"Full traceback: {traceback.format_exc()}")
                        return UploadResult.FAILED
                    
                    try:
                        ch_client.insert(
                            'samples_catalog',
                            data,
                            column_names=column_names,
                            column_type_names=column_type_names,
                            settings={'input_format_values_interpret_expressions': 0}
                        )
                        logger.info(f"Successfully processed file: {sha256}")
                        return UploadResult.CORRECTLY
                    except Exception as e:
                        logger.error(f"Error inserting metadata: {str(e)}")
                        import traceback
                        logger.error(f"Full traceback: {traceback.format_exc()}")
                        return UploadResult.FAILED

            except Exception as e:
                logger.error(f"Error processing file: {str(e)}")
                import traceback
                logger.error(f"Full traceback: {traceback.format_exc()}")
                return UploadResult.FAILED

    except Exception as e:
        logger.error(f"Error in worker: {str(e)}")
        return UploadResult.FAILED
    finally:
        # if s3_client:
        #     s3_client.close()
        if ch_client:
            ch_client.close()
        if base_logger and base_logger.handlers:
            for handler in base_logger.handlers:
                handler.close()
            base_logger.handlers.clear()

class S3Uploader:
    def __init__(self, path: str, repository: str, notes: Optional[str] = None):
        """
        Initialize S3 uploader
        
        Args:
            path: Path to file or directory to process
            repository: Repository name
            notes: Optional notes to add to all files
        """
        self.path = path
        self.repository = repository
        self.notes = notes
        
        # Initialize multiprocessing components
        self.manager = multiprocessing.Manager()
        # self.log_queue = self.manager.Queue()
        self.total_results = self.manager.dict({result: 0 for result in UploadResult})
        
        # Set up logging
        self.today = datetime.today().strftime("%Y%m%dT%H%M%S")
        # self.log_file = os.getenv("LOG_FILE_PATH", "./logs/") + f"{self.today}-{self.repository}-upload.txt"
        log_base_path = os.getenv("LOG_FILE_PATH")
        if not log_base_path:
            log_base_path = os.path.join(os.getcwd(), "logs")  # Default to ./logs directory
        self.log_file = os.path.join(log_base_path, f"{self.today}-{self.repository}-upload.txt")
        os.makedirs(os.path.dirname(self.log_file), exist_ok=True)

        # Initialize S3 client
        self.s3_client = Minio(
            os.getenv("S3_ENDPOINT"),
            access_key=os.getenv("S3_ACCESS_KEY"),
            secret_key=os.getenv("S3_SECRET_KEY"),
            secure=True  # Set to False if not using HTTPS
        )
        self.bucket_name = os.getenv("S3_BUCKET")

        # Initialize ClickHouse client
        self.ch_client = clickhouse_connect.get_client(
            host=os.getenv('CLICKHOUSE_HOST'),
            port=os.getenv('CLICKHOUSE_PORT'),
            username=os.getenv('CLICKHOUSE_USER'),
            password=os.getenv('CLICKHOUSE_PASSWORD'),
            database=os.getenv('CLICKHOUSE_DATABASE'),
            verify=os.getenv('CLICKHOUSE_ENFORCE_SSL', 'False')
        )
        
        # Ensure database table exists
        self._create_table()

    def _create_table(self):
        """Create the file catalog table if it doesn't exist"""
        create_table_query = """
            CREATE TABLE IF NOT EXISTS samples_catalog (
                sha256 String,
                filename String,
                repository LowCardinality(String),
                file_size UInt64,
                filetype String,
                filetype_mime String,
                filetype_magika String,
                upload_date DateTime64(3, 'UTC'),
                first_seen DateTime64(3, 'UTC'),
                version_date DateTime64(3, 'UTC'),  -- This will be inverted timestamp
                s3_bucket String,
                s3_key String,
                notes Nullable(String),
                PRIMARY KEY (sha256, repository)
            ) ENGINE = ReplacingMergeTree(version_date)
            ORDER BY (sha256, repository)
        """
        self.ch_client.command(create_table_query)

    # # def upload(self):
    # #     """Main upload function with multiprocessing support"""
    # #     start_time = time.time()

    # #     config = {
    # #         "s3_endpoint": os.getenv("S3_ENDPOINT"),
    # #         "s3_access_key": os.getenv("S3_ACCESS_KEY"),
    # #         "s3_secret_key": os.getenv("S3_SECRET_KEY"),
    # #         "s3_bucket": os.getenv("S3_BUCKET"),
    # #         "clickhouse_host": os.getenv("CLICKHOUSE_HOST"),
    # #         "clickhouse_port": os.getenv("CLICKHOUSE_PORT"),
    # #         "clickhouse_user": os.getenv("CLICKHOUSE_USER"),
    # #         "clickhouse_password": os.getenv("CLICKHOUSE_PASSWORD"),
    # #         "clickhouse_database": os.getenv("CLICKHOUSE_DATABASE"),
    # #         "clickhouse_verify": os.getenv("CLICKHOUSE_ENFORCE_SSL", "False"),
    # #         "repository": self.repository,
    # #         "notes": self.notes
    # #     }

    # #     try:
    # #         # Get list of files
    # #         if os.path.isfile(self.path):
    # #             files = [self.path] if not self.path.endswith('.txt') else [
    # #                 line.strip() for line in open(self.path) 
    # #                 if line.strip() and not os.path.basename(line.strip()).startswith('.')
    # #             ]
    # #         elif os.path.isdir(self.path):
    # #             files = [
    # #                 os.path.join(root, file)
    # #                 for root, _, files in os.walk(self.path)
    # #                 for file in files
    # #                 if not file.startswith('.')
    # #             ]
    # #         else:
    # #             print(f"[ERR] Invalid path: {self.path}")
    # #             return

    # #         total_files = len(files)
    # #         print(f"Found {total_files} files to process")

    # #         with Pool(processes=max(1, multiprocessing.cpu_count() - 1)) as pool:
    # #             results = pool.map(
    # #                 process_single_file,
    # #                 [(f, i + 1, total_files, config, self.log_file) for i, f in enumerate(files)]
    # #             )

    # #         # Update statistics
    # #         for result in results:
    # #             if result is not None:
    # #                 self.total_results[result] += 1

    # #         # Generate summary
    # #         end_time = time.time()
    # #         elapsed_time = end_time - start_time
    # #         elapsed_time_pretty = time.strftime("%H:%M:%S", time.gmtime(elapsed_time))

    # #         summary = (
    # #             f"\n\nUpload finished for {self.path}"
    # #             f"\nTime required: {elapsed_time_pretty}"
    # #             f"\nResults:"
    # #             f"\n- Total processed: {sum(self.total_results.values())}"
    # #             f"\n- Successfully uploaded: {self.total_results[UploadResult.CORRECTLY]}"
    # #             f"\n- Skipped (already exists): {self.total_results[UploadResult.SKIPPED]}"
    # #             f"\n- Failed: {self.total_results[UploadResult.FAILED]}\n"
    # #         )

    # #         with open(self.log_file, "a") as f:
    # #             f.write(summary)

    # #         print(summary)

    # #     except Exception as e:
    # #         print(f"[ERR] Error in upload process: {str(e)}")

    # def upload(self):
    #     """Main upload function with multiprocessing support"""
    #     start_time = time.time()
    #     BATCH_SIZE = 1000  # Process files in batches of 1000

    #     config = {
    #         "s3_endpoint": os.getenv("S3_ENDPOINT"),
    #         "s3_access_key": os.getenv("S3_ACCESS_KEY"),
    #         "s3_secret_key": os.getenv("S3_SECRET_KEY"),
    #         "s3_bucket": os.getenv("S3_BUCKET"),
    #         "clickhouse_host": os.getenv("CLICKHOUSE_HOST"),
    #         "clickhouse_port": os.getenv("CLICKHOUSE_PORT"),
    #         "clickhouse_user": os.getenv("CLICKHOUSE_USER"),
    #         "clickhouse_password": os.getenv("CLICKHOUSE_PASSWORD"),
    #         "clickhouse_database": os.getenv("CLICKHOUSE_DATABASE"),
    #         "clickhouse_verify": os.getenv("CLICKHOUSE_ENFORCE_SSL", "False"),
    #         "repository": self.repository,
    #         "notes": self.notes
    #     }

    #     try:
    #         # Get list of files
    #         if os.path.isfile(self.path):
    #             files = [self.path] if not self.path.endswith('.txt') else [
    #                 line.strip() for line in open(self.path) 
    #                 if line.strip() and not os.path.basename(line.strip()).startswith('.')
    #             ]
    #         elif os.path.isdir(self.path):
    #             files = [
    #                 os.path.join(root, file)
    #                 for root, _, files in os.walk(self.path)
    #                 for file in files
    #                 if not file.startswith('.')
    #             ]
    #         else:
    #             print(f"[ERR] Invalid path: {self.path}")
    #             return

    #         total_files = len(files)
    #         print(f"Found {total_files} files to process")

    #         # Process files in batches
    #         for i in range(0, total_files, BATCH_SIZE):
    #             batch_files = files[i:i + BATCH_SIZE]
    #             batch_start = i + 1
    #             batch_end = min(i + BATCH_SIZE, total_files)
    #             print(f"\nProcessing batch {batch_start}-{batch_end} of {total_files}")
                
    #             with Pool(processes=max(1, multiprocessing.cpu_count() - 1)) as pool:
    #                 batch_results = pool.map(
    #                     process_single_file,
    #                     [(f, j + 1, total_files, config, self.log_file) 
    #                         for j, f in enumerate(batch_files, start=i)]
    #                 )
                    
    #                 # Update statistics for this batch
    #                 for result in batch_results:
    #                     if result is not None:
    #                         self.total_results[result] += 1

    #                 # Print intermediate summary
    #                 print(f"Batch {batch_start}-{batch_end} completed:")
    #                 print(f"- Successfully uploaded: {sum(1 for r in batch_results if r == UploadResult.CORRECTLY)}")
    #                 print(f"- Skipped (exists): {sum(1 for r in batch_results if r == UploadResult.SKIPPED)}")
    #                 print(f"- Failed: {sum(1 for r in batch_results if r == UploadResult.FAILED)}")

    #         # Generate final summary
    #         end_time = time.time()
    #         elapsed_time = end_time - start_time
    #         elapsed_time_pretty = time.strftime("%H:%M:%S", time.gmtime(elapsed_time))

    #         summary = (
    #             f"\n\nUpload finished for {self.path}"
    #             f"\nTime required: {elapsed_time_pretty}"
    #             f"\nResults:"
    #             f"\n- Total processed: {sum(self.total_results.values())}"
    #             f"\n- Successfully uploaded: {self.total_results[UploadResult.CORRECTLY]}"
    #             f"\n- Skipped (already exists): {self.total_results[UploadResult.SKIPPED]}"
    #             f"\n- Failed: {self.total_results[UploadResult.FAILED]}\n"
    #         )

    #         with open(self.log_file, "a") as f:
    #             f.write(summary)

    #         print(summary)

    #     except Exception as e:
    #         print(f"[ERR] Error in upload process: {str(e)}")

    def upload(self):
        """Main upload function with multiprocessing support"""
        start_time = time.time()
        BATCH_SIZE = 1000  # Process files in batches of 1000

        config = {
            "s3_endpoint": os.getenv("S3_ENDPOINT"),
            "s3_access_key": os.getenv("S3_ACCESS_KEY"),
            "s3_secret_key": os.getenv("S3_SECRET_KEY"),
            "s3_bucket": os.getenv("S3_BUCKET"),
            "clickhouse_host": os.getenv("CLICKHOUSE_HOST"),
            "clickhouse_port": os.getenv("CLICKHOUSE_PORT"),
            "clickhouse_user": os.getenv("CLICKHOUSE_USER"),
            "clickhouse_password": os.getenv("CLICKHOUSE_PASSWORD"),
            "clickhouse_database": os.getenv("CLICKHOUSE_DATABASE"),
            "clickhouse_verify": os.getenv("CLICKHOUSE_ENFORCE_SSL", "False"),
            "repository": self.repository,
            "notes": self.notes
        }

        try:
            # Get all existing filenames for this repository
            query = f"SELECT filename FROM samples_catalog WHERE repository = '{self.repository}'"
            # existing_files = set()
            # for row in self.ch_client.query(query).result_rows:
            #     filename = os.path.splitext(row[0])[0]  # Strip extension
            #     existing_files.add(filename)
            existing_files = set(row[0] for row in self.ch_client.query(query).result_rows)
            print(f"Found {len(existing_files)} existing files in repository")

            # Get list of files to process
            if os.path.isfile(self.path):
                files = [self.path] if not self.path.endswith('.txt') else [
                    line.strip() for line in open(self.path) 
                    if line.strip() and not os.path.basename(line.strip()).startswith('.')
                ]
            elif os.path.isdir(self.path):
                files = [
                    os.path.join(root, file)
                    for root, _, files in os.walk(self.path)
                    for file in files
                    if not file.startswith('.')
                ]
            else:
                print(f"[ERR] Invalid path: {self.path}")
                return

            # Filter out files that are already in the repository
            files_to_process = []
            for f in files:
                basename = os.path.splitext(os.path.basename(f))[0]
                if basename not in existing_files:
                    files_to_process.append(f)

            total_files = len(files_to_process)
            skipped_files = len(files) - total_files
            print(f"Found {total_files} new files to process (skipped {skipped_files} existing files)")

            if total_files == 0:
                print("No new files to process")
                return

            # Process files in batches
            for i in range(0, total_files, BATCH_SIZE):
                batch_files = files_to_process[i:i + BATCH_SIZE]
                batch_start = i + 1
                batch_end = min(i + BATCH_SIZE, total_files)
                print(f"\nProcessing batch {batch_start}-{batch_end} of {total_files}")
                
                with Pool(processes=max(1, multiprocessing.cpu_count() - 1)) as pool:
                    batch_results = pool.map(
                        process_single_file,
                        [(f, j + 1, total_files, config, self.log_file) 
                        for j, f in enumerate(batch_files, start=i)]
                    )
                    
                    # Update statistics for this batch
                    for result in batch_results:
                        if result is not None:
                            self.total_results[result] += 1

                    # Print intermediate summary
                    print(f"Batch {batch_start}-{batch_end} completed:")
                    print(f"- Successfully uploaded: {sum(1 for r in batch_results if r == UploadResult.CORRECTLY)}")
                    print(f"- Skipped (exists): {sum(1 for r in batch_results if r == UploadResult.SKIPPED)}")
                    print(f"- Failed: {sum(1 for r in batch_results if r == UploadResult.FAILED)}")

            # Generate final summary
            end_time = time.time()
            elapsed_time = end_time - start_time
            elapsed_time_pretty = time.strftime("%H:%M:%S", time.gmtime(elapsed_time))

            summary = (
                f"\n\nUpload finished for {self.path}"
                f"\nTime required: {elapsed_time_pretty}"
                f"\nResults:"
                f"\n- Files already in repository: {skipped_files}"
                f"\n- New files processed: {total_files}"
                f"\n- Successfully uploaded: {self.total_results[UploadResult.CORRECTLY]}"
                f"\n- Skipped (already exists): {self.total_results[UploadResult.SKIPPED]}"
                f"\n- Failed: {self.total_results[UploadResult.FAILED]}\n"
            )

            with open(self.log_file, "a") as f:
                f.write(summary)

            print(summary)

        except Exception as e:
            print(f"[ERR] Error in upload process: {str(e)}")
            import traceback
            print(traceback.format_exc())

if __name__ == "__main__":
    import argparse

    parser = argparse.ArgumentParser(description='Upload malware samples to S3 and catalog them.')
    parser.add_argument('path', help='Path to file or directory to process')
    parser.add_argument('repository', help='Repository name')
    parser.add_argument('--notes', help='Optional notes to add to all files', default=None)
    
    args = parser.parse_args()
    
    uploader = S3Uploader(args.path, args.repository, args.notes)
    uploader.upload()