boto3-assist 0.6.0__py3-none-any.whl → 0.7.0__py3-none-any.whl

boto3_assist/s3/s3_object.py

@@ -0,0 +1,606 @@
+"""
+Geek Cafe, LLC
+Maintainers: Eric Wilson
+MIT License.  See Project Root for the license information.
+"""
+
+import os
+import tempfile
+import time
+import io
+from typing import Any, Dict, Optional, List
+
+from aws_lambda_powertools import Logger
+from botocore.exceptions import ClientError
+
+from boto3_assist.errors.custom_exceptions import InvalidHttpMethod
+from boto3_assist.s3.s3_connection import S3Connection
+from boto3_assist.utilities.datetime_utility import DatetimeUtility
+from boto3_assist.utilities.file_operations import FileOperations
+from boto3_assist.utilities.http_utility import HttpUtility
+
+
+logger = Logger(child=True)
+
+
+class S3Object:
+    """S3 Object Actions"""
+
+    def __init__(self, connection: S3Connection):
+        self.connection = connection or S3Connection()
+
+    def delete(self, *, bucket_name: str, key: str) -> Dict[str, Any]:
+        """
+        Deletes an object key
+
+        Args:
+            bucket_name (str): The AWS Bucket Name
+            key (str): The Object Key
+        """
+        s3 = self.connection.client
+        # see if the object exists
+        try:
+            response = s3.head_object(Bucket=bucket_name, Key=key)
+            response = s3.delete_object(Bucket=bucket_name, Key=key)
+        except s3.exceptions.NoSuchKey:
+            response = {"ResponseMetadata": {"HTTPStatusCode": 404}}
+        except s3.exceptions.ClientError as e:
+            if e.response.get("Error", {}).get("Code") == "404":
+                response = {"ResponseMetadata": {"HTTPStatusCode": 404}}
+            else:
+                raise e
+
+        return dict(response)
+
+    def delete_all_versions(
+        self, *, bucket_name: str, key: str, include_deleted: bool = False
+    ) -> List[str]:
+        """
+        Deletes an object key and all the versions for that object key
+
+        Args:
+            bucket_name (str): The AWS Bucket Name
+            key (str): The Object Kuye
+            include_deleted (bool, optional): Should deleted files be removed as well.
+                If True it will look for the object keys with the deleted marker and remove it.
+                Defaults to False.
+        """
+        s3 = self.connection.client
+        paginator = s3.get_paginator("list_object_versions")
+        files: List[str] = []
+
+        for page in paginator.paginate(Bucket=bucket_name, Prefix=key):
+            # Delete object versions
+            if "Versions" in page:
+                for version in page["Versions"]:
+                    s3.delete_object(
+                        Bucket=bucket_name,
+                        Key=version["Key"],
+                        VersionId=version["VersionId"],
+                    )
+
+                    files.append(f"{version['Key']} - {version['VersionId']}")
+
+                if include_deleted:
+                    # delete a previous files that may have just been a soft delete.
+                    if "DeleteMarkers" in page:
+                        for marker in page["DeleteMarkers"]:
+                            s3.delete_object(
+                                Bucket=bucket_name,
+                                Key=marker["Key"],
+                                VersionId=marker["VersionId"],
+                            )
+
+                            files.append(
+                                f"{marker['Key']}:{marker['VersionId']}:delete-marker"
+                            )
+            else:
+                response = self.delete(bucket_name=bucket_name, key=key)
+                if response["ResponseMetadata"]["HTTPStatusCode"] == 404:
+                    return files
+
+                files.append(key)
+
+        return files
+
+    def generate_presigned_url(
+        self,
+        *,
+        bucket_name: str,
+        key_path: str,
+        file_name: str,
+        meta_data: Optional[dict] = None,
+        expiration: int = 3600,
+        method_type: str = "POST",
+        user_id: Optional[str] = None,
+    ) -> Dict[str, Any]:
+        """
+        Create a signed URL for uploading a file to S3.
+        :param bucket_name: The name of the S3 bucket.
+        :param user_id: The user ID of the user uploading the file.
+        :param file_name: The file name of the file being uploaded.
+        :param aws_profile: The name of the AWS profile to use.
+        :param aws_region: The name of the AWS region to use.
+        :param expiration: The number of seconds the URL is valid for.
+        :return: The signed URL.
+        """
+        start = DatetimeUtility.get_utc_now()
+        logger.debug(
+            f"Creating signed URL for bucket {bucket_name} for user {user_id} and file {file_name} at {start} UTC"
+        )
+
+        file_extension = FileOperations.get_file_extension(file_name)
+
+        local_meta = {
+            "user_id": f"{user_id}",
+            "file_name": f"{file_name}",
+            "extension": f"{file_extension}",
+            "method": "pre-signed-upload",
+        }
+
+        if not meta_data:
+            meta_data = local_meta
+        else:
+            meta_data.update(local_meta)
+
+        key = key_path
+        method_type = method_type.upper()
+
+        signed_url: str | Dict[str, Any]
+        if method_type == "PUT":
+            signed_url = self.connection.client.generate_presigned_url(
+                "put_object",
+                Params={
+                    "Bucket": f"{bucket_name}",
+                    "Key": f"{key}",
+                    # NOTE: if you include the ContentType or Metadata then its required in the when they upload the file
+                    # Otherwise you will get a `SignatureDoesNotMatch` error
+                    # for now I'm commenting it out.
+                    #'ContentType': 'application/octet-stream',
+                    #'ACL': 'private',
+                    # "Metadata": meta_data,
+                },
+                ExpiresIn=expiration,  # URL is valid for x seconds
+            )
+        elif method_type == "POST":
+            signed_url = self.connection.client.generate_presigned_post(
+                bucket_name,
+                key,
+                ExpiresIn=expiration,  # URL is valid for x seconds
+            )
+        elif method_type == "GET":
+            signed_url = self.connection.client.generate_presigned_url(
+                "get_object",
+                Params={
+                    "Bucket": f"{bucket_name}",
+                    "Key": f"{key}",
+                },
+                ExpiresIn=expiration,  # URL is valid for x seconds
+            )
+        else:
+            raise InvalidHttpMethod(
+                f'Unknown method type was referenced.  valid types are "PUT", "POST", "GET" , "{method_type}" as used '
+            )
+
+        end = DatetimeUtility.get_utc_now()
+        logger.debug(f"Signed URL created in {end-start}")
+
+        response = {
+            "signed_url": signed_url,
+            "key": key,
+            "meta_data": meta_data,
+        }
+
+        return response
+
+    def upload_file_obj(self, *, bucket: str, key: str, file_obj: bytes | str) -> str:
+        """
+        Uploads a file object to s3. Returns the full s3 path s3://<bucket>/<key>
+        """
+
+        if key.startswith("/"):
+            # remove the first slash
+            key = key[1:]
+
+        logger.debug(
+            {
+                "metric_filter": "upload_file_to_s3",
+                "bucket": bucket,
+                "key": key,
+            }
+        )
+        try:
+            # convert if necessary
+            file_obj: bytes = (
+                file_obj.encode("utf-8") if isinstance(file_obj, str) else file_obj
+            )
+            self.connection.client.upload_fileobj(
+                Fileobj=io.BytesIO(file_obj), Bucket=bucket, Key=key
+            )
+
+        except ClientError as ce:
+            error = {
+                "metric_filter": "upload_file_to_s3_failure",
+                "s3 upload": "failure",
+                "bucket": bucket,
+                "key": key,
+            }
+            logger.error(error)
+            raise RuntimeError(error) from ce
+
+        return f"s3://{bucket}/{key}"
+
+    def upload_file(
+        self,
+        *,
+        bucket: str,
+        key: str,
+        local_file_path: str,
+        throw_error_on_failure: bool = False,
+    ) -> str | None:
+        """
+        Uploads a file to s3. Returns the full s3 path s3://<bucket>/<key>
+        """
+
+        if key.startswith("/"):
+            # remove the first slash
+            key = key[1:]
+
+        # build the path
+        s3_path = f"s3://{bucket}/{key}"
+
+        logger.debug(
+            {
+                "metric_filter": "upload_file_to_s3",
+                "bucket": bucket,
+                "key": key,
+                "local_file_path": local_file_path,
+            }
+        )
+        try:
+            self.connection.client.upload_file(local_file_path, bucket, key)
+
+        except ClientError as ce:
+            error = {
+                "metric_filter": "upload_file_to_s3_failure",
+                "s3 upload": "failure",
+                "bucket": bucket,
+                "key": key,
+                "local_file_path": local_file_path,
+            }
+            logger.error(error)
+
+            if throw_error_on_failure:
+                raise RuntimeError(error) from ce
+
+            return None
+
+        return s3_path
+
+    def download_file(
+        self,
+        *,
+        bucket: str,
+        key: str,
+        local_directory: str | None = None,
+        local_file_path: str | None = None,
+        retry_attempts: int = 3,
+        retry_sleep: int = 5,
+    ) -> str:
+        """Download a file from s3"""
+        exception: Exception | None = None
+
+        if retry_attempts == 0:
+            retry_attempts = 1
+
+        for i in range(retry_attempts):
+            exception = None
+            try:
+                path = self.download_file_no_retries(
+                    bucket=bucket,
+                    key=key,
+                    local_directory=local_directory,
+                    local_file_path=local_file_path,
+                )
+                if path and os.path.exists(path):
+                    return path
+
+            except Exception as e:  # pylint: disable=w0718
+                logger.warning(
+                    {
+                        "action": "download_file",
+                        "result": "failure",
+                        "exception": str(e),
+                        "attempt": i + 1,
+                        "retry_attempts": retry_attempts,
+                    }
+                )
+
+                exception = e
+
+                # sleep for a bit
+                attempt = i + 1
+                time.sleep(attempt * retry_sleep)
+
+        if exception:
+            logger.exception(
+                {
+                    "action": "download_file",
+                    "result": "failure",
+                    "exception": str(exception),
+                    "retry_attempts": retry_attempts,
+                }
+            )
+
+            raise exception from exception
+
+        raise RuntimeError("Unable to download file")
+
+    def download_file_no_retries(
+        self,
+        bucket: str,
+        key: str,
+        local_directory: str | None = None,
+        local_file_path: str | None = None,
+    ) -> str:
+        """
+        Downloads a file from s3
+
+        Args:
+            bucket (str): s3 bucket
+            key (str): the s3 object key
+            local_directory (str, optional): Local directory to download to. Defaults to None.
+            If None, we'll use a local tmp directory.
+
+        Raises:
+            e:
+
+        Returns:
+            str: Path to the downloaded file.
+        """
+
+        decoded_object_key: str
+        try:
+            logger.debug(
+                {
+                    "action": "downloading file",
+                    "bucket": bucket,
+                    "key": key,
+                    "local_directory": local_directory,
+                }
+            )
+            return self.__download_file(bucket, key, local_directory, local_file_path)
+        except FileNotFoundError:
+            logger.warning(
+                {
+                    "metric_filter": "download_file_error",
+                    "error": "FileNotFoundError",
+                    "message": "attempting to find it decoded",
+                    "bucket": bucket,
+                    "key": key,
+                }
+            )
+
+            # attempt to decode the key
+            decoded_object_key = HttpUtility.decode_url(key)
+
+            logger.error(
+                {
+                    "metric_filter": "download_file_error",
+                    "error": "FileNotFoundError",
+                    "message": "attempting to find it decoded",
+                    "bucket": bucket,
+                    "key": key,
+                    "decoded_object_key": decoded_object_key,
+                }
+            )
+
+            return self.__download_file(bucket, decoded_object_key, local_directory)
+
+        except Exception as e:
+            logger.error(
+                {
+                    "metric_filter": "download_file_error",
+                    "error": str(e),
+                    "bucket": bucket,
+                    "decoded_object_key": decoded_object_key,
+                }
+            )
+            raise e
+
+    def stream_file(self, bucket_name: str, key: str) -> Dict[str, Any]:
+        """
+        Gets a file from s3 and returns the response.
+        The "Body" is a streaming body object.  You can read it like a file.
+        For example:
+
+        with response["Body"] as f:
+            data = f.read()
+            print(data)
+
+        """
+
+        logger.debug(
+            {
+                "source": "download_file",
+                "action": "downloading a file from s3",
+                "bucket": bucket_name,
+                "key": key,
+            }
+        )
+
+        response: Dict[str, Any] = {}
+        error = None
+
+        try:
+            response = dict(
+                self.connection.client.get_object(Bucket=bucket_name, Key=key)
+            )
+
+            logger.debug(
+                {"metric_filter": "s3_download_response", "response": str(response)}
+            )
+
+        except Exception as e:  # pylint: disable=W0718
+            error = str(e)
+            logger.error({"metric_filter": "s3_download_error", "error": str(e)})
+            raise RuntimeError(
+                {
+                    "metric_filter": "s3_download_error",
+                    "error": str(e),
+                    "bucket": bucket_name,
+                    "key": key,
+                }
+            ) from e
+
+        finally:
+            logger.debug(
+                {
+                    "source": "download_file",
+                    "action": "downloading a file from s3",
+                    "bucket": bucket_name,
+                    "key": key,
+                    "response": response,
+                    "errors": error,
+                }
+            )
+
+        return response
+
+    def __download_file(
+        self,
+        bucket: str,
+        key: str,
+        local_directory: str | None = None,
+        local_file_path: str | None = None,
+    ):
+        if local_directory and local_file_path:
+            raise ValueError(
+                "Only one of local_directory or local_file_path can be provided"
+            )
+
+        if local_directory and not os.path.exists(local_directory):
+            FileOperations.makedirs(local_directory)
+
+        if local_file_path and not os.path.exists(os.path.dirname(local_file_path)):
+            FileOperations.makedirs(os.path.dirname(local_file_path))
+
+        file_name = self.__get_file_name_from_path(key)
+        if local_directory is None and local_file_path is None:
+            local_path = self.get_local_path_for_file(file_name)
+        elif local_directory:
+            local_path = os.path.join(local_directory, file_name)
+        else:
+            local_path = local_file_path
+
+        logger.debug(
+            {
+                "source": "download_file",
+                "action": "downloading a file from s3",
+                "bucket": bucket,
+                "key": key,
+                "file_name": file_name,
+                "local_path": local_path,
+            }
+        )
+
+        error: str | None = None
+        try:
+            self.connection.client.download_file(bucket, key, local_path)
+
+        except Exception as e:  # pylint: disable=W0718
+            error = str(e)
+            logger.error({"metric_filter": "s3_download_error", "error": str(e)})
+
+        file_exist = os.path.exists(local_path)
+
+        logger.debug(
+            {
+                "source": "download_file",
+                "action": "downloading a file from s3",
+                "bucket": bucket,
+                "key": key,
+                "file_name": file_name,
+                "local_path": local_path,
+                "file_downloaded": file_exist,
+                "errors": error,
+            }
+        )
+
+        if not file_exist:
+            raise FileNotFoundError("File Failed to download (does not exist) from S3.")
+
+        return local_path
+
+    def __get_file_name_from_path(self, path: str) -> str:
+        """
+        Get a file name from the path
+
+        Args:
+            path (str): a file path
+
+        Returns:
+            str: the file name
+        """
+        return path.rsplit("/")[-1]
+
+    def get_local_path_for_file(self, file_name: str):
+        """
+        Get a local temp location for a file.
+        This is designed to work with lambda functions.
+        The /tmp directory is the only writeable location for lambda functions.
+        """
+        temp_dir = self.get_temp_directory()
+        # use /tmp it's the only writeable location for lambda
+        local_path = os.path.join(temp_dir, file_name)
+        return local_path
+
+    def get_temp_directory(self):
+        """
+        Determines the appropriate temporary directory based on the environment.
+        If running in AWS Lambda, returns '/tmp'.
+        Otherwise, returns the system's standard temp directory.
+        """
+        if "AWS_LAMBDA_FUNCTION_NAME" in os.environ:
+            # In AWS Lambda environment
+            return "/tmp"
+        else:
+            # Not in AWS Lambda, use the system's default temp directory
+            return tempfile.gettempdir()
+
+    def encode(
+        self, text: str, encoding: str = "utf-8", errors: str = "strict"
+    ) -> bytes:
+        """
+        Encodes a string for s3
+        """
+        return text.encode(encoding=encoding, errors=errors)
+
+    def decode(
+        self, file_obj: bytes, encoding: str = "utf-8", errors: str = "strict"
+    ) -> str:
+        """
+        Decodes bytes to a string
+        """
+        return file_obj.decode(encoding=encoding, errors=errors)
+
+    def list_versions(self, bucket: str, prefix: str = "") -> List[str]:
+        """
+        List all versions of objects in an S3 bucket with a given prefix.
+
+        Args:
+            bucket (str): The name of the S3 bucket.
+            prefix (str, optional): The prefix to filter objects by. Defaults to "".
+
+        Returns:
+            list: A list of dictionaries containing information about each object version.
+        """
+        versions = []
+        paginator = self.connection.client.get_paginator("list_object_versions")
+        page_iterator = paginator.paginate(Bucket=bucket, Prefix=prefix)
+
+        for page in page_iterator:
+            if "Versions" in page:
+                versions.extend(page["Versions"])
+
+        return versions