Skip to main content
A table-valued function (TVF) that lists objects and their metadata from Amazon S3. The function can use either a location object (recommended) or direct credentials to access the data. Each row in the output contains four columns: object_name (TEXT), object_type (TEXT), object_bytes (BIGINT), and file_timestamp (TIMESTAMPTZ).

Syntax

Parameters

Return Type

The output is a table with four columns:
  • object_name (TEXT)
  • object_type (TEXT)
  • object_bytes (BIGINT)
  • last_modified (TIMESTAMPTZ)
Column Descriptions
  • object_name: Contains both the full path and the file extension.
  • object_type: Can be either “file” or “folder”.
    • If object_type = “folder”, the object_bytes and last_modified columns will contain NULL values, as folders do not have associated sizes or timestamps.
    • If object_type = “file”, the following apply:
      1. The last_modified column is populated from the LastModified attribute in Amazon S3. Note that AWS does not expose the creation timestamp, so the values in this column only differ from the creation time if an immutable object has been overwritten.
      2. The object_bytes column contains the size of the file in bytes.
Amazon S3 is not a traditional filesystem. In AWS, what is commonly referred to as “folders” is called “common_prefixes,” and what are typically considered “files” are referred to as “objects.”

Examples

Using LOCATION object to store credentials

Best practice Firebolt recommends using a LOCATION object to store credentials for authentication. The following code example retrieves all objects from the specified LOCATION using the LIST_OBJECTS function:
Returns:

Using static credentials

The following code examples show how to list all objects (folders) that start with a specified prefix within an Amazon S3 bucket. You can specify either a part of the prefix or the full prefix. For example, the URLs ending with the prefixes fire or firebolt_sample_dataset both return identical results because both are valid matches for the firebolt_sample_dataset folder as follows:
and
Both return: The following code example shows how to list all objects such as files, folders, and associated metadata, that start with a specified prefix in an Amazon S3 bucket:
Returns: The following code examples show how to list all objects such as files and associated metadata, that start with a specified prefix in an Amazon S3 bucket. The URLs ending with the prefixes lev or levels.csv return identical results because both are valid matches for the levels.csv file as follows:
and
Both Return: The following code example shows how to use your Amazon credentials to list objects in an Amazon S3 bucket that is not publicly accessible:
Returns: