aws-dynamodb-parallel-scan

aws-dynamodb-parallel-scan

Amazon DynamoDB parallel scan paginator for boto3.

Installation

Install from PyPI with pip

pip install aws-dynamodb-parallel-scan

or with the package manager of choice.

Usage

The library is a drop-in replacement for boto3 DynamoDB Scan Paginator. Example:

import aws_dynamodb_parallel_scan
import boto3

# Create DynamoDB client to use for scan operations
client = boto3.resource("dynamodb").meta.client

# Create the parallel scan paginator with the client
paginator = aws_dynamodb_parallel_scan.get_paginator(client)

# Scan "mytable" in five segments. Each segment is scanned in parallel.
for page in paginator.paginate(TableName="mytable", TotalSegments=5):
    items = page.get("Items", [])

Notes:

• paginate() accepts the same arguments as boto3 DynamoDB.Client.scan() method. Arguments are passed to DynamoDB.Client.scan() as-is.

• paginate() uses the value of TotalSegments argument as parallelism level. Each segment is scanned in parallel in a separate thread.

• paginate() yields DynamoDB Scan API responses in the same format as boto3 DynamoDB.Client.scan() method.

See boto3 DynamoDB.Client.scan() documentation for details on supported arguments and the response format.

CLI

This package also provides a CLI tool (aws-dynamodb-parallel-scan) to scan a DynamoDB table with parallel scan. The tool supports all non-deprecated arguments of DynamoDB Scan API. Execute aws-dynamodb-parallel-scan -h for details

Here's some examples:

# Scan "mytable" sequentially
$ aws-dynamodb-parallel-scan --table-name mytable
{"Items": [...], "Count": 10256, "ScannedCount": 10256, "ResponseMetadata": {}}
{"Items": [...], "Count": 12, "ScannedCount": 12, "ResponseMetadata": {}}

# Scan "mytable" in parallel (5 parallel segments)
$ aws-dynamodb-parallel-scan --table-name mytable --total-segments 5
{"Items": [...], "Count":32, "ScannedCount":32, "ResponseMetadata": {}}
{"Items": [...], "Count":47, "ScannedCount":47, "ResponseMetadata": {}}
{"Items": [...], "Count":52, "ScannedCount":52, "ResponseMetadata": {}}
{"Items": [...], "Count":34, "ScannedCount":34, "ResponseMetadata": {}}
{"Items": [...], "Count":40, "ScannedCount":40, "ResponseMetadata": {}}

# Scan "mytable" in parallel and return items, not Scan API responses (--output-items flag)
$ aws-dynamodb-parallel-scan --table-name mytable --total-segments 5 \
    --output-items
{"pk": {"S": "item1"}, "quantity": {"N": "99"}}
{"pk": {"S": "item24"}, "quantity": {"N": "25"}}
...

# Scan "mytable" in parallel, return items with native types, not DynamoDB types (--use-document-client flag)
$ aws-dynamodb-parallel-scan --table-name mytable --total-segments 5 \
    --output-items --use-document-client
{"pk": "item1", "quantity": 99}
{"pk": "item24", "quantity": 25}
...

# Scan "mytable" with a filter expression, return items
$ aws-dynamodb-parallel-scan --table-name mytable --total-segments 5 \
    --filter-expression "quantity < :value" \
    --expression-attribute-values '{":value": {"N": "5"}}' \
    --output-items
{"pk": {"S": "item142"}, "quantity": {"N": "4"}}
{"pk": {"S": "item874"}, "quantity": {"N": "1"}}

# Scan "mytable" with a filter expression using native types, return items
$ aws-dynamodb-parallel-scan --table-name mytable --total-segments 5 \
    --filter-expression "quantity < :value" \
    --expression-attribute-values '{":value": 5}' \
    --use-document-client --output-items
{"pk": "item142", "quantity": 4}
{"pk": "item874", "quantity": 1}

Development

Requires Python 3 and uv. Useful commands:

# Run tests (integration test requires rights to create, delete and use DynamoDB tables)
make test

# Run linters
make -k lint

# Format code
make format

License

MIT

Credits

• Alex Chan, Getting every item from a DynamoDB table with Python