[Python / DynamoDB / boto3] List of operations I tried

Overview

** Minimize the number of lines and characters as much as possible **, and cover (want to) how to operate DynamoDB from Python

• Reason ** I don't want to read the document one by one ** ← ** It is troublesome to find the necessary information from other people's articles every time ** ←←← (terrible)

I will write from what I actually tried (\ * ˙︶˙ \ *) و Good!

Operation list

Preparations common to all operations

import boto3
from boto3.dynamodb.conditions import Attr, Key

# region_For name, write the name of your region
dynamodb = boto3.resource('dynamodb', region_name='us-east-2')

1. create_table () ... create table

Create table

table_name = 'hoge_table'

table = dynamodb.create_table(
    TableName = table_name,
    KeySchema = [
        {'AttributeName': 'primary_key_name', 'KeyType': 'HASH'},
        {'AttributeName': 'column_name', 'KeyType': 'RANGE'}
    ],
    AttributeDefinitions = [
        {'AttributeName': 'primary_key_name', 'AttributeType': 'N'},
        {'AttributeName': 'column_name', 'AttributeType': 'S'}
    ],
    ProvisionedThroughput = {
        'ReadCapacityUnits': 1, 'WriteCapacityUnits': 1
    }
)
table.meta.client.get_waiter('table_exists').wait(TableName=table_name)

• KeySchema --Specify column name with AttributeName 'primary_key_name' and'column_name' are just sample names> _ <; --'KeyType':'HASH' will be the partition key --'KeyType':'RANGE' is the sort key
• AttributeDefinitions --'AttributeType':'S' makes it a string type column --'AttributeType':'N' makes it a numeric column

2. list_tables () ... Get table list

Get table list

# region_For name, write the name of your region
client = boto3.client('dynamodb', region_name='us-east-2')

client.list_tables()['TableNames']

3. get_item () ... Get 1

Get 1 record

table_name = 'hoge_table'
table = dynamodb.Table(table_name)

table.get_item(
    Key={'primary_key_name': 1, 'column_name': 'string_value'}
).get('Item')

4. scan () ... Filtering after fetching all

After getting all items, filter by Attribute It is said that it consumes extra resources compared to query () (AWS usage fee is extra)

Filtering after getting all

table_name = 'hoge_table'
table = dynamodb.Table(table_name)

table.scan(
    FilterExpression=Attr('column_name').eq('string_value'),
    Limit=1 #You can specify the number of cases. There is no need to have one separately.
).get('Items')

5. query () ... Conditional search

Get only records that meet the conditions

filtered search

table_name = 'hoge_table'
table = dynamodb.Table(table_name)

#Column name is appropriate
table.query(
    KeyConditionExpression=Key('column_name1').eq('string_value') & Key('column_name2').between(from_str, to_str)
)

(I want to start a new line at some point, but I'm angry with the Lint tool ...)

6. batch_writer () ... insert all at once

table_name = 'hoge_table'
table = dynamodb.Table(table_name)
items_source = [
    {'primary_key_name': 1, 'column_name': 'hoge1'},
    {'primary_key_name': 2, 'column_name': 'hoge2'},
           ...(Abbreviation)
    {'primary_key_name': 9, 'column_name': 'hoge9'}
]

# `overwrite_by_pkeys`For the column specified in, if it is duplicated, the record will be overwritten.
#If this is not specified, an error will be thrown when the value is duplicated.
with table.batch_writer(overwrite_by_pkeys=['primary_key_name', 'column_name']) as batch:
    for item in items_source:
        batch.put_item(Item=item)

Referenced articles

-I tried to operate DynamoDB with Boto3 -Create table and register data in DynamoDB Local with Python (boto3)