How to pull a list of items from a DynomoDB table without using primary key in Go.

0

Below is a single item in JSON, there are hundreds more. The account_name and ami_name are the partition key and sort key. I want to be able to query a list based on the ami_name and filter that list by the environment value, in this case, "test". I have an example of what I am trying so far.



package main

import (
    "context"
    "fmt"
	"log"
    "github.com/aws/aws-sdk-go-v2/aws"
    "github.com/aws/aws-sdk-go-v2/config"
    "github.com/aws/aws-sdk-go-v2/service/dynamodb"
    "github.com/aws/aws-sdk-go-v2/service/dynamodb/types"
)

func main() {
    cfg, err := config.LoadDefaultConfig(context.TODO(), config.WithRegion("us-east-1"))
    if err != nil {
        log.Fatalf("unable to load SDK config, %v", err)
    }

	tableName   := "aws-webhooks-table-test"
	// account_name := "aws-computer-test"
	ami_name     := "win2019core"
    environment := "test"
    svc := dynamodb.NewFromConfig(cfg)
    out, err := svc.Query(context.TODO(), &dynamodb.QueryInput{
        TableName:              aws.String(tableName),
        KeyConditionExpression: aws.String("ami_name = :ami_name"),
		FilterExpression: aws.String("environment = :environment"),
        ExpressionAttributeValues: map[string]types.AttributeValue{
            ":ami_name":  &types.AttributeValueMemberS{Value: ami_name},
			":environment":  &types.AttributeValueMemberS{Value: environment},
        },
        ExpressionAttributeNames: map[string]string{
            "ami_name": ami_name,
        },
}

{
  "account_name": {
    "S": "aws-computer-test"
  },
  "ami_name": {
    "S": "win2019core"
  },
  "account_id": {
    "S": "7666666666"
  },
  "aws_region": {
    "S": "us-east-1"
  },
  "environment": {
    "S": "test"
  },
  "lob": {
    "S": "computer"
  },
  "payload_url": {
    "S": "https://codebuild.us-east-1.amazonaws.com/webhooks?t=eyJlbmNyeXBlotsoflettersKPIkeo"
  },
  "project_name": {
    "S": "aws-computer-test-win2019-ami-project"
  },
  "repo_url": {
    "S": "https://git.nylcloud.com/Cloud-Team/packer-aws-win2019core"
  },
  "secret": {
    "S": "dq7bHlotsofletterZKPIkeo"
  }
}
已提問 1 年前檢視次數 4380 次
2 個答案
2

In DynamoDB you must always provide the Partition Key when reading items, unless using Scan operation which will read every item in the table and is best to be avoided if possible.

For your use-case you would need to create a Global Secondary Index (GSI) and specify ami_name as the partition key. Optionally you can use environment as your sort key.

Now you would target the GSI and use a Query operation to read the data (with partition key only):

    out, err := svc.Query(context.TODO(), &dynamodb.QueryInput{
        TableName:              aws.String(tableName),
        IndexName:              aws.String(indexName),
        KeyConditionExpression: aws.String("ami_name = :ami_name"),
	FilterExpression: aws.String("environment = :environment"),
        ExpressionAttributeValues: map[string]types.AttributeValue{
            ":ami_name":  &types.AttributeValueMemberS{Value: ami_name},
	    ":environment":  &types.AttributeValueMemberS{Value: environment},
        },
        ExpressionAttributeNames: map[string]string{
            "ami_name": ami_name,
        },
}

If you choose to use environment as sort-key on your index then you would include that in your KeyConditionExpression and remove the FilterExpression.

profile pictureAWS
專家
已回答 1 年前
-2

Yes you can use the pagination method which uses 'Scan' operation, however this operation will read every item in the table and is best to be avoided if possible due to two reasons :

  1. Increase in cost.
  2. Time consuming.

Therefore you must always provide the Partition Key when reading items.

AWS
支援工程師
已回答 1 年前

您尚未登入。 登入 去張貼答案。

一個好的回答可以清楚地回答問題並提供建設性的意見回饋,同時有助於提問者的專業成長。

回答問題指南