I know there are no folders in Amazon S3, but we can emulate them by using "/" on the key name. Given that, is it possible using the AWS SDK for Go v2 to calculate the size of a folder? Or do I have to retrieve all objects in the folder and then calculate one by one the size?

CodePudding user response：

Not sure if the easiest way, however you can iterate over your objects list of interest - https://docs.aws.amazon.com/AmazonS3/latest/API/API_ListObjectsV2.html, and aggregate the size locally.

CodePudding user response：

Given that example, and the Object types documentation here

It is possible to compute the size occupied by items within a bucket

package main

import (
    "context"
    "flag"
    "fmt"
    "log"

    "github.com/aws/aws-sdk-go-v2/config"
    "github.com/aws/aws-sdk-go-v2/service/s3"
)

var (
    bucketName      string
    objectPrefix    string
    objectDelimiter string
    maxKeys         int
)

func init() {
    flag.StringVar(&bucketName, "bucket", "", "The `name` of the S3 bucket to list objects from.")
    flag.StringVar(&objectPrefix, "prefix", "", "The optional `object prefix` of the S3 Object keys to list.")
    flag.StringVar(&objectDelimiter, "delimiter", "",
        "The optional `object key delimiter` used by S3 List objects to group object keys.")
    flag.IntVar(&maxKeys, "max-keys", 0,
        "The maximum number of `keys per page` to retrieve at once.")
}

// Lists all objects in a bucket using pagination
func main() {
    flag.Parse()
    if len(bucketName) == 0 {
        flag.PrintDefaults()
        log.Fatalf("invalid parameters, bucket name required")
    }

    // Load the SDK's configuration from environment and shared config, and
    // create the client with this.
    cfg, err := config.LoadDefaultConfig(context.TODO())
    if err != nil {
        log.Fatalf("failed to load SDK configuration, %v", err)
    }

    client := s3.NewFromConfig(cfg)

    // Set the parameters based on the CLI flag inputs.
    params := &s3.ListObjectsV2Input{
        Bucket: &bucketName,
    }
    if len(objectPrefix) != 0 {
        params.Prefix = &objectPrefix
    }
    if len(objectDelimiter) != 0 {
        params.Delimiter = &objectDelimiter
    }

    // Create the Paginator for the ListObjectsV2 operation.
    p := s3.NewListObjectsV2Paginator(client, params, func(o *s3.ListObjectsV2PaginatorOptions) {
        if v := int32(maxKeys); v != 0 {
            o.Limit = v
        }
    })

    // Iterate through the S3 object pages, printing each object returned.
    var i int
    var total int64
    log.Println("Objects:")
    for p.HasMorePages() {
        i  

        // Next Page takes a new context for each page retrieval. This is where
        // you could add timeouts or deadlines.
        page, err := p.NextPage(context.TODO())
        if err != nil {
            log.Fatalf("failed to get page %v, %v", i, err)
        }

        // Log the objects found
        for _, obj := range page.Contents {
            // fmt.Println("Object:", *obj.Key)
            total  = obj.Size
        }
    }
    fmt.Println("total", total)
}

Then, if I am correct, reading at s3.ListObjectsV2Input documentation, it appears to me that you can configure the Prefix member of the s3.ListObjectV2Input instance to select a specific folder. The example already demonstrates that if you pass in the flag -prefix=...