A science-focused, more humane R interface to AWS.
Installation
CRAN version
# install.packages("pak")
pak::pak("sixtyfour")Development version
# install.packages("pak")
pak::pak("getwilds/sixtyfour")Load package
Authentication
We leverage the package paws to handle AWS
authentication. There are many ways to configure paws to
authenticate with AWS - see paws
documentation.
One way to authenticate is using these three environment variables:
AWS_ACCESS_KEY_IDAWS_SECRET_ACCESS_KEYAWS_REGION
You can set these three environment variables within R for the current R session like:
Sys.setenv(
AWS_ACCESS_KEY_ID = "",
AWS_SECRET_ACCESS_KEY = "",
AWS_REGION = "us-west-2"
)Or set environment variables in a variety of ways to be available across R sessions. See the R Startup chapter of What They Forgot to Teach You About R book for more details.
Package API Overview
-
aws_billing*: manage AWS billing details -
aws_bucket*: manage S3 buckets -
aws_file_*: manage files in S3 buckets -
aws_user*: manage AWS users -
aws_group*: manage AWS groups -
aws_role*: manage AWS roles -
aws_policy*/aws_policies*: manage AWS policies -
aws_db*: interact with AWS database services Redshift and RDS -
aws_secrets*: secrets manager -
aws_vpc*: VPC security groups -
con_*: connection classes to various AWS services -
six_*: magical methods
Working with S3
We cover S3 in detail in another vignette
(vignette("s3")), so we’ll briefly touch on it here.
Buckets
Make a random bucket name
bucket <- random_bucket()Create a bucket - check if it exists first
exists <- aws_bucket_exists(bucket)
if (!exists) {
aws_bucket_create(bucket)
}
#> [1] "http://bucket-lphnzotejwfyuxkm.s3.amazonaws.com/"
aws_bucket_delete(bucket)Files
First, create a bucket:
my_bucket <- random_bucket()
aws_bucket_create(my_bucket)
#> [1] "http://bucket-cdzsxyjiofheptnr.s3.amazonaws.com/"Then, upload some files
temp_files <- replicate(n = 3, tempfile(fileext = ".txt"))
for (i in temp_files) cat(letters, "\n", file = i)
remote_files <- s3_path(my_bucket, basename(temp_files))
aws_file_upload(path = temp_files, remote_path = remote_files)
#> [1] "s3://bucket-cdzsxyjiofheptnr/file16cf21ea161d1.txt"
#> [2] "s3://bucket-cdzsxyjiofheptnr/file16cf2751cf00c.txt"
#> [3] "s3://bucket-cdzsxyjiofheptnr/file16cf23ed3dfe5.txt"List files in the bucket
obs <- aws_bucket_list_objects(my_bucket)
obs
#> # A tibble: 3 × 8
#> bucket key uri size type etag lastmodified storageclass
#> <glue> <chr> <glu> <fs:> <chr> <chr> <dttm> <chr>
#> 1 bucket-cdzsxyj… file… s3:/… 53 file "\"a… 2025-03-20 20:52:34 STANDARD
#> 2 bucket-cdzsxyj… file… s3:/… 53 file "\"a… 2025-03-20 20:52:34 STANDARD
#> 3 bucket-cdzsxyj… file… s3:/… 53 file "\"a… 2025-03-20 20:52:34 STANDARDBilling
Billing is covered in the vignette("billing")
vignette.
IAM
IAM stands for Identity Access Management. IAM covers four resource types:
- Users: via
aws_user*functions - Groups: via
aws_group*functions - Roles: via
aws_role*functions - Policies: via
aws_policy*/aws_policies*functions
As a brief example let’s create a group, a user, and a role. We’ll assign the user to the group. Then clean it all up.
First, create a group
group_name <- random_string("gr-")
group_name
#> gr-gmpcjatq
aws_group_create(group_name)
#> # A tibble: 1 × 5
#> GroupName GroupId Path Arn CreateDate
#> <chr> <chr> <chr> <chr> <dttm>
#> 1 gr-gmpcjatq AGPA22PL7JXXYJ5WLSESJ / ***** 2025-03-20 20:52:35Then create a user
name <- random_user()
name
#> [1] "RepellentFinale"
aws_user_create(name)
#> # A tibble: 1 × 6
#> UserName UserId Path Arn CreateDate PasswordLastUsed
#> <chr> <chr> <chr> <chr> <dttm> <dttm>
#> 1 RepellentFinale AIDA22PL7… / ***** 2025-03-20 20:52:35 NANow add the user to the group
aws_user_add_to_group(name, group_name)
#> $user
#> # A tibble: 1 × 6
#> UserName UserId Path Arn CreateDate PasswordLastUsed
#> <chr> <chr> <chr> <chr> <dttm> <dttm>
#> 1 RepellentFinale AIDA22PL7… / ***** 2025-03-20 20:52:35 NA
#>
#> $policies
#> # A tibble: 0 × 0
#>
#> $attached_policies
#> # A tibble: 0 × 0
#>
#> $groups
#> # A tibble: 1 × 5
#> GroupName GroupId Path Arn CreateDate
#> <chr> <chr> <chr> <chr> <dttm>
#> 1 gr-gmpcjatq AGPA22PL7JXXYJ5WLSESJ / ***** 2025-03-20 20:52:35Now create a role
role_name <- random_role()
trust_policy <- list(
Version = "2012-10-17",
Statement = list(
list(
Sid = "Statement1",
Effect = "Allow",
Principal = list(
Service = "lambda.amazonaws.com"
),
Action = "sts:AssumeRole"
)
)
)
doc1 <- jsonlite::toJSON(trust_policy, auto_unbox = TRUE)
aws_role_create(role_name,
assume_role_policy_document = doc1,
description = "test role A"
)
#> # A tibble: 1 × 7
#> RoleName RoleId Path Arn CreateDate Description
#> <chr> <chr> <chr> <chr> <dttm> <lgl>
#> 1 BridgedLiquidati AROA22PL7JXX5ETK… / ***** 2025-03-20 20:52:36 NA
#> # ℹ 1 more variable: AssumeRolePolicyDocument <chr>Finally, cleanup
aws_role_delete(role_name)
six_user_delete(name)
#> ! No access keys found for RepellentFinale
#> ℹ Group gr-gmpcjatq detached
#> ℹ RepellentFinale deleted
six_group_delete(group_name)
#> ℹ group gr-gmpcjatq deletedDatabases
Databases are covered in the vignette("databases")
vignette.
Magical methods
Magical methods are higher level functions to make your life easier.
As an example, six_user_create() creates a new user,
including a) creating the appropriate policy if it doesn’t exist yet, b)
attaching the policy to the created user, and c) creating an AWS Access
Key for the user (and copy an email template to your clipboard with the
details the new user needs).
name <- random_user()
name
#> [1] "GildedArchery"
six_user_create(name)
#> ℹ Added policy UserInfo to GildedArchery
#> ✔ Key pair created for GildedArchery
#> ℹ AccessKeyId: *****
#> ℹ SecretAccessKey: *****
#> ℹ Email template copied to your clipboardThere are other magical methods. See the docs
and the vignette("six") vignette for more information.