Read from Amazon S3 — tl_read

Downloads a file from an S3 bucket and reads it into a tidylearn_data object. The file format is auto-detected from the key's extension, or can be specified explicitly. Requires the paws.storage package and valid AWS credentials.

Usage

tl_read_s3(source, format = NULL, region = NULL, ..., trust_rds = FALSE)

Arguments

source: An S3 URI (e.g., "s3://bucket/path/to/file.csv"). Zip archives are not read from S3: download the object and read it with tl_read_zip().
format: Optional format override for the downloaded file. If NULL, auto-detected from the S3 key extension.
region: AWS region. If NULL, uses the default from your AWS configuration.
...: Additional arguments passed to the format-specific reader.
trust_rds: Logical. Read an .rds, .rdata or .rda object on R older than 4.4.0? Those versions can run code embedded in a crafted file as it is read (CVE-2024-27322), so such objects are refused there unless this is TRUE. It has no effect on R 4.4.0 or later. Default FALSE.

Value

A tidylearn_data object containing the downloaded data.

Reading R serialisation files

An .rds or .rdata object is rebuilt with readRDS() or load(), which recreate whatever R objects the file describes. Read them only from a bucket you trust, on any version of R.

Examples

if (FALSE) { # \dontrun{
# Needs AWS credentials
tl_read_s3("s3://my-bucket/data/sales.csv")
tl_read_s3("s3://my-bucket/data/results.parquet", region = "eu-west-1")
} # }