Skip to contents

cloudfs is an R package developed by Gradient Metrics that offers a unified interface for simplifying cloud storage interactions. cloudfs supports uploading, downloading, reading, and writing files for both Google Drive and Amazon S3.

Installation

# from CRAN
install.packages("cloudfs")

# from GitHub
remotes::install_github("g6t/cloudfs")

Key Features

  1. Relative path simplicity

    Use paths relative to your project’s main cloud folder.

    cloud_drive_download("raw_data/transactions.xlsx")
  2. Unified interface for Google Drive and S3

    Downloading from S3? The process is just as straightforward.

    cloud_s3_download("raw_data/transactions.xlsx")
  3. Extension-aware functions

    The package automatically selects the right read or write function based on the file extension, simplifying your interactions.

    cloud_s3_write(glmnet_model, "models/glmnet.rds")

    If you don’t like the default function, you can use a different one by explicitly calling it.

    cloud_s3_write(glmnet_model, "models/glmnet.rds", fun = readr::write_rds)
  4. Effortless cloud navigation

    Open folders in browser

    or list contents in console.

    cloud_s3_ls("data")
  5. Bulk file management

    Easily retrieve all data from a folder in one go

    all_data <- 
      cloud_s3_ls("data") %>%
      cloud_s3_read_bulk()

    or push multiple files to the cloud at once.