Utility Function get_all_project_files(): Integration and Examples

2024-11-01

The get_all_project_files() function provides an efficient way to retrieve detailed information about all files within a specified directory on the Seven Bridges File System (SBFS). This utility function is ideal for preparing data frames that can be integrated with the File Picker Module, enhancing file selection and management in Shiny applications.

Function Overview

This function fetches all files from a given directory path recursively, including their metadata if available. It constructs a comprehensive data frame containing file names, paths, sizes, and associated metadata, making it ready for use with the file picker module.

Parameters:

• path: The path to the project-files directory. This is a required parameter specifying the location from which to retrieve file information.

• …: Additional parameters passed to the list.files() function. Options such as pattern and include.dirs can be used to customize the file search.

The function returns a comprehensive data frame containing file names, paths, sizes, and associated metadata (if available), making it ready for use with the file picker module.

Note on Dependencies

The get_all_project_files() function depends on the xattrs packagefor extended attribute support. This package is not installed automatically with *sbShinyModules, so you will need to install xattrs separately if you wish to use this function.

On Unix-based systems, you can install xattrs using:

remotes::install_github("hrbrmstr/xattrs")

If you encounter installation issues, it may be due to missing system libraries. For instance, on Ubuntu Linux, you can resolve this by installing the libattr1-dev library with:

sudo apt-get update
sudo apt-get install libattr1-dev

For macOS users, additional system libraries are typically not required for xattrs, but ensure your system is up-to-date.

Please note that the get_all_project_files() function is not available on Windows systems. This is because it depends on the xattrs package, which requires extended attribute functions that are not supported on Windows.

Integration with File Picker Module

The utility function for loading project files and their metadata get_all_project_files() is intended to be used when developing Shiny apps within Data Studio, or to be deployed as on-demand Shiny apps using Data Studio infrastructure.

Data Studio File Organization

Data Studio environments (RStudio or Jupyter) come equipped with the Seven Bridges File System (SBFS) tool installed, allowing you to access their project files right from the IDE/Data Studio. This means that project files are mounted on the local file system and can be imported into your analysis without any download or upload.

The mounted directory is called sbgenomics and contains these subdirectories:
- project-files
- output-files - workspace

The /project-files subdirectory contains all your project files (the project from which you started the application).

However, the /output-files subdirectory aims to “save” the files you wish to write/export from your analysis/application into your project; therefore, you would have “write” permissions here. The files you store here will be exported to the project files once you stop your analysis/application.

The /workspace subdirectory can serve as a working directory where you can store temporary files produced during the app execution, which you do not want to export. The content of this subdirectory will be saved and available the next time the app is initialized, but we strongly suggest not storing anything here.

This means that you should assume the existence of /sbgenomics/project-files directory which represents the mounted directory on the reserved instance for running Data Studio analysis (in our case - Shiny app). This directory contains all project files that can be accessed and used within the app.

The data frame produced by get_all_project_files() can be directly used as an input for the files_df parameter of the file picker module (check vignette("file-pickers") for more details). This streamlines the process of populating the file picker with accurate and up-to-date file information.

By utilizing get_all_project_files() function, you can efficiently gather and prepare file information, enhancing the functionality of Shiny applications that involve file management and selection.

Note: To run this example, ensure that both the sbShinyModules and xattrs packages are installed. Please be aware that this will not work on Windows systems.

# Create a mock directory on the sbgenomics/project-files path and populate it
# with some test fles to test the get_all_project_files() function
all_files_df <- sbShinyModules::get_all_project_files(
  path = "/sbgenomics/project-files"
)

head(all_files_df)