orderly.sharedfile

This package contains a small plugin for using files from some shared location within orderly2. Unlike orderly2::orderly_shared_resource, the assumption here is that the files are somewhere out of the source tree, for example a mounted shared drive that multiple people access.

We expect that the files are to be read into a running task/report are read only and are a source of information, not a destination to write things to.

Configuration

This may change as and when we overhaul orderly2’s configuration; we would like to entirely remove the yaml, and are open to suggestions.

Edit your orderly_config.yml to add a section like

plugins:
  orderly.sharedfile:
    database:
      path: /path/to/database

where /path/to/database is the full (absolute) path to the files that you want to use from within orderly tasks/reports. As a shorthand you could also write database: /path/to/database.

You can have more than one entry here, for example:

plugins:
  orderly.sharedfile:
    malaria: /path/to/malaria
    dengue: /path/to/dengue

uses the shorthand form to set up two shared locations, one called malaria and the other dengue.

If you are not the only person that is going to be running things that use this shared resource, you will need to make it configurable per-user. The easiest way to do this is to set paths within a file orderly_envir.yml (which should be excluded git) and reference environment variables from here within the configuration. So for example you might have an orderly_envir.yml that contains:

PATH_MALARIA: /path/to/malaria
PATH_DENGUE: /path/to/dengue

and then within the orderly_config.yml:

plugins:
  orderly.sharedfile:
    malaria: $PATH_MALARIA
    dengue: $PATH_DENGUE

Usage

After the plugin is configured as above, you can use it from a task/report:

path_db <- orderly.sharedfile::sharedfile_path(
  "database/v1.mdb", from = "malaria")

The from argument is optional if you only have a single shared location configured. After running this statement, the variable path_db contains the absolute path to your database and you can pass this your db driver.

You might use a similar approach to access large shapefiles, genomic sequences, etc.

You can return vectors of paths, for example:

path <- orderly.sharedfile::sharedfile_path(c("a", "b"))

will return a vector of length two, with the paths for files a and b in a configuration that uses only a single shared directory.

Metadata

We save the hash of all files found, along with their full names. If the returned path is a directory, all contents are hashed.

Installation

To install orderly.sharedfile:

remotes::install_github("mrc-ide/orderly.sharedfile", upgrade = FALSE)

License

MIT © Imperial College of Science, Technology and Medicine