Scopes #

A workspace can contain a large number of packages, but often only a few are relevant at any given point in time.

Many dryad commands offer some combination of filters (like --include and --exclude) to select a specific part of a workspace to operate on. But, it can be annoying to remember and typeextra arguments all the time.

To make it easier to logically partition parts of a workspace, dryad offers the ability to create shortcut commands through scopes. A scope is a collection of command or argument “aliases”.

Most dryad commands have a --scope option, which allows you to specify the scope to run a command in. When a scope is provided that specifies arguments for that command, dryad will_rewrite the command arguments before execution, using the scoped settings for that command.

You can also set a default scope using dryad scopes default set. If no scope is provided, dryad will automatically use the default scope.

The scope name none is reserved to use as an escape hatch from scoping. Using --scope=none will bypass the default scope and run the command as-is.

Scripts #

Dryad scope settings are by default plain-text files that include extra arguments to add to a command. The one exception is the dryad run <name> / dryad script run <name> command. This command interprets the corresponding scope setting as a standalone script to be executed, which allows you to add any commands you’d like to dryad run. Keep in mind these scripts are scoped to the current scope, so you can use this to do things like create two different dryad run build commands for two different scopes.

You can create a scoped script using the dryad script edit <name> command, and see which scoped scripts are available to you using dryad scripts list.