As of 1.0, the logic to check the custom files/scripts directories to see if they are empty occurs in the actual build step instead of validation. There's no real driving reason for this; it was largely missed in the last major validation rework which was primarily focused on the definition itself.
It's not entirely trivial to migrate this into the validation, which exists in the image package and, as previously stated, focuses more on the definition. The logic for what these sorts of directories are live in the combustion-specific component areas and I'm hesitant to introduce that dependency outside of a larger post-1.0 architecture revisit (we already know of a number of things that we need to rethink given new information, in which case this may become moot).
This issue tracks the movement of that logic, along with the logic checking the rpms directory for emptiness. The latter directly stems from #242 which will follow the current model the custom files uses when its fix lands.
As of 1.0, the logic to check the custom files/scripts directories to see if they are empty occurs in the actual build step instead of validation. There's no real driving reason for this; it was largely missed in the last major validation rework which was primarily focused on the definition itself.
It's not entirely trivial to migrate this into the validation, which exists in the image package and, as previously stated, focuses more on the definition. The logic for what these sorts of directories are live in the combustion-specific component areas and I'm hesitant to introduce that dependency outside of a larger post-1.0 architecture revisit (we already know of a number of things that we need to rethink given new information, in which case this may become moot).
This issue tracks the movement of that logic, along with the logic checking the
rpms
directory for emptiness. The latter directly stems from #242 which will follow the current model the custom files uses when its fix lands.