Directory Structure

Why do we need it?

As we wanted the end-user to be able to work from everywhere into the filesystem, we created a logic which will create and keep the output/ directory which complies with our code.

How does it work?


Want to read the code ? It’s here Constructor()!

After each version, the maintainer does a --production which will prepare the repository for production. That has the side effect to map the maintainer version of the output/ directory into a file called dir_structure_production.json.

Once pushed, on the end-user side, when testing for file, that file is downloaded into a file called dir_structure.json which is then used to restore/create a a perfect copy of the output directory the maintainer had when pushing the new version.


If you find yourself in a case that a directory is not found, please try first to delete the dir_structure*.json files to force a resynchronization.

How to generate it manually?

You can’t. But using the --dir-structure argument will do the job on purpose.