If you are storing files from several projects inside a unique repository from where you take the data, the hash function, will apply the same hash to every project, this due to how the assigning is done:
Only if the name is CodeNet will work because takes the second part of the name, meanwhile in every other case it will take only the first part of the path which will be the name of the repository not the name of the projects stored inside as it would be for the CodeNet case.
If you are storing files from several projects inside a unique repository from where you take the data, the hash function, will apply the same hash to every project, this due to how the assigning is done:
Only if the name is CodeNet will work because takes the second part of the name, meanwhile in every other case it will take only the first part of the path which will be the name of the repository not the name of the projects stored inside as it would be for the CodeNet case.