The repository management can be used to load all metadata from a repository, analyze them, and provide a bunch of information about the whole repository state.
Since loading all metadata from a repository is not a light task, this engine should only be used on a machine having good access to the repository (on the same filesystem being usually the best suited).
To access information, you usually have before to call a method to init the data: {@link #load()}is used to load repository metadata, {@link #analyze()} is used to analyze them. These methodsbeing very time consuming, they must always be called explicitly.
On a large repository, this engine can be very memory consuming to use, it is not suited to be used in a long running process, but rather in short process loading data and taking action about the current state of the repository.
This engine is not intended to be used concurrently with publish, the order of repository loaded being undeterministic and long, it could end up in having an inconsistent in memory state.
For better performance, we strongly suggest using this engine with cache in useOrigin mode.