Development of Dantalian has been somewhat bumpy as I was figuring out how exactly it should be and what it should do, as well as figuring how best to structure the project in terms of documentation, websites, files, and info. Now that I have most of it figured out and Dantalian is relatively stable, here are my plans for the future:
-
Add caching to Dantalian. With large libraries, doing tag searches becomes costly time-wise, and the cost is unavoidable. However, the searching can be done in advance and cached, with the downside of requiring regular updating and not always being up to date. The dynamic is similar to that of
findandlocatein GNU findutils. -
Move FUSE features to a separate application, probably written in Go. Writing such a service in Python is unacceptable performance-wise, especially with CPython’s poor threading.
Those are the most pressing and clearest goals, but there are some
other minor things as well (such as implementing find in Python to
remove the dependency on findutils and more comprehensive tests).