-
Notifications
You must be signed in to change notification settings - Fork 73
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Modularize dependencies #238
Comments
Unfortunately it is not possible to specify default features in python: pypa/setuptools#1139 So if the dependencies are modularized, it would be a breaking change and dependent projects would have to specify all of the features that they need. |
Yes, I'm sorry about this. I should have made the RPC dependencies, pyperclip, pynput and others optional from the start. One way I could see around this problem is to have a separate |
I would be in support of this as well. Ultimately would help diversify Dragonfly use case to different applications. It would be nice not to have to split releases but it looks like we are a constrained with packaging technical limitations. |
I'm glad that this is something that's open to being fixed. Another idea might be to split dragonfly into a bunch of sub packages and publish those separately, like dragonfly_grammar, dragonfly_kaldi, etc. Just an idea. |
Sorry for the late response! It's certainly possible to split Dragonfly into a bunch of sub-packages to be published separately. Some of us are considering splitting Dragonfly's actions and related OS control code into a separate dictation-toolbox repository with similar code from Unimacro. |
I believe modularisation of dependencies should be done for Dragonfly version 1.0.0. |
I'm writing a program that only needs the grammar and kaldi backend and it would be nice if I only had to pull in dependencies for those two features. Kaldi is already hidden behind a feature flag which is nice, but I still don't need rpc features, pyperclip, and pynput. I think it would be nice if the sub packages were modularized behind feature flags. Thanks!
The text was updated successfully, but these errors were encountered: