Closed farooqkz closed 1 year ago
You can use busted or mineunit for testing mods.
I don't see how this differs from #13070 which was closed.
I don't see how this differs from #13070 which was closed.
I believe I didn't well explain my idea in that issue. And it was more focused on creating bots. Also ruben didn't explain to me why it was closed. Perhaps I had to comment on that issue instead of creating a new one?
@rubenwardy
TBH, I still don't understand why you are not going to implement this? Or are you proposing that another client must be built from scratch with rendering, mods and everything just to be programmable?
Minetest is open source, you can fork the engine and implement this yourself - as a client-side modding API or your socket suggestion. Dragonfire is an example of a fork of Minetest with more client-side modding features, it probably allows you to control player movement
We intentionally have not added :set_pos or an API to control player movement to client-side modding as allowing players to script their own movement isn't a goal of Minetest
We don't have time to implement everything, we need to prioritise and our priorities are with giving more control to games and mods.
This issue is definitely a duplicate - you could ask another core developer to give their opinion on #13070
Duplicate of #13070
Minetest is open source, you can fork the engine and implement this yourself - as a client-side modding API or your socket suggestion. Dragonfire is an example of a fork of Minetest with more client-side modding features, it probably allows you to control player movement
We intentionally have not added :set_pos or an API to control player movement to client-side modding as allowing players to script their own movement isn't a goal of Minetest
We don't have time to implement everything, we need to prioritise and our priorities are with giving more control to games and mods
If I implement it myself, will it get merged into upstream?
Asking for opinion from @celeron55
I think what I am requesting is related with #4064
sfan5's view here: https://irc.minetest.net/minetest-dev/2023-03-13#i_6063012
If I implement it myself, will it get merged into upstream?
No, implementation is usually less than half the work - most of the work is in maintenance and keeping things working. Implementing it yourself in a fork would allow you to use it without us incurring the cost of maintenance
sfan5's view here: https://irc.minetest.net/minetest-dev/2023-03-13#i_6063012
If I implement it myself, will it get merged into upstream?
No, implementation is usually less than half the work - most of the work is in maintenance and keeping things working. Implementing it yourself in a fork would allow you to use it without us incurring the cost of maintenance
@sfan5 @rubenwardy
You are absolutely right about maintenance costs. But my point is that with this not so big feature, a lot of things will be possible which now require forking the client and adding to the engine. A brand new plugin system will be introduced for Minetest Client which basically a lot of things would be possible with it and it will be language agonstic.
Automatic mod testing is just one small example.
Problem
Testing mods is actually a very bad experience, since you have to test everything yourself. Also to my knowledge, Minetest does not support plugins. E.g. a voice typing plugin or a shortcut to send something specific to the server.
Solutions
Additional context
Implementation of this feature will open a big door. Possible use cases could be: