Finalize outstanding Sector topics before merge of v2 to develop #3144
Labels
Category: Doc
Requests, Issues and Changes targeting javadoc and module documentation
Status: Needs Investigation
Requires to be debugged or checked for feasibility, etc.
Topic: Architecture
Requests, Issues and Changes related to software architecture, programming patterns, etc.
Akin to #3143, the other large piece of v2.0.0 as it stands right now is @Vizaxo's item on Sectors, which also introduces zones and tweaks at entity scope. While a lot of that item is foundation work for future enhancements like multi-world and truly independent entity + world pools that won't really matter to the current API level there are some outstanding issues and changes that might cause trouble or require follow-ups in the near term that could trigger API violations.
The new behavior tree system (see #3143) came with a bunch of PRs for affected modules so merging everything together should leave nothing broken. But far fewer modules use BT in the first place so being sure everything is fixed is easier than what Sectors and especially the entity scope change may affect. For instance the magic dome in Light & Shadow uses
EntityRef.setAlwaysRelevant
(inMagicDomeSystem
) which changes with Sectors. It does look like that part has been updated and should work now, but I don't think we've gone looking exhaustively for examples.My v2 test workspace for GSOC didn't really grow to cover a lot of modules beyond what was actively worked for that project, so if nothing else we should try the v2 branch with a lot more modules that may be affected. All the world generator and world plugin modules in particular. Beyond that I'll list some other potential things to review.
inspectionTool.prefab
from Core) - maybe L&S would be an especially interesting test when combined with BTs and its changes to L&S making the creatures functional again (what happens when they unload or try to pathfind into unloaded chunks? Would they interact any with the dome?)Much like the other big issue like this one (#3143) it may be no action is needed on this before the v2 merge, but we at least should get another round of review in on it so we can be more confident. So marking blocker for the v2 release yet in a mild fashion as it could be a non-issue.
Some reference material:
The text was updated successfully, but these errors were encountered: