[`CaseInsensitiveCommands`] Casexile, but a Tweak!

[KazWolfe]

Someone should probably verify this can be "adopted"... Anyways.

• Uses the same failed command handler logic as Dalamud does. This should always run post-Dalamud.
• Instead of just forcing the first word to lowercase, intelligently look up the command from Lumina.
  • This only makes a difference for GM commands, but hey. It's technically more correct!
  • It may make sense (possibly?) to cache these resolutions to not run a LINQ query every time. However, this isn't running per-frame so I'm not sure if it's a big deal.
• Add optional resolution for Dalamud commands as well, so people can use /XLPlugins and all.
  • This will also intelligently search through Dalamud's command registry, meaning plugins that register uppercase commands will be used if found first. (e.g. a plugin registering /CoolCommand will now be accessible via /coolcommand).

Command resolution order is: as-is command, case-insensitive game command, case-insensitive plugin command, default error response. Meaning:

• The first match for any case-insensitive search is found, so if two plugins register /someCommand and /SomeCommand, the earlier one registered will take priority when using /somecommand.
• If a plugin registers /MountGuide, /MOUNTGUIDE will always resolve to the game's variant.
  • If the user calls /MountGuide, it'll still be handled by the plugin since it's correct case.

I do not know how this plugin will interact with other plugins that perform their own command hooking; I suspect load order would kick in. So if, e.g., a plugin uses the same hook to listen for /sam for class-switch purposes, I'm not sure this one will necessarily resolve to that if the user types in /SAM.

[Caraxi]

I'll take a look at this in a couple weeks (maybe) once things are stable. Don't want a hotfix delayed because of a new tweak in the code lol

[KazWolfe]

Makes perfect sense. I'd like to give @RyouBakura some grace time to bring their plugin to API11 - or give consent for this migration as-is. Don't think this is anything critical, so it can hang around for a bit.

[RyouBakura]

Hey. I'm happy that you want to adopt this. Feel free to use any code that seems useful to you: https://gitlab.com/Bakura/casexile/-/blob/master/SlashCommandInterceptor.cs?ref_type=heads

Makes perfect sense. I'd like to give @RyouBakura some grace time to bring their plugin to API11 - or give consent for this migration as-is. Don't think this is anything critical, so it can hang around for a bit.

Casexile is broken since DT because the repo in which the XivCommon dependency is hosted doesn't accept PRs. And I didn't think it would be appropriate to ship Casexile with my own updated version of XivCommon because some poor maintainer would have to dig through all the bloat that comes with it to make sure I didn't add any malicious code lines.

[YsEmei]

Would it be possible to add Command Alias support too?

[KazWolfe]

Hey. I'm happy that you want to adopt this. Feel free to use any code that seems useful to you.

Neat, thank you for the confirmation. I did take a look at your repo but decided to rewrite from scratch to better handle certain edge cases.

Would it be possible to add Command Alias support too?

This is a question for Cara, but I suspect that it'd be better to add case insensitivity to that Tweak rather than perform a double-hijack here. I could probably add support to this tweak directly, but I'm unsure how that would affect the execution flow and if it'd further mess with context.

[Caraxi]

I added support for command alias

[KazWolfe]

Looks good and that should work, since you'd still be redirecting the alias prior to the command handler's initial invokes.