Just to have it said since people have made me aware of a certain spreadsheet:
Yes, SLR Translator uses the same ".trans" file format as Translator++, I've made some tiny changes to it, but nothing that would impact this topic, and as such you can use LinguaGacha to add LLM translations to it.
But there are still some serious problems with that.
If you parse a project with SLR and then use the prepare for batch translation button to initiate getting rid of harmful or empty objects and tag everything, it tags stuff according to rules in the escaper paired with scripts in post processing, but you never actually use that escaper when you run LinguaGacha, it has its own translation engine. That means stuff can be tagged green in SLR that doesn't actually have any protection in LG and break, or stuff can be tagged red in SLR, but actually be completely fine for LG and just does not have the necessary stuff in my code, yet. So essentially even if you get rid of all translations in gray, red, dark red, yellow, gold, blue and aqua tagged cells after using LG on the .trans file, stuff could still be broken and crash your game.
The next problem is that currently post processing and the automatic checking for errors scripts are the same button, which means you can't run them independently. As a result it will attempt to do hundreds of "fixes" for Sugoi that do not apply for non Sugoi translation results and might actually make the translation significantly worse, but if you don't run it, it will also not do the process for checking for errors, which seems like a really bad idea after already playing lottery with the tagging.
I'm looking into ways to make SLR Translator better compatible with LinguaGacha, but the issue is really that SLR is just built from the core to run with Sugoi, and since you are editing the .trans outside of the tool, it obviously does not know what you did or didn't do.
