Inconsistent sorting compared with other libraries

[Digote]

I don't think it's a bug, but rather a question about how the text extraction is organized.

In the file PouchContent.msbt, we have item descriptions and their names.

When we use a tool that has been around for some time, like (Kuriimu), these strings come organized as follows: item description, item name.

However, when we use the msbt_yml tool, it ends up organizing these strings differently, making it a bit more complicated to translate because we have to search for the corresponding description for each specific item.

Once the text passes through the msbt_yml tool, it will always be organized in that way.

I'll attach some screenshots of how the extraction looks before and after passing through msbt_yml."

Before

After

Thanks a lot for the tool, it's been very useful to use the .yml files on a review site.

[ArchLeaders]

Labels are sorted using this hash algorithm: https://github.com/EPD-Libraries/msbt/blob/main/src/sections/label_section.cpp#L7-L14

Source is from CLMS (which iirc is based off official RE): https://github.com/Trippixyz/CLMS/blob/main/LMS.cs#L178-L191

[ArchLeaders]

Although, it doesn't look like that sorts within groups, in which case it's not sorted at all and is just written in the same order as the yaml. Could be wrong though, it's been a while since I've looked at this code.

[Digote]

Since I still can't help you directly with the code, I'll share another example of a tool that seems to extract data in the same way as Kurrimu, but it's written in pure C.

https://github.com/AlgebraManiacABC/msbt2txt

[ArchLeaders]

As far as I can tell, that doesn't sort LBL either. Also it's read-only, so there won't be much difference there.

[ArchLeaders]

When I have more time I'll try to look into this further.