In general, it also breaks filename-based file type detection

What do you mean by this?

In general, it also breaks filename-based file type detection

What do you mean by this?

I mean that it changes the filename's file extension. Video Name.srt - eng (incorrect extension) has a different file extension type from Video Name - eng.srt (SubRip Text (SRT) subtitle file; correct).

- Video Name.srt - eng
+ Video Name - eng.srt

May or may not matter to Kodi, because Kodi gets "told" it's a subtitle file (https://github.com/putdotio/putio-kodi/blob/v3.1.0/main.py#L135), but I certainly consider it a bug.

tree ~/.kodi/temp/plugin.video.putio

~/.kodi/temp/plugin.video.putio
└── subtitles
    └── 0000000000
        ├── abc_de
        │   └── Video Name.srt - eng
        ...

Got it.

We have 3 types of subtitles:

• Folder subtitles: We look at current directory for a .srt file. So, name always have extension. We don't detect language for these, always unknown.
• File subtitles: We extract these from file and save as ...-{lang}.srt. So, name always have extension and language code. {lang} here is how it is tagged in MKV, so may be improper (seems < 0.2%).
• OpenSubtitles subtitles: These always have .srt extension. Name can include language code (depends how it's uploaded), but we have language code separately.

What are your thought on when we should append language code?

I think:

• In cases we have language we append language. It's simple and shown name is always the actual subtitle name (minus extension). If name has right/wrong language in it, it stays in the name.
• For folder subtitles we append und except name has explicit language in form {name}.{lang}.srt. Identifying the language wrong is IMO worse than saying Unknown. I saw this with a HI (hearing impaired) subtitle identified as Hindi. It also matches the behavior of our other apps (currently always marked Unknown without exception).
• In all cases, make the filename {name}.{lang}.srt.