Unusually-formatted episode number not parsed correctly

[lcdt22890158]

The release group "Aurora" seems to have used a weird naming convention for Gintama. Episodes are named like this:

"[Aurora] Gintama 第195話 (BD 1920x1080 x264 AAC).mp4"

Currently guessit fails to recognise the episode numbers. Medusa reports the following:

release_group: Aurora, title: Gintama 第195話, source: BDRip, screen_size: 1080p, aspect_ratio: 1.778, video_codec: H.264, video_encoder: x264, audio_codec: AAC, container: mp4, mimetype: video/mp4, type: episode, parsing_time: 0.009290456771850586, season: None, episode: [], absolute_episode: [], quality: Unknown, total_time: 0.044719696044921875

Maybe this is an unreasonable format to support, but it would be nice if it were handled correctly.

Thanks!

[Toilal]

I could add 第/話 as separators, but what doesn't it mean ?

[Toilal]

So https://github.com/guessit-io/guessit/commit/40f43b133cf5fb43d79c1e2b886a3620830a4a37 goes in the right direction, title is now OK but the episode is still missing. We need enhance the options.json configuration features and make seps configurable this way, but it requires some refactoring.

I place this task in high priority though.

[lcdt22890158]

Thanks for working towards this. I don't think it needs to be high priority though, it's really just a nice-to-have.