Implement Languagetool protocol

[thomas-mc-work]

Hello!

The other day this idea came into my mind: Did you ever think about implementing the server part of the Languagetool protocol? The would immediately enable a lot of clients to interact with your service (LibreOffice, Mozilla Thunderbird) since you can configure the target server (mostly).

The effort might not be too difficult since it's Http based.

[elijah-potter]

It's something I've considered, but never seriously attempted. I'd be interested in taking a crack at it. How would you want this to be distributed? I'm imagining two possibilities, each with their own pros and cons.

• A cloud service (maybe in a Cloudflare Worker or something).
• An installable binary that runs on your device.

Which would you prefer?

[ccoVeille]

Installable binary for me.

[thomas-mc-work]

Basically my issue isn't about the distribution but about the implementation of an existing protocol which has a very broad adoption already. The server side would be a self-hosted binary. But there could also exist a cloud service, maybe even paid to get you some income.

[hippietrail]

I've never used LanguageTool but if there seems to be value in having some version of Harper available in the cloud somehow with a web API of some sort. This could be used in all manner of ways.

A LanguageTool binary would still require a local server to communicate using its existing protocols right? Unless it can run over any arbitrary protocol?

[BradKML]

Would love to see where this goes, since if there are also ways to copy all the grammatical rules from LanguageTools + some of the PRs/forks with missing functions of LT

[thomas-mc-work]

I think this is a different topic. I just like to have harper talk the existing LanguageTool protocol in order to be able to reuse the existing clients.

[BradKML]

Probably can start elsewhere #1465

[thomas-mc-work]

Just found the API documentation here:

https://languagetool.org/http-api/

Should make the re-implementation a lot easier.