I have been using two of the most useful checks included in the LanguageTool extension for OmegaT: flagging target segments that are too long or too short with regards to the source and flagging segments where the source and the target have different final punctuation.
I was doing that in a Spanish translation and it was helpful to catch segmentstion issues. However, I had to do the same in Finnish but I saw no LT checks available for this language.
I came LT site and saw that there were no active rules for Finnish but still I was surprised because the two checks above are language-independent. In other words, the same check can be applied regardless of the languages compared.
So the question is: Would it be possible to have a language-blind library for those languages for which there are no language-spcific rules?
I guess what I mean is that the LT plugin in OmegaT should fetch language rules from the library for the target language in the project settings PLUS generic rules from the generic library regardless of the project settings. The former might not exist but the latter should always apply.
Well, to me that sounds so simple that it could be implemented either in OmegaT itself or in the LT add-on for OmegaT. Making changes to the LT core because of that sounds like overkill to me.
Thank you for your answer, Daniel. I understand your point.
The two checks that I mentioned were just an example but there might be other language-independent checks, e.g. empty target, unaltered segment, unaltered source words in the target segment, target segment consistency, source segment consistency, etc. (haven’t checked which ones actually exist already for LT).
Could you please give more information about implementing those checks in the LT add-on for OmegaT? Is that “so simple” that a simple user can do it, or would it be more a task for a Java developer?