issue about detect incorrectlly

[mickeyzo12]

I try with many text of korean but CLD3 is unable to detect it.

for example:
Korean text: "이 회의에서는 업계 전반의" => output: vi => should be ko

English text: "hello world" => output: ky => should be en

how can CLD3 detect language more accurately?

thank you very much.

[Balajanovski]

I have also run into this issue. cld3 is unusable with these bugs.