In my opinion, if discorse automatically puts a space before an emoticon, problems will be solved in all Chinese, Korean and Japanese.
Initially, this problem arises because Japanese and Chinese people insert pictograms without putting a space at the end of the sentence or in the middle of the sentence.
This is because it is based on the language notation that it creates sentences without spaces, so there is no choice.
There is no need to let the program decide whether it is Chinese or not. Would not it be okay for the program to judge only whether or not there is space before emoji?
こんにちは:smiley: