Unicode Hash tag word boundary isnt recognised correctly

In the language Telugu ( South India ) while posting a hash tag the end of a word should ideally be when it encounters a space. But the issue here is it is recognising the end of a word for every vowel. Please check unicode spec on word boundaries. Mastodon doesn't have this issue and federated posts from a masto instance show these hash tags nicely. This issue exists with both pleroma and mastodon frontends.