Open aonsager opened 2 years ago
Ah yes, thanks for pointing this out @aonsager . This is indeed the case and it's a problem. Not just Japanese, CJK scripts do not work currently.
This has to do with how the FTS system tokenizer [1]. This can be configured through and is in the backlog
Have been rewriting the backend for this to use an alternate search system. In my limited testing CJK works roughly as expected. In the case of Chinese there's no special handling for word separation so individual characters are treated as terms. Could be improved but definitely better than the status quo.
It's not exactly the same tool but here's the link: https://github.com/iansinnott/browser-gopher
When typing a search query in Japanese I get zero search results. The text is visible in search results when I find the item through English queries, so I thought there may be some filter it's not getting through when parsing either the query or the results.
I understand that this may be very low-priority, so please handle as you see fit. Thanks!