I have a rather odd situation where it might be preferable to not use stopwords and to have a value of ft_min_word_len equal to 1. The index in question would be dealing with only a few words per row as an average and it is important to get scores back that can distinguish between rows with data such as "who" and "who is" and return as a result against a search string such as "Who is this Matt bloke?" (The table is a big list of sentence stubs for an experimental chatbot idea).

The default settings (which are good for most everything else) would block such tiny stubs from the index. However, if this is the global DB setting then it would make fulltext indexes very slow and noisy for most other uses.

Is there a way to have fulltext indexes and use match on that table only that have an empty stopwords list and allow tiny words? Or is there another way of finding partial matches on large lists of sentences stubs to match full sentences against?

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.