Problem using fuzzy search on HTML content in my database

OpenCitizen

Hi everyone,
Hope you're all well!

So first of all thanks to the Xano team for everything as this is an incredible product that I would recommend a hundred times.

I currently have one small issue that I can't seem to get around: I try to do a fuzzy search on a field that contains HTML content.

It works really well actually, but as the language of my app is French I run across accentuated character here and there that the search field doesn't recognize.

For instance, everytime you would have an 'é' you get a 'é' in the database instead. I manage to transform the 'é' from the front-end into a 'é' that goes through the JSON well but then the fuzzy search doesn't match it with the content in the database.

It actually returns an empty result, so it's not like it ignores it (which could be an OK workaround but words with multiple accentuated characters would get lost in the process). Exact matching didn't help as well unfortunately.

Is there a way to get around this? Or are these HTML entity numbers considered stop words or something like that?

My database is 50.000+ records, growing everyday and each HTML is pretty long so converting everything to accentuated characters doesn't seem appropriate.

I have a "fuzzy search" plugin in my front-end that matches these quite well actually but above 300 records it starts to seriously slow down so leveraging Xano would be ideal here as the performance is just great.

Thanks for the help! Have a great day!

Lefteris - blupry.com

Hello @OpenCitizen,

The easier way to get over that problem is probably to just allow html to parse and save the characters to the backend with the encoding that it uses, and then from the frontend implementation, just decode the html data you retrieve from the backend.

Considering that there are about 15 special French characters, the decoding shouldn't take too much time.