minisearch VS uFuzzy

I wrote an online catalog a while back (and I need to get back on adding graphics and products at some point). It’s written using Eleventy and the minisearch library. The source and data are available on Github if you want to see how I did things. I’m not a professional web designer either, but it was a fun project.

You could have a look at https://github.com/lucaong/minisearch/

If you're dealing with thousands of records or less, searching titles and summaries rather than long bodies of text, I recommend looking into client-side solutions. Nothing beats the responsiveness of search-as-you-type entirely on the client side. It can be fairly sophisticated fulltext search. For example, I've built had great success with MiniSearch.

I quite enjoy minisearch[1] which is also 0 dependencies, actively maintained, and I expect would work well in a worker environment. I dropped it into a service worker and plugged it with a simple point in polygon script to enable geosearch for a recent project[2] and it played v. nicely.

[1] https://github.com/lucaong/minisearch

No magic here, the underlying engine is Minisearch, which uses the BM25 algorithm (the de facto standard among search libraries). Omnisearch adds a magic sauce during indexing by converting notes into custom objects, with the following fields: - body (the plain markdown text) - filename & yaml aliases - level 1 headers - level 2 headers - level 3 headers

It's actually really simple! Minisearch did most of the heavy lifting so all we needed to do was the crawling, storing and UI etc. I'd check that out if you're interested in the search part!

We use MiniSearch for searching, while fast-fuzzy is used for highlighting of detected search terms.

Thank you. We need more libs like that. I just researched the field yesterday and https://github.com/leeoniya/uFuzzy looked pretty good. But there is a gap in the market of such libs. Just few allow to send the whole html document, serialize and deserialize index to be used in browser, highlighting the matches is desired feature.

Most importantly very few fuzzy search libs can get a simple substring match as a priority, which is understandable but not helpful. Imagine searching for “xample” and not having “example” among the results.

interesting, i made something much more stupid but brutally effective: https://github.com/leeoniya/uFuzzy

i guess if you wanted to do sorting by contact frequency you can just keep the original haystack sorted by frequency, and get Richard match first.

allow me to self-promote https://github.com/leeoniya/uFuzzy

Keep the tech simple and minimal. Just the basics: javascript, css, HTML. No react or other frameworks. No tech for tech's sake. I did use one very excellent search library uFuzzy.

maybe habit, maybe less devDependencies, maybe wanting to hand-author a readable/commented form of https://github.com/leeoniya/uFuzzy/blob/main/dist/uFuzzy.d.ts?