Full text search on the generated site.

[leira]

Currently, neuron generated site only supports search with title and meta data. Is there any plan to support fuzzy full text search? I think it is more important and useful than CLI fuzzy search, As research in the notes is probably done through the web interface.

[leira]

Did some random search found tinysearch. It is a Rust based wasm implementation of Bloom Filter algorithm. It seems to have a good space and time performance. I didn't get into the details yet, maybe we can either use it directly, or seek some Haskell Bloom Filter wasm solution.

[NightMachinery]

Sourcegraph is good for websearching on one's own repo. I personally prefer CLI searching. ugrep seems promising in this regard, especially if the author polishes its interactive tooling.

[GenrihGrigoryan]

Here are some projects, that have full-text search:

https://mathieudutour.github.io/foam-gatsby-template/
https://github.com/cotes2020/jekyll-theme-chirpy
https://github.com/Maxence-L/notenote.link

[srid]

Full-text will easily become problematic for larger Zettelkastens. The browser will have to fetch megabytes of JSON index.

Metadata-only search (graph search) is a good compromise; see #108 (comment)