kmschaal

Space-filling curves as a traveling salesman heuristic is a really nice application I was not aware of. I have mostly seen them in the context of domain decomposition for parallel computing, where you cut the curve at n points to obtain n+1 disjoint domains.

Tested it and you are right, thanks for the clarification.

OP here, I wanted to add one technical note I forgot to mention in the post.

The prefix search uses the same suffix array as the substring search. This approach might also be useful for other search libraries that rely on suffix arrays. It can improve the search experience with minimal additional effort.

Happy to discuss the implementation details if anyone’s curious!

μFuzzy has a great comparison project that could serve as a reference for all fuzzy search implementations. My fuzzy searcher (v1) is already included and will soon be updated to v2 (PR is open).

Thank you! I will do a comparison in the near future and update the readme.

Thank you for your explanation, I get your point. Regarding your last suggestion: I think it would be great to measure how often the correct result is near the top. However, don't we face the same issue as before? What is the correct result? Is it the term the user has in mind when writing the query? But what if they make a typo, and the term with the typo also exists in the dataset? Or they just type half of the term they are thinking of, but there are many terms in the dataset with the same prefix?

So, in the end, I believe it's worthwhile to try different implementations and share our subjective experiences.

Hi, thank you for your questions. Unfortunately there are no comparisons yet. The gif is simply a looped screen recording created with Camtasia. Best regards!

Glad to hear that it works well, will look into it.

Thank you for your comment! It is indeed a problem in plain fuzzy search libraries (like this one) that substring matches can have a lower quality than unequal strings of similar length. A solution to that is to implement a higher level search controller that queries the fuzzy searcher, as well as a suffix array searcher. The controller than mixes the matches and returns the best matches across the two searchers. One can even add more searchers, e.g. a phonetic one. With the correct parameters this approach works well.

Thank you, sounds great!

Hi, thanks for sharing! No, I haven't taken wasm into consideration. It would be interesting to see the performance gain, in particular for indexing.

Sorry for the confusion. I think you could generate the memento each time you compile your blog into HTML. The memento can be stored as a json file and served statically by your fileserver. When a user visits your page, retrieve the memento from the server. Then, initialize the searcher with it. In this way you avoid indexing content at runtime.

As a bonus you could cache the memento in the local storage or session storage.

Thank you for your kind feedback! That's a great idea. I have implemented a memento object that can be used for transferring the serialized state of the searcher. The intent of the implementation was to transfer the searcher between a web worker and the main thread. You could try to serve the Memento from the server and store it in an index db. You may have a look at the web worker example I provide in the repository.