Adding the complete architecture for search benchmarking

[Rahban1]

this closes #2417.

[Hetarth02]

Since we already have a file, can't we just do node file.js?

[Rahban1]

but we're replacing the placeholders with their values here

wrapper_js = replace(wrapper_js, "__SEARCH_INDEX__" => JSON.json(search_index_data))
wrapper_js = replace(wrapper_js, "__QUERY__" => "\"" * query * "\"")

[mortenpi]

A couple of thoughts:

• We could add a make search-benchmarks command: https://github.com/JuliaDocs/Documenter.jl/blob/master/Makefile
• We could add a simple CI job that runs these benchmarks on CI. It should be fine to just copy-paste a thing here for now: https://github.com/JuliaDocs/Documenter.jl/blob/master/.github/workflows/CI.yml
• It would be good to have a more realistic reference manual / search index we test against. Options: Documenter's manual or Julia manual. As we discussed on the call, let's start with Documenter.

[mortenpi]

This duplicates the search.js, so it would be good to somehow avoid the duplication here by reusing the search.js in the tests. It would likely require some changes on how we handle the search.js file.

Documenter.jl/assets/html/js/search.js

Lines 189 to 216 in 022c013

	let index = new MiniSearch({
	fields: ["title", "text"], // fields to index for full-text search
	storeFields: ["location", "title", "text", "category", "page"], // fields to return with results
	processTerm: (term) => {
	let word = stopWords.has(term) ? null : term;
	if (word) {
	// custom trimmer that doesn't strip @ and !, which are used in julia macro and function names
	word = word
	.replace(/^[^a-zA-Z0-9@!]+/, "")
	.replace(/[^a-zA-Z0-9@!]+$/, "");
	word = word.toLowerCase();
	}
	return word ?? null;
	},
	// add . as a separator, because otherwise "title": "Documenter.Anchors.add!", would not
	// find anything if searching for "add!", only for the entire qualification
	tokenize: (string) => string.split(/[\s\-\.]+/),
	// options which will be applied during the search
	searchOptions: {
	prefix: true,
	boost: { title: 100 },
	fuzzy: 2,
	},
	});
	index.addAll(data);

[mortenpi]

I think I would be happy to merge this as the first iteration if we could get the duplication & version number sorted as well!

[mortenpi]

Could we make this use the Node JLL? This means that the user wouldn't have to install Node just to run the benchmarks -- we can pull it in via the Julia package manager.

• https://juliahub.com/ui/Packages/General/NodeJS_22_jll

There's some example usage here for example: https://github.com/JuliaComputing/MultiDocumenter.jl/blob/e112da8b744f3393d037a7380e544797c2a41953/src/search/pagefind.jl#L31-L55

[mortenpi]

If we switch to the Node JLL, we can remove these.

Suggested change

	- uses: actions/setup-node@v4
	with:
	node-version: '20.x'
	- name: Install Node.js dependencies
	run: npm install
	working-directory: test/search

[Rahban1]

how's this implementation @mortenpi ?

[mortenpi]

@Rahban1 LGTM! I think we should work a bit on getting more test cases in etc. But I link this infra, so let's get this merged!