Stork
Impossibly fast web search, made for static sites.
Stork is two things. First, it's an indexer: it indexes your loosely-structured content and creates a file that you can upload to your web server. Second, it's a Javascript + WebAssembly frontend for that index file: Stork will hook into an <input> on your web page, download the index you've specified, and display the best search results immediately to your user, as they type. The precomputed index and WebAssembly frontend module make the entire Stork engine very good, and very fast.
Currently in development by James Little
Getting Started
Let's put a search box online that searches within the text of the Federalist Papers.
See this demo live at https://stork-search.net.
<!DOCTYPE html>
<html lang="en">
<head>
<title>Federalist Search</title>
</head>
<body>
<div class="stork-wrapper">
<input data-stork="federalist" class="stork-input" />
<div data-stork="federalist-output" class="stork-output"></div>
</div>
<script src="https://files.stork-search.net/stork.js"></script>
<script>
stork.register(
"federalist",
"http://files.stork-search.net/federalist.st"
);
</script>
</body>
</html>
Step 1: Include the HTML
Stork hooks into existing HTML that you include on your page. Each Stork instance has to have an input hook and a results list; those two elements should be placed in a wrapper, though the wrapper is optional.
The input hook should have the data-stork="federalist" attribute, where federalist is the name with which you register that search instance. (This way, you can have multiple, independent search boxes on a page, all pointing to different instances.) It doesn't have to be federalist -- you can change it to whatever you want.
The results list should be an empty <div> tag with the attribute data-stork="federalist-results". Again, here, you can change federalist to whatever you want.
The classes in the example above (stork-input, stork-output) are for the theme. Most Stork themes assume the format above; the theme documentation will tell you if it requires something different. You can also design your own theme, at which point the styling and class names are up to you.
Step 2: Include the Javascript
You need to include stork.js, which you can either load from the Stork CDN or host yourself. This will load the Stork WebAssembly blob and create the Stork object, which will allow for registering and configuring indices.
Then, you should register at least one index:
stork.register("federalist", "http://files.stork-search.net/federalist.st");
The search index you build needs to be stored somewhere with a public URL. To register
This registers the index stored at http://files.stork-search.net/federalist.st under the name federalist; the data-stork attributes in the HTML will hook into this name.
Finally, you can set some configuration options for how your search bar will interact with the index and with the page.
Building your own index
You probably don't want to add an interface to your own website that lets you search through the Federalist papers. Here's how to make your search bar yours.
To build an index, you need the Stork executable on your computer. On a Mac, you can run $ brew install jameslittle230/stork-tap/stork to download it. If that doesn't work for you, check out the installation documentation to learn how to build it manually.
The search index is based on a document structure: you give Stork a list of documents on disk and include some metadata about those documents, and Stork will build its search index based on the contents of those documents.
First, you need a configuration file that describes, among other things, that list of files:
[input]
base_directory = "test/federalist"
files = [
{path = "federalist-1.txt", url = "/federalist-1/", title = "Introduction"},
{path = "federalist-2.txt", url = "/federalist-2/", title = "Concerning Dangers from Foreign Force and Influence"},
{path = "federalist-3.txt", url = "/federalist-3/", title = "Concerning Dangers from Foreign Force and Influence 2"},
{path = "federalist-4.txt", url = "/federalist-4/", title = "Concerning Dangers from Foreign Force and Influence 3"},
{path = "federalist-5.txt", url = "/federalist-5/", title = "Concerning Dangers from Foreign Force and Influence 4"},
{path = "federalist-6.txt", url = "/federalist-6/", title = "Concerning Dangers from Dissensions Between the States"},
{path = "federalist-7.txt", url = "/federalist-7/", title = "Concerning Dangers from Dissensions Between the States 2"},
{path = "federalist-8.txt", url = "/federalist-8/", title = "The Consequences of Hostilities Between the States"},
{path = "federalist-9.txt", url = "/federalist-9/", title = "The Union as a Safeguard Against Domestic Faction and Insurrection"},
{path = "federalist-10.txt", url = "/federalist-10/", title = "The Union as a Safeguard Against Domestic Faction and Insurrection 2"}
]
This TOML file describes the base directory of all your documents, then lists out each document along with the web URL at which that document will be found, along with that document's title.
From there, you can build your search index by running:
$ stork build --input federalist.toml --output federalist.st
This will create a new file at federalist.st. You can search through it with the same command line tool:
$ stork search --index federalist.st --query "liberty"
To embed a Stork search interface on your website, first upload the index file to your web server, then pass its URL to the stork.register() function in your web page's Javascript.
Going Further
You can read more documentation and learn more about customization at the project's website: https://stork-search.net.
Development
To build Stork, you'll need:
The repository is structured like a typical Cargo package, with some modifications.
srcholds the Rust source codejsholds the Javascript source codetestholds files used for development, including test indexes and the static files for the development page.test-assetshold binary assets required by Stork's functional tests.
You can build the project using either the Rust entrypoint or the Javascript entrypoint. After you've built the project, you'll see three more directories:
targetholds Rust build artifactspkgholds WASM build artifactsdistholds the final build artifacts and files required for the development server.
If you're interested in extracting the final Stork build artifacts, you can extract the following files after building the project with yarn build:
/target/release/stork/dist/stork.js/dist/stork.wasm
Building the Project
Rust
cargo buildwill build thestorkexecutablecargo runwill run the executable.
Javascript
yarn developwill build an index, build the WASM, and build the Javascript file all in the development configuration. It will then start a development server.yarn buildwill build all the release assets, including the binary, the WASM, and the Javascript. (Note: This does not work on M1 Macs because of a lack of wasm-pack compatibility.)
You can see more Javascript build options by running yarn run.
2.2k Dec 31, 2022
5 Mar 9, 2022
31.6k Dec 31, 2022
17.4k Jan 2, 2023
679 Jan 1, 2023
2k Jan 5, 2023
22 Sep 21, 2022
116 Dec 10, 2022
359 Nov 27, 2022
70 Dec 9, 2022
7.4k Dec 28, 2022
1.8k Dec 26, 2022
69 Jan 2, 2023
3.8k Jan 7, 2023
1.8k Dec 31, 2022
7 Nov 23, 2022
41 Oct 28, 2022
342 Dec 26, 2022