Skip to content

Minimalist wrapper Svelte Stores around Stork Search, also provides customizable Components

License

Notifications You must be signed in to change notification settings

novacbn/svelte-stork

Repository files navigation

svelte-stork

Description

Minimalist wrapper Svelte Stores around Stork Search, also provides customizable Components.

Demo

See a demo of the Svelte Components at novacbn.github.io/svelte-stork.

Getting Started

NOTE: To learn how to build your own custom search indices and other concepts, visit the Stork documentation.

To get started with this library, you need to install Stork Search on your Web Application. Alternatively, see the below sample on how to have this library handle that for you.

Usage

Stores

import {initialize, register, search} from "svelte-stork";

// Before we can do anything with `search`, you need to
// initialize Stork Search
await initialize({
    // If you want this library to handle mounting / loading
    // of the main Stork JS library after page load, you can
    // uncomment the following line
    //
    // script_url: "/path/to/stork.js",

    // The WASM blob is not optional however, due to how
    // Browsers handle loading them
    wasm_url: "/path/to/stork.wasm",
});

// Next, you need to register your search index with Stork Search
await register({
    // All search indices are globally registered to the webpage, so
    // you need to choose a unique name. Preferably versioned as well
    index_name: "my-search-index_v1.0.0",

    // Make sure this path publically accessible
    index_url: "/path/to/search_index.v1.0.0.st",
});

// Now if we initialized and registered successfully, we can create the Svelte Store
const store = search({
    // You can customize how long in milliseconds the Store after
    // receiving input to update the query results
    //
    // DEFAULT: 250ms (0.25s)
    debounce: 250,

    // You also need to reference your registered search index
    index_name: "my-search-index_v1.0.0",

    // Lastly, you can customize the minimum amount of characters a
    // end-user needs to input before queries are performed
    //
    // DEFAULT: 3 (three characters)
    minimum_length: 3,
});

// Like any other Svelte Store, we update the input query via `set`
store.set("my search query");

// Also subscriptions
store.subscribe((query) => {
    // Finally retrieve your query results, visit the link below to see the typings
    // https://github.com/novacbn/svelte-stork/blob/afc328ca576f83a7c903027f0b117f5a1a566f12/src/lib/stores/search.ts#L115-L130
    console.log(query.results);
});

Components

NOTE: This sample expects initialization / registration of Stork Search was already performed.

<script>
    import {search, selectable} from "svelte-stork";
    import * as Stork from "svelte-stork/components";

    // Just input the same options you passed into the Javascript sample
    const store = search(...);

    // This caches the `HTMLElement` reference to the `<div>` element from `<Output>`
    let output = null;

    let value = "";

    // After the initial configuration / load, it's really simple to get started. Just
    // use Svelte's reactive blocks pattern to update from input and listen for query results
    $: $store = value;
    $: query = $store;
</script>

<!--
    By using the `selectable` Svelte Action here, this will allow the end-user
    to select the next / previous highlighted result entries via their arrow keys.
    And also be able to navigate to the highlighted entry via the enter key
-->

<input
    type="text"
    placeholder="...search"
    bind:value
    use:selectable={{output}}
/>

<!--
    Now we can just pass in the query output into the built-in basic search UI

    Properties:
        - `excerpts_maximum` / `results_maximum` — When set greater than `-1`, they limit the respective amount of
        excerpts and results items displayed in search outputs
-->

<Stork.Output
    bind:element={output}
    excerpts_maximum={-1}
    results_maximum={-1} 
    {query}
/>

Developer

Installation

Open your terminal and install via npm:

npm install @novacbn/svelte-stork

Documentation

TODO

CSS Custom Properties

Name Default
--svst-attribution-color currentColor
--svst-attribution-font-size 0.8em
--svst-attribution-padding 0.5rem 1em
--svst-attribution-opacity 0.8
--svst-attribution-text-decoration underline
--svst-attribution-hover-opacity 1
--svst-attribution-link-color hsl(192, 82%, 44%)
--svst-attribution-visited-color hsl(271, 63%, 34%)


--svst-entry-color currentColor
--svst-entry-cursor pointer
--svst-entry-padding 1em
--svst-entry-text-decoration none
--svst-entry-hover-background-color hsl(192, 82%, 78%)
--svst-entry-hover-opacity 1


--svst-entry-title-spacing 0.8em
--svst-entry-title-font-size 1.1em
--svst-entry-title-font-weight bold


--svst-excerpt-line-height 1
--svst-excerpt-spacing 0.8em


--svst-highlight-background-color hsl(59, 98%, 58%)
--svst-highlight-color currentColor


--svst-message-color currentColor
--svst-message-padding 0.5em 1em
--svst-message-font-size 1em
--svst-message-opacity 0.8


--svst-output-color black
--svst-output-font-family -apple-system, BlinkMacSystemFont, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji"
--svst-output-font-size 1em
--svst-output-font-weight 400
--svst-output-max-height auto


--svst-result-border 1px solid hsla(0, 0%, 90%, 1)
--svst-result-font-size 0.8em


--svst-results-border 1px solid hsla(0, 0%, 80%, 1)

API

  • Functions

    • initalize
    • register
  • Stores

    • search
  • Components

    • Output

    • Message

    • Attribution

    • Highlight

    • Result

    • Results

    • Entry

    • EntryTitle

    • Excerpt