Distributable files.
This directory contains distributable files for use in browser environments. Each distributable is a standalone UMD bundle which, if no recognized module system is present, will expose bundle contents to the global scope.
<script type="text/javascript" src="/path/to/<bundle>.js"></script>where <bundle> corresponds to the desired bundle. For example, to include the main un-minified bundle exposing a flat namespace
<script type="text/javascript" src="/path/to/stdlib-flat.js"></script>If no recognized module system is present, access bundle contents via the global scope. For example, assuming the flat namespace bundle sourced above,
<script type="text/javascript">
// `stdlib` is a global variable...
var erf = stdlib.erf;
console.log( erf( 0.5 ) );
</script>-
Bundles are one of two namespace types:
flatortree. Atreenamespace is a nested object namespace which mirrors the project's layout (e.g.,stdlib.math.base.special.erf. Aflatnamespace uses the global alias namespace, where each package has a unique alias (e.g.,stdlib.base.erf). Which namespace is preferred depends on personal taste and application context. -
Each minified bundle has a corresponding gzip-compressed bundle. The gzip compression level for each compressed bundle is
9, which is the highest (and most optimal) compression level. Deciding between uncompressed and compressed bundles depends on the application and whether compression is handled elsewhere in the application stack (e.g., nginx, CDN, et cetera). -
While you are strongly encouraged to vendor bundles and host with a CDN/provider which can provide availability guarantees, especially for production applications, bundles are available via unpkg for quick demos, proof-of-concepts, and instructional material. For example,
<script type="text/javascript" src="https://unpkg.com/@stdlib/stdlib/dist/stdlib-repl.min.js"></script>
Please be mindful that unpkg is a free, best-effort service relying on donated infrastructure which does not provide any availability guarantees. Under no circumstances should you abuse or misuse the service. You have been warned.
The main bundles, stdlib-flat and stdlib-tree, contain all browser compatible packages except for
@stdlib/repl: REPL environment.@stdlib/namespace: global alias namespace.@stdlib/datasets/*: datasets.
The excluded packages can significantly inflate bundle size, and, if desired, should be bundled and sourced separately.
<script type="text/javascript" src="/path/to/stdlib-flat.min.js"></script>
<script type="text/javascript">
// If no recognized module system present, exposed to global scope:
var erf = stdlib.base.erf;
console.log( erf( 0.5 ) );
</script>| stdlib-flat.js | stdlib-flat.min.js | stdlib-flat.min.js.gz |
|---|---|---|
| 5.62 MB | 1.356 MB | 299.266 kB |
| stdlib-tree.js | stdlib-tree.min.js | stdlib-tree.min.js.gz |
|---|---|---|
| 5.64 MB | 1.371 MB | 299.874 kB |
The dataset bundle, stdlib-datasets-tree, contains all datasets and exposes them as a tree namespace. Unless an application depends on these datasets, they should not be sourced, and, if needed, consider bundling only those datasets which are necessary.
<script type="text/javascript" src="/path/to/stdlib-datasets-tree.min.js"></script>
<script type="text/javascript">
// If no recognized module system present, exposed to global scope:
var datasets = stdlib_datasets.datasets;
console.log( datasets.AFINN_111 );
</script>| stdlib-datasets-tree.min.js | stdlib-datasets-tree.min.js.gz |
|---|---|
| 52.205 MB | 15.988 MB |
The REPL bundle, stdlib-repl, contains all browser compatible packages exposed via the project REPL and exposes a single function which attaches exports to a provided context.
<script type="text/javascript" src="/path/to/stdlib-repl.min.js"></script>
<script type="text/javascript">
// If no recognized module system present, exposed to global scope:
var repl = stdlib_repl.repl;
// Extend the `window` context with REPL contents:
repl( window );
// Print the help docs:
help( base.erf );
</script>| stdlib-repl.min.js | stdlib-repl.min.js.gz |
|---|---|
| 59.259 MB | 16.937 MB |