diff options
Diffstat (limited to 'school/node_modules/@polka/url/readme.md')
| -rw-r--r-- | school/node_modules/@polka/url/readme.md | 86 |
1 files changed, 86 insertions, 0 deletions
diff --git a/school/node_modules/@polka/url/readme.md b/school/node_modules/@polka/url/readme.md new file mode 100644 index 0000000..dda8379 --- /dev/null +++ b/school/node_modules/@polka/url/readme.md @@ -0,0 +1,86 @@ +# @polka/url [](https://npmjs.org/package/@polka/url) + +> Super fast, memoized `req.url` parser; _not_ limited to [Polka][polka]! + +Parses the `url` from a [`IncomingMessage`](https://nodejs.org/api/http.html#http_class_http_incomingmessage) request. The returned object will always only contain the following keys: `search`, `query`, `pathname`, `path`, `href`, and `_raw`. + +> **Note:** This library does not process `protocol`, `hostname`, `port`, etc.<br>This is because the incoming `req.url` value only begins with the path information. + +Parsed requests will be mutated with a `_parsedUrl` key, containing the returned output. This is used for future memoization, so as to avoid parsing the same `url` value multiple times. + +## Install + +``` +$ npm install --save @polka/url +``` + +## Usage + +```js +const parse = require('@polka/url'); + +let req = { url: '/foo/bar?fizz=buzz' }; +let foo = parse(req); +//=> { search: '?fizz=buzz', +//=> query: 'fizz=buzz', +//=> pathname: '/foo/bar', +//=> path: '/foo/bar?fizz=buzz', +//=> href: '/foo/bar?fizz=buzz', +//=> _raw: '/foo/bar?fizz=buzz' } + +// Attaches result for future memoization +assert.deepEqual(foo, req._parsedUrl); //=> true +``` + +## API + +### url(req) +Returns: `Object` or `undefined` + +> **Important:** The `req` must have a `url` key, otherwise `undefined` will be returned.<br>If no input is provided at all, a `TypeError` will be thrown. + +#### req +Type: `IncomingMessage` or `Object` + +The incoming HTTP request (`req`) or a plain `Object` with a `url` key. + +> **Note:** In Node.js servers, the [`req.url`](https://nodejs.org/api/http.html#http_message_url) begins with a pathname & does not include a `hash`. + + +## Benchmarks + +> Running the `parseurl` benchmark suite on Node 10.9.0 + +``` +Parsing: "/foo/bar?user=tj&pet=fluffy" + nativeurl x 3,496,593 ops/sec ±0.78% (194 runs sampled) + parseurl x 5,702,515 ops/sec ±0.59% (194 runs sampled) + @polka/url x 11,510,281 ops/sec ±1.93% (192 runs sampled) + +REPEAT: "/foo/bar?user=tj&pet=fluffy" + nativeurl x 3,344,884 ops/sec ±0.13% (191 runs sampled) + parseurl x 20,386,848 ops/sec ±0.22% (192 runs sampled) + @polka/url x 21,088,923 ops/sec ±0.58% (191 runs sampled) + +Parsing: "/foo/bar" + nativeurl x 9,808,119 ops/sec ±0.51% (190 runs sampled) + parseurl x 26,186,627 ops/sec ±0.16% (195 runs sampled) + @polka/url x 43,946,765 ops/sec ±0.55% (194 runs sampled) + +Parsing: "/" + nativeurl x 15,698,746 ops/sec ±0.79% (192 runs sampled) + parseurl x 36,861,339 ops/sec ±0.19% (195 runs sampled) + @polka/url x 48,295,119 ops/sec ±0.51% (194 runs sampled) +``` + + +## Support + +Any issues or questions can be sent to the [Polka][polka] repository.<br>However, please specify that your inquiry is about `@polka/url` specifically. + + +## License + +MIT © [Luke Edwards](https://lukeed.com) + +[polka]: https://github.com/lukeed/polka |