URL.canParse

By  on  

Parsing of URLs on the client side has been a common practice for two decades. The early days included using illegible regular expressions but the JavaScript specification eventually evolved into a new URL method of parsing URLs. While URL is incredibly useful when a valid URL is provided, an invalid string will throw an error -- yikes! A new method, URL.canParse, will soon be available to validate URLs!

Providing a malformed URL to new URL will throw an error, so every use of new URL would need to be within a try/catch block:

// The correct, safest way
try {
  const url = new URL('https://davidwalsh.name/pornhub-interview');
} catch (e) {
  console.log("Bad URL provided!");
}

// Oops, these are problematic (mostly relative URLs)
new URL('/');
new URL('../');
new URL('/pornhub-interview');
new URL('?q=search+term');
new URL('davidwalsh.name');

// Also works
new URL('javascript:;');

As you can see, strings that would work properly with an <a> tag sometimes won't with new URL. With URL.canParse, you can avoid the try/catch mess to determine URL validity:

// Detect problematic URLs
URL.canParse('/'); // false
URL.canParse('/pornhub-interview'); // false
URL.canParse('davidwalsh.name'); //false

// Proper usage
if (URL.canParse('https://davidwalsh.name/pornhub-interview')) {
  const parsed = new URL('https://davidwalsh.name/pornhub-interview');
}

We've come a long way from cryptic regexes and burner <a> elements to this URL and URL.canParse APIs. URLs represent so much more than location these days, so having a reliable API has helped web developers so much!

Recent Features

Incredible Demos

  • By
    Advanced CSS Printing &#8211; Using JavaScript Double-Click To Remove Unwanted DIVs

    Like any good programmer, I'm constantly searching around the internet for ideas and articles that can help me improve my code. There are thousands of talented programmers out there so I stumble upon some great articles and code snippets that I like to print out...

  • By
    MooTools HTML Police: dwMarkupMarine

    We've all inherited rubbish websites from webmasters that couldn't master valid HTML. You know the horrid markup: paragraph tags with align attributes and body tags with background attributes. It's almost a sin what they do. That's where dwMarkupMarine comes in.

Discussion

  1. picker

    It doesn’t work in Typescript (with node v22 and @types/node v22) because:
    TS2339: Property ‘canParse’ does not exist on type ‘{ new (url: string | URL, base?: string | URL | undefined): URL; prototype: URL; createObjectURL(obj: Blob | MediaSource):
    string; revokeObjectURL(url: string): void; }’.

Wrap your code in <pre class="{language}"></pre> tags, link to a GitHub gist, JSFiddle fiddle, or CodePen pen to embed!