Properly parse urls without "//" after the protocol

[iamtheib]

Fixes issue #8

let url = parse('mailto:some@one.co'); // starting out with no slashes
url.set('protocol', 'https');
url.href === 'https://some@one.co'; // true

url = parse('https://github.com'); // starting out with slashes
url.set('protocol', 'sip', true);
url.href === 'sip:github.com'; // true

url.protocol = 'http' // this will not behave as intended
url.toString() === 'http://github.com'; // false

[coveralls]

Coverage remained the same at 100.0% when pulling 0838acd on iamtheib:master into 7d03c18 on unshiftio:master.

[lpinca]

I would not add this, it will throw an error anyway if the given argument is not function.
It will save few bytes on the browser.

[3rd-Eden]

Agreed

[lpinca]

Left some comments but LGTM. It would be nice to have another pair of eyes here anyway.
Thank you.

[3rd-Eden]

Could you JSDoc comment this function? So it documented what it returns, and what the function expects to receive?

[iamtheib]

It is meant to be an internal/private function not exposed in the API. I can still use JSDoc for it I guess

[3rd-Eden]

Minor comments, but LGTM otherwise

[iamtheib]

Made the suggested changes. @lpinca I left the for loop in the test

[coveralls]

Coverage remained the same at 100.0% when pulling c707f0e on iamtheib:master into 7d03c18 on unshiftio:master.

[lpinca]

It was ok with just Object but thanks for the effort.

[lpinca]

👍

[3rd-Eden]

Ill release a new minor in a bit. Thanks for your contribution

[iamtheib]

No problem. Happy to help.

[lpinca]

This test fails on IE, we need to fix it.

[iamtheib]

Yes. Just seeing it. It has to do with the expected error message.

[lpinca]

On my way.

[iamtheib]

End of this line can probably be changed to throws(/function/i). Or just assume it throws(), if that's possible

[lpinca]

02a9296

[3rd-Eden]

Published as 1.1.0