extract method does not work with '#' since v6.5

[albert-gonzalez]

Hi!

I am trying to update the library from v6.2 and I have found an issue since v6.5. The extract method returns an empty string when I pass a value with a #. For example

• extract('https://wwww.domain.com#hello?param=1'); // returns ''
• extract('#hello?param=1'); // returns '';

I think the problem is in the function "removeHash", introduced in the v6.5. This function removes all the string that is after the #. I wrote this simple fiddle with an example: https://jsfiddle.net/5Ljnwc07/

Do you think it's a bug? Or I am doing something wrong?

Thank you!

[serhii12]

Any updates on this?

[samuelpetroline]

It seems that this is the intended behaviour:

query-string/test/extract.js

Line 8 in a1d108f

t.is(queryString.extract('https://foo.bar#top?foo=bar'), '');

[baile320]

Additional potential support that this is intended behavior: see this stackoverflow answer (which references to RFC 3986). Percent encoding can escape the #, if that is something you have control over.

[sindresorhus]

Closing as this is working as intended.