Add missing characters to URL regex

[daniel-stoneuk]

Added the unreserved character ~ and the reserved characters '(), to the URL matching regex. If I'm understanding RFC 3986 correctly, "reserved" characters are allowed unencoded in URIs (& there already exist some of these characters in the regex) and so are "unreserved" characters like ~.

The tilde looks like its an unsafe character to have unencoded in RFC 1738, though it appears this is now superseded in most applications - and regardless, if this regex is only used to determine if a link is clickable I don't think there's much harm in including it. I'd love to get some more opinions on this.

This should fix #4705 and improve detection of clickable URLs. I have checked that this is the only place this REGEX string is used but since this is my first contribution, please can someone else check this as well.

Changelog(Fixes): Fixed an issue where URLs in JSON responses were not properly clickable if they included certain unreserved characters

[johnwchadwick]

Looks like it works for me.

Before:

After:

I don't see a reason why this would break anything either. Obviously, it will break if a URL is followed directly by any of the newly added characters, but that's not a big deal I don't think, especially given that in Insomnia, we're mostly dealing with URLs inside machine-readable formats.

LGTM.

[daniel-stoneuk]

Apologies for the force push. I messed up trying to rebase on top of develop now that the directory these changes are in has been renamed 😨

[johnwchadwick]

Apologies for the force push. I messed up trying to rebase on top of develop now that the directory these changes are in has been renamed fearful

Thanks for the rebase. Force pushing on PR branches is fine and not a problem in my opinion.