Url does not pass as expected

[digitalanalogue]

This url: https://www.php.net/manual/en/language.oop5.overloading.php#object.call

Gives these results:

{ "value": { "url": "https://www.php.net/manual/en/language.oop5.overloading.php#object.call", "normalizedUrl": "https://www.ph/manual/en/language.oop5.overloading.php#object.call", "removedTailOnUrl": "", "protocol": "https", "onlyDomain": "www.ph", "onlyParams": null, "onlyUri": "/manual/en/language.oop5.overloading.php#object.call", "onlyUriWithParams": "/manual/en/language.oop5.overloading.php#object.call", "onlyParamsJsn": null, "type": "domain", "port": null }, "area": "text" }

When used like so:

Pattern.TextArea.extractAllFuzzyUrls('https://www.php.net/manual/en/language.oop5.overloading.php#object.call');

I'm using this within a browser (firefox 86 on OSX).

I expect the onlyDomain to be www.php.net

Great project btw, just found this bug when messing about with pasting in urls.

[digitalanalogue]

Also found domains with hypens like this http://sub.do-main.co.uk/ do not work and come out like http://sub.do.main.co.uk/

Not sure if this is due to using extractAllFuzzyUrls and not extractAllUrls, if I use the later it works as expected. Not sure if this is really an error or to be expected when using fuzzy - my aim was to catch as many bad url's as possible but extractAllUrls works well enough for what I want.

[patternknife]

You are right. extractAllUrls is currently a recommended method to extract URLs, but extractAllFuzzyUrls is kind of an experimental thing to extract corrupted ones. I will look into extractAllFuzzyUrls for the issue later.