Perhaps worth adding that verifying an e-mail address via regex is difficult and not actually that useful.
Difficult, because a lot of e-mail addresses are syntactically valid. For example, a@b is valid. As is a@1.2.3.4. As is "spaces are cool 😎"@b (with the quotes).
And it’s not that useful, because you need to send out an e-mail either way, if you want to know that it’s the correct address and the user has access to it. A regex can’t catch a typo in the actual words, for example.
If you want a basic sanity check to catch when the user types their name into the e-mail field, then you can require that it contains an @ and, if you want, that there’s at least one symbol before and after.
If you go beyond that, you either need a massive regex or you’re going to lock out users with unusual e-mail addresses.
Yep, these email address validation tools are the bane of my existence. I use a non-typical TLD and it sometimes gets flagged as invalid because it’s not .com, .net, or .org.
I wrote one parser that checked the top level domain against a txt of all top level domains downloaded from Icann once a day. Was kind of proud of that one.
Perhaps worth adding that verifying an e-mail address via regex is difficult and not actually that useful.
Difficult, because a lot of e-mail addresses are syntactically valid. For example,
a@bis valid. As isa@1.2.3.4. As is"spaces are cool 😎"@b(with the quotes).And it’s not that useful, because you need to send out an e-mail either way, if you want to know that it’s the correct address and the user has access to it. A regex can’t catch a typo in the actual words, for example.
If you want a basic sanity check to catch when the user types their name into the e-mail field, then you can require that it contains an
@and, if you want, that there’s at least one symbol before and after.If you go beyond that, you either need a massive regex or you’re going to lock out users with unusual e-mail addresses.
Yep, these email address validation tools are the bane of my existence. I use a non-typical TLD and it sometimes gets flagged as invalid because it’s not .com, .net, or .org.
I wrote one parser that checked the top level domain against a txt of all top level domains downloaded from Icann once a day. Was kind of proud of that one.
That’s a pretty clever way of doing it!
I’ve found many systems don’t work with the .company TLD!
or some u.s.-centric ones that break with .co.uk