user@domain.c

validates to true when checking whether it's a valid email (using the most commonly used regex) since the underlying spec for email addresses apparently allows one letter TLDs even though they aren't used. my question for anyone willing to answer is whether OUR email validation function should allow single letter TLDs per the spec or whether OUR validation should return false on the premise that no person would have such an email (or hamfisted entering a 'valid' one)

Agree

with which?

We should not allow as valid for 1 letter TLD

i'm inclined to agree, but still torn

or ability to define custom but then it becomes a custom validator

the main issue is trying to protect against mistyped emails

Usually for domains with typos there is a need to have a suggested if the domain looks “close” to gmail when typing “gmial” for example; “Did you mean to have gmail.com?"

sure, will consider that as well

similar how smart tv’s offer the remaining @domain.com when trying to Auth to an app like Netflix etc

not all emails i need to validate are user input per se, most are coming in through integrations, but still need to be validated.

haha, yea lots of algorithms in that fiasco

My local newspaper flags my email address as invalid even though it's in the form abc@abcd.com.