4 ms·

The actual relevant question is: how many of you make a point of disallowing this? You have to go out of your way to make that not work, given that + is a valid

The actual relevant question is: how many of you make a point of disallowing this? You have to go out of your way to make that not work, given that + is a valid character in email addresses.

throw0101d 2y ago
> You have to go out of your way to make that not work, given that + is a valid character in email addresses. Per another comment, all one would need for it to not work is have a regex of [a-zA-Z0-9]@[a-z.]+, or using POSIX-compat, [:alnum:]. * https://en.wikipedia.org/wiki/Regular_expression#Character_classes https://en.wikipedia.org/wiki/Regular_expression#Character_c...
- nielsole 2y ago
  > [a-zA-Z0-9]@[a-z.]+ Only a small minority of my email addresses I've ever had matches that pattern.
  - throw0101d 2y ago
    > Only a small minority of my email addresses I've ever had matches that pattern. The example is illustrative, and not definitive, as to how the simple and innocuous decision about a regex can determine functionality, and it is not about someone going "out of your way to make that not work" to break things.
    - dxdm 2y ago
      I don't know if that's what you're trying to say, but what it illustrates to me is how easy it is to quickly cook up a simple regex to filter out certain patterns in email addresses and feel like it's doing the job, while unintentionally breaking things for a large number of people you don't have on your radar.
      - throw0101d 2y ago
        > […] while unintentionally breaking things for a large number of people you don't have on your radar. And you can unintentionally cause sub-addressing to not work without having "to go out of your way to make that not work" like the GGGP wrote.
        dxdm 2y ago
        I think I misunderstood your point before. I thought you wanted to show a regex that _intentionally_ disables it. It also happens to disable a lot more, making it a bad choice. Now I'm thinking that you might actually agree with that.
      - codetrotter 2y ago
        For example, an email like David.Calhoun@boeing.com would be completely normal. Yet the above regex would not match it. Whoops. Likewise, can_you_believe_it@hotmail.com which is also completely normal and typical, would fail the regex as well. But I guess if the only kind of customers you want are the fred@hotstartup.com and jane@some-vc.com people then sure, this otherwise wholly inadequate regex would serve you well.
- LocalPCGuy 2y ago
  This is why my preferred email regex is just .*@.* (I've been talked into .*@.*\..{2,} before as a slightly more restrictive pattern). And I realize there may be times/places this is not satisfactory, but it works for most general cases.

The actual relevant question is: how many of you make a point of disallowing this? You have to go out of your way to make that not work, given that + is a valid character in email addresses.

throw0101d 2y ago

> You have to go out of your way to make that not work, given that + is a valid character in email addresses. Per another comment, all one would need for it to not work is have a regex of [a-zA-Z0-9]@[a-z.]+, or using POSIX-compat, [:alnum:]. * https://en.wikipedia.org/wiki/Regular_expression#Character_classes https://en.wikipedia.org/wiki/Regular_expression#Character_c...

nielsole 2y ago
> [a-zA-Z0-9]@[a-z.]+ Only a small minority of my email addresses I've ever had matches that pattern.
- throw0101d 2y ago
  > Only a small minority of my email addresses I've ever had matches that pattern. The example is illustrative, and not definitive, as to how the simple and innocuous decision about a regex can determine functionality, and it is not about someone going "out of your way to make that not work" to break things.
  - dxdm 2y ago
    I don't know if that's what you're trying to say, but what it illustrates to me is how easy it is to quickly cook up a simple regex to filter out certain patterns in email addresses and feel like it's doing the job, while unintentionally breaking things for a large number of people you don't have on your radar.
    - throw0101d 2y ago
      > […] while unintentionally breaking things for a large number of people you don't have on your radar. And you can unintentionally cause sub-addressing to not work without having "to go out of your way to make that not work" like the GGGP wrote.
      - dxdm 2y ago
        I think I misunderstood your point before. I thought you wanted to show a regex that _intentionally_ disables it. It also happens to disable a lot more, making it a bad choice. Now I'm thinking that you might actually agree with that.
    - codetrotter 2y ago
      For example, an email like David.Calhoun@boeing.com would be completely normal. Yet the above regex would not match it. Whoops. Likewise, can_you_believe_it@hotmail.com which is also completely normal and typical, would fail the regex as well. But I guess if the only kind of customers you want are the fred@hotstartup.com and jane@some-vc.com people then sure, this otherwise wholly inadequate regex would serve you well.
LocalPCGuy 2y ago
This is why my preferred email regex is just .*@.* (I've been talked into .*@.*\..{2,} before as a slightly more restrictive pattern). And I realize there may be times/places this is not satisfactory, but it works for most general cases.