Help understanding preg_match ?

[ShiggyWiggy]

I'm using a function that checks if an e-mail address is of a valid format eg [e-mail].

It uses the following to check if its in a valid format.

(!preg_match("/^.+@([?) [a-zA-Z0-9-.]+.([a-zA-Z]{2,3}|[0-9]{1,3})(]?)$/", $email))

I got this code from a book but it dosn't explain it too well and I can't really work out whats happening using the manual so I was hoping someone on here might be able to break it down and point out what exactly it does.

Thanks

[laserlight]

/ => starting delimiter
Here's what I think it should be, when broken down:

^ => assert start of subject (i.e. pattern matched must start here)
.+ => match 1 or more of any character
@ => literal '@'
([?) => subpattern that matches 0 or 1 of '['
[a-zA-Z0-9-.]+ => match one or more alphanumeric characters, '-' and '.'
. => literal '.'
( => start subpattern
[a-zA-Z]{2,3} => match 2 or 3 alphabetic characters
| => or
[0-9]{1,3} => match 1 to 3 numeric characters
) => end subpattern
(]?) => subpattern that matches 0 or 1 of ']'
$ => assert end of subject (i.e. pattern matched must end here)
/ => closing delimiter

These are some examples that I think it should match:
[e-mail]
[e-mail]
test@[example.com]
[e-mail]
[e-mail]
[e-mail]
test@.example.123
test@[example.1
test@example.12]

These are some examples that I think it shouldnt match:
example
test@example
test@example.
test@example.a
[e-mail]
test@example.1234

Note that I'm not particularly good at regex, so I could well be blatantly wrong.

[ShiggyWiggy]

Thanks

Makes a lot more sense now.

[bodzan]

...go to this online tutorial....

[mtmosier]

I found it rather funny that the above tutorial link goes to "regular-expressions.info", a domain which this regular expression would reject as invalid.

Might consider changing it to read:

(!preg_match("/^.+\@(\[?) [a-zA-Z0-9\-\.]+\.([a-zA-Z]{2,4}|[0-9]{1,3})(\]?)$/", $email))