2

I have just started learning to code both PHP as well as HTML and had a look at a few tutorials on regular expressions however have a hard time understanding what these mean. I appreciate any help.

For example, I would like to validate the email address [email protected]. I start off with the code and I get the message invalid email address.

  1. What am I doing wrong?
  2. I know that the metacharacters such as ^ denote the start of a string and $ denote the end of a string however what does this mean? What is the start of a string and what is the end of a string?
  3. When do I group regular expressions?

 

$emailaddress = '[email protected]'; if(preg_match('/^[a-zA-z0-9]+@[a-zA-z0-9]+\.[a-zA-z0-9]$/', $emailaddress)) { echo 'Great, you have a valid email address'; } else { echo 'boo hoo, you have an invalid email address'; }
Improve this question
3
  • 4
    I'd recommend using a tool such as gskinner.com/RegExr to help understand and craft regular expressions Commented Apr 20, 2011 at 22:21
  • 1
    Rather than specifying A-Za-z, make the regex case-insensitive with the /i modifier. Commented Apr 20, 2011 at 22:32
  • What about: [email protected] thats a valid e-mail that fails your test. Commented Apr 20, 2011 at 23:21

6 Answers 6

Reset to default
5

What you have written works with some small modifications if that is what you want to use, however you miss a '+' at the end.

1)

 ^[a-zA-Z0-9]+@[a-zA-Z0-9]+\.[a-zA-Z0-9]+$

The caret and dollar character match positions rather than characters, ^ is equal to the beginning of line and $ is equal to the end of line, they are used to anchor your regex. If you write your regex without those two you will match email addresses everywhere in your text, not only the email addresses which is on a single line in this case. If you had written only the ^ (caret) you would have found every email address which is on the start of the line and if you had written only the $ (dollar) you would have found only the email addresses on the end of the line.

Blah blah blah [email protected] blah blah

would not give you a match because you do NOT have a email address at the beginning of line and the line does not terminate with it either so in order to match it in this context you would have to drop ^ and $.

  1. Grouping is used for two reasons as far I know: Back referencing and... grouping. Grouping is used for the same reasons as in math, 1 + 3 * 4 is not the same as (1 + 3) * 4. You use parentheses to constrain quantifiers such as '+', '*' and '?' as well as alternation '|' etc.

You also parentheses for back referencing, but since I can't explain it better I would link you to: http://www.regular-expressions.info/brackets.html

I will encourage you to take a look at this book, even though you only read the first 2-3 chapters you will learn a lot and it is a great book! http://oreilly.com/catalog/9781565922570

And as the commentators say, this regex is not perfect but it works and show you what you had forgotten. You were not far away!

UPDATED as requested:

The '+', '*' and '?' are quantifiers. And is also a good example where you group.

  • '+' mean match whatever charachter preceeds it or group 1 or n times.
  • '*' mean match whatever charachter preceeds it 0 or n times.
  • '?' mean match whatever charachter preceeds it or the group 0 or 1 time.

n times meaning (indefinitely)

The reason why you use [a-zA-Z0-9]+ is without the '+' it will only match one character. With the + it will match many but it must match at least one. With * it match many but also 0, and ? will match 1 character at most but also 0.

Improve this answer
Sign up to request clarification or add additional context in comments.

29 Comments

This is the correct answer :) Just answer 2) and 3) and you're done!
Codemonkey's suggested regex is way better than this one!
@Bodgan Agree, but this changes completely the logic of the OP's regex. There are lots of regex for validating emails, but I don't think that is the idea. The OP was asking: What am I doing wrong?. I think he prefers learning Regular Expressions than just validating an email address.
well said @Oscar - @Codemonkey's regex is much better, but still not good. It is very difficult to match an email address, especially as the specification changes over time, and there are some very strange use cases - like + signs in gmail addresses used for filtering for example. Generally I think validation is over-used and does not help anyone. It does not validate that the email exists - only that it could exist at best - and even then, I have never seen a watertight regex to do that.
I think reading a thorough tutorial is what you need. If you read through this tutorial and check out the examples, this comment thread will likely be redundant.
|
5

Your regex doesn't match email addresses. Try this one:

/\b[\w\.-]+@[\w\.-]+\.\w{2,4}\b/

I recommend you read through this tutorial to learn about Regular Expressions.

Also, RegExr is great for testing them out.

As for your second question; the ^ character means that the regular expression must start matching from the first character in the string you input. The $ means that the regular expression must end at the final character in the string you input. In essence, this means that your regular expression will match the following string:

[email protected]

but NOT the following string:

My email address is [email protected], and I love it!

Grouping regular expressions has lots of use cases. Using matching groups will also make your expression cleaner and more readable. It's all explained quite well in the tutorial I linked earlier.

As CanSpice points out, matching all possible email addresses isn't all that easy. Using the RFC2822 Email Validation expression will do a better job:

/[a-z0-9!#$%&'*+/=?^_`{|}~-]+(?:\.[a-z0-9!#$%&'*+/=?^_`{|}~-]+)*@(?:[a-z0-9](?:[a-z0-9-]*[a-z0-9])?\.)+[a-z0-9](?:[a-z0-9-]*[a-z0-9])?/

There are many alternatives, but even the simplest ones will do a fair job as most email addresses end in .com (or other 2-4 character length top domains).

The only reason your original expression doesn't work is that you're limiting the number of characters behind the period (.) in your expressions to 1. Changing your expression to:

/^[a-zA-z0-9]+@[a-zA-z0-9]+\.[a-zA-z0-9]+$/

Will allow for an infinite amount of characters behind the last period.

/^[a-zA-z0-9]+@[a-zA-z0-9]+\.[a-zA-z0-9]{2,4}$/

Will allow 2 to 4 characters behind the last period. That would match:

[email protected]

[email protected]

but not:

[email protected]

Improve this answer

3 Comments

That regex doesn't match all email addresses either.
Thanks. I am still stumbling to understand why My email address is [email protected], and I love it! would not work if they are in/on the same line.
Because ^p means that the beginning of the string input MUST be the letter p. Similarly, p$ means that the input string MUST end with the letter p. In other words, putting ^ and $ at the beginning and end of an expression, respectively, means that the expression MUST match the entire input string. So "[email protected]" will be matched successfully, but "My address is [email protected]!" will not, because the string contains MORE than just the address. Do you understand now? :)
1

The top level domain (".com," ".net," ".museum") can be from 2 to 6 characters. So you should be saying 2,6 instead of 2,4.

I wrote an extremely good email address regular expression a few years ago:

^\w+([-+._]\w+)@(\w+((-+)|.))\w{1,63}.[a-zA-Z]{2,6}$

A lot of research went into that. But I have some basic tips:

DON'T JUST COPY-PASTE! If someone says "here's a great regex for that," don't just copy paste it! Understand what's going on! Regular expressions are not that hard. And once you learn them well, it'll pay dividends forever. I got good at them by taking a class in Perl back in college. Since then, I've barely gotten any better and am WAY better than the vast majority of programmers I know. It's sad. Anyways, learn it!

Start small. Instead of building a giant regex and testing it when you're done, test just a few characters. For example, when writing an email validator, why not try \w+@\w+.\w+ and see how good that is? Add in a few more things and re-test. Like ^\w+@\w+.[A-Za-z]{2,6}$

Improve this answer

Comments

0

The start and end of a regex string means that nothing can come before or after the characters you specify. Your regex string needs to account for underscores, needs capitals Zs with your capital ranges, and other adjustments.

/^[a-zA-Z_0-9]+@[a-zA-Z0-9]+\.[a-zA-z0-9]{2,4}$/

{2,4} says the top level domain is between 2 and 4 characters.

Improve this answer

2 Comments

@Andreas - Thanks. Sorry, I don't follow what you mean by the start and end of a string. As my commet to Oscar Mederos does it mean start] This is my email address [email protected]. This is also my email address [email protected] [end] assuming that they are on the same line?
If I said '\d{1}' instead of '^\d{1}$', then it would accept a digit plus anything before or after the digit. The latter makes sure you get only the one digit (nothing before or after it).
0

This will validate ANY email address (at least i've tried a lot )

preg_match("/^[a-z0-9._-]{2,}+\@[a-z0-9_-]{2,}+\.([a-z0-9-]{2,4}|[a-z0-9-]{2,}+\.[a-z0-9-]{2,4})$/i", $emailaddress);

Hope it works!

Improve this answer

1 Comment

Doesn't match üñîçøðé@travel for example. See en.wikipedia.org/wiki/Email_address#Valid_email_addresses
-1

Make sure you ALWAYS escape metacharacters (like dot):

if(preg_match('/^[a-zA-z0-9]+@[a-zA-z0-9]+\.[a-zA-z0-9]$/', $emailaddress)) {
Improve this answer

1 Comment

yeap, I forgot things that our friends already have meantioned; like plus after last list [a-zA-z0-9]+

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.