[objective-c] Check that an email address is valid on iOS

Possible Duplicate:
Best practices for validating email address in Objective-C on iOS 2.0?

I am developing an iPhone application where I need the user to give his email address at login.

What is the best way to check if an email address is a valid email address?

To check if a string variable contains a valid email address, the easiest way is to test it against a regular expression. There is a good discussion of various regex's and their trade-offs at regular-expressions.info.

Here is a relatively simple one that leans on the side of allowing some invalid addresses through: ^[A-Z0-9._%+-]+@[A-Z0-9.-]+\.[A-Z]{2,6}$

How you can use regular expressions depends on the version of iOS you are using.

iOS 4.x and Later

You can use NSRegularExpression, which allows you to compile and test against a regular expression directly.

iOS 3.x

Does not include the NSRegularExpression class, but does include NSPredicate, which can match against regular expressions.

NSString *emailRegex = ...;
NSPredicate *emailTest = [NSPredicate predicateWithFormat:@"SELF MATCHES %@", emailRegex];
BOOL isValid = [emailTest evaluateWithObject:checkString];

Read a full article about this approach at cocoawithlove.com.

iOS 2.x

Does not include any regular expression matching in the Cocoa libraries. However, you can easily include RegexKit Lite in your project, which gives you access to the C-level regex APIs included on iOS 2.0.

Heres a good one with NSRegularExpression that's working for me.

[text rangeOfString:@"^.+@.+\\..{2,}$" options:NSRegularExpressionSearch].location != NSNotFound;

You can insert whatever regex you want but I like being able to do it in one line.

to validate the email string you will need to write a regular expression to check it is in the correct form. there are plenty out on the web but be carefull as some can exclude what are actually legal addresses.

essentially it will look something like this

^((?>[a-zA-Z\d!#$%&'*+\-/=?^_`{|}~]+\x20*|"((?=[\x01-\x7f])[^"\\]|\\[\x01-\x7f])*"\x20*)*(?<angle><))?((?!\.)(?>\.?[a-zA-Z\d!#$%&'*+\-/=?^_`{|}~]+)+|"((?=[\x01-\x7f])[^"\\]|\\[\x01-\x7f])*")@(((?!-)[a-zA-Z\d\-]+(?<!-)\.)+[a-zA-Z]{2,}|\[(((?(?<!\[)\.)(25[0-5]|2[0-4]\d|[01]?\d?\d)){4}|[a-zA-Z\d\-]*[a-zA-Z\d]:((?=[\x01-\x7f])[^\\\[\]]|\\[\x01-\x7f])+)\])(?(angle)>)$

Actually checking if the email exists and doesn't bounce would mean sending an email and seeing what the result was. i.e. it bounced or it didn't. However it might not bounce for several hours or not at all and still not be a "real" email address. There are a number of services out there which purport to do this for you and would probably be paid for by you and quite frankly why bother to see if it is real?

It is good to check the user has not misspelt their email else they could enter it incorrectly, not realise it and then get hacked of with you for not replying. However if someone wants to add a bum email address there would be nothing to stop them creating it on hotmail or yahoo (or many other places) to gain the same end.

So do the regular expression and validate the structure but forget about validating against a service.