Email Address Standards

[jesh]

Does anyone know what organizing body manages the standards for the format of an email address? I checked the W3C but wasn't able to find the information there. Is there even such a body or is it ruled by anarchy? :)I'm not having any luck with Google because just about every website on the Internet has "email address" on it somewhere.EDIT: To clarify, I want to be able to validate an email address using regular expressions. I don't need to know how to write the regular expression - I can do that on my own. What I'm looking for is a definition of what, exactly, constitutes a valid email address. For example, is jesh_@somedomain.com a valid email address or not? What about jesh___________________@somedomain.com? Is there a standard out there?

[aspnetguy]

you mean whether it should befirst.last@yourdomain.comflast@yourdomain.comIs that what you mean?

[jesh]

I think I was adding my edit while you were replying. I want to know if there is a standard regarding the format of the local-part (i.e. "jesh" in jesh@somedomain.com).

[pulpfiction]

Is this something you are looking for....http://tools.ietf.org/html/rfc2822#section-3.4.1check this out too...http://en.wikipedia.org/wiki/E-mail_address

[jesh]

Hmm, I managed to find a few links:http://www.cs.tut.fi/~jkorpela/rfc/822addr.htmlhttp://www.freesoft.org/CIE/RFC/822/index.htmhttp://www.faqs.org/rfcs/rfc2822.htmlBasically, what I am beginning to understand is, there isn't a standard.

The local-part of an addr-spec in a mailbox specification (i.e., the host's name for the mailbox) is understood to be whatever the receiving mail protocol server allows.

And

A host that is forwarding the message but is not the destination host implied by the right-hand side "domain" must not interpret or modify the "local-part" of the address.

So, this means that "jesh$7____23^! bring it!"@somedomain.com is a valid email address if somedomain.com accepts it as such?

Is this something you are looking for....http://tools.ietf.org/html/rfc2822#section-3.4.1check this out too...http://en.wikipedia.org/wiki/E-mail_address

Thanks! I'll continue reading.

[justsomeguy]

There isn't a standard per se, these types of things are governed by RFCs. I see there's already a link for RFC 2822, which does define email addresses. They aren't standards so much as definitions. They are defined using a grammar, like this:addr-spec = local-part "@" domainAn addr-spec is defined as a "local-part", then "@", then "domain". A "local-part" is defined this way:dot-atom / quoted-string / obs-local-partA dot-atom is defined as dot-atom-text, which is defined like this:1*atext *("." 1*atext)So it is 1 or more atext, followed by 0 or more of the group [dot, 1 or more atext]. atext is any of these:

atext		   =	   ALPHA / DIGIT /; Any character except controls,						"!" / "#" /	;  SP, and specials.						"$" / "%" /	;  Used for atoms						"&" / "'" /						"*" / "+" /						"-" / "/" /						"=" / "?" /						"^" / "_" /						"`" / "{" /						"|" / "}" /						"~"

A local part can also be a quoted-string, which is loosely defined as double quotes enclosing an alphanumeric string, minus a few special characters. The domain part can be a dot-atom or a domain-literal. Basically, what all this means is that an address is any number of dot-atoms with a @ in there somewhere. So these are all legal:a@a.aa.a@a.aa@a.a.a.a.a.a.a.a.aaaa.aaa.aaa.aaa.aaa.aaa@a.aMultiple dots in a row are not allowed, since a dot atom is defined as a dot followed by one or more atext characters. And a dot cannot appear at the end, since a dot atom has to be a dot followed by atext.So, in terms of a regexp, you can think of it as something like this (using the same grammar as above):atext (0 or more ["." atext]) "@" atext (0 or more ["." atext])

[jesh]

Basically, what all this means is that an address is any number of dot-atoms with a @ in there somewhere. So these are all legal:a@a.aa.a@a.aa@a.a.a.a.a.a.a.a.aaaa.aaa.aaa.aaa.aaa.aaa@a.a

So, too, would be ____________________________________@a.a.Thanks everyone for your help. I was just having the hardest time figuring out how to word my search.@pulpfiction - I had even tried wikipedia, I guess I just didn't think of searching for something as freakishly simple as "Email Address". All the links helped, but I think this one helped the most: http://en.wikipedia.org/wiki/E-mail_address

[pulpfiction]

@pulpfiction - I had even tried wikipedia, I guess I just didn't think of searching for something as freakishly simple as "Email Address". All the links helped, but I think this one helped the most: http://en.wikipedia.org/wiki/E-mail_address

Yeah sometimes it happens, glad could help you...

[reportingsjr]

So, too, would be ____________________________________@a.a.

No, the "atom dot" or whatever it is called has to be followed by a letter or whatever Steve said.Offtopic: Pulpfiction, I have been wondering if you based your name off of the movie by Quentin Tarintino? Good movie

[justsomeguy]

No, the "atom dot" or whatever it is called has to be followed by a letter or whatever Steve said.

I think the dot at the end is for the end-of-sentence, not necessarily part of the pattern. But that's correct, the address cannot end in a dot.

[jesh]

No, the "atom dot" or whatever it is called has to be followed by a letter or whatever Steve said.

Not according to this:

According to RFC 2822, the local-part of the e-mail may use any of these ASCII characters: * Uppercase and lowercase letters (case insensitive) * The digits 0 through 9 * The characters ! # $ % & ' * + - / = ? ^ _ ` { | } ~ * The character "." provided that it is not the first or last character in the local-part.

The atom can be made up of any of the above-mentioned characters except for the period (.). So this too should be a valid email address:$$$@somedomain.comOr even (eek):`'`@somedomain.com

[justsomeguy]

That's right, but the dot-atom is defined to be a dot followed by any number of "atext" (which is what you have listed up there, other then the dot). So all of these are a "dot-atom":.asdf.a.1.!#$%^&*()But you can't have a dot without an atext following it.