[Solar-talk] ValidateEmail filter

[Robert Gonzalez]

You know, I may just be wrong about this formatting thing. I had read 
once (or so I thought) that a domain name could be valid even if there 
is not .extension on it (as in localhost or other variants of mapped 
hostnames). The Email RFC says that the domain part must follow the RFC 
for domain names. But there is nothing in the email RFC that would 
suggest that localhost (or some other not dot separated hostname) is a 
valid domain part. In one place I read that the domain part consists of 
one or more dot separated components.

So before I make a complete fool of myself stating something this 
emphatically I think it would be best if I (or someone else) found out 
for sure that this is the case.

I do hold to my stance that user at hostname is valid. But I kinda want to 
prove it to myself now. Of course, it is a little after 0200 right now, 
so I could just be rambling because I am tired. :)

Robert

Robert Gonzalez wrote:
> What would happen if Solar were used as an intranet backbone where 
> emails were sent to just the domain name? If name at host didn't validate 
> the framework would not be doing something it should do, which is 
> validate the email address per the spec.
>
> I am not trying to be argumentative. I am just saying that name at domain 
> (even if domain is x) is still a perfectly valid email address and as 
> such should be considered valid by any validator.
>
> Robert
>
> Jeff Surgeson wrote:
>>>>> Is there a reason why validateEmail() filter no longer forces the domain
>>>>> part to include a dot com or dot whatever. When it was part of
>>>>> Solar_DataFilter it did but now not. 
>>>>>         
>>
>>   
>>>> If I remember correctly the .domain part is not a requirement in an email
>>>> address. But I think that a dot at the end is invalid per the email spec.
>>>> I could be wrong.
>>>>
>>>> Robert
>>>>       
>>
>>   
>>> I would agree. If there a period after the hostname it must be followed by
>>> a suffix (com, edu, etc). Forcing the suffix is not entirely a good idea if
>>> you have email addresses like rkolbe at localhost and that is it (or something
>>> to that extent).
>>>     
>>
>> Maybe I am missing something here, in my rather small world if you attempted 
>> to send an email with x at x and no TLD it would not be accepted by any mail 
>> server other than a local intranet or network as far as I am aware. And in my 
>> humble opinion the reason to validate it is to make sure it is a valid usable 
>> email address that can be used to send and receive email, attempting to send 
>> to x at x will fail. So I would say forcing it to validate for the full 
>> domainname made up of SLD and TLD is important.
>>
>> In all the email validation stuff I have looked at, and solar email filter 
>> used to do this as well, it would validate for a minimum of 5 characters, in 
>> the format of: x at x.x
>>
>> This being the shortest possible and still valid email address, so the 
>> question still remains, Solar email filter used to validate it for x at x.x but 
>> no longer does.
>>
>> Is there a reason it was removed that I am not aware of?
>> Or am I just missing the plot?
>>
>>   
> ------------------------------------------------------------------------
>
> _______________________________________________
> Solar-talk mailing list
> Solar-talk at lists.solarphp.com
> http://mailman-mail3.webfaction.com/listinfo/solar-talk
>