[Top] [All Lists]

Re: Parsing a header

2007-12-03 16:04:01

[Not clear on why this is on ietf-smtp instead of ietf-822, but anyway...]

On 12/3/07 at 7:53 PM +0100, Frank Ellermann wrote:

>Paul Smith wrote:
>>syntactically correct formations of the line could be:
>>From: "Joe (Joseph) Bloggs" <joe(_at_)joe(_dot_)com>
>>From: (Joe \(Joseph\) Bloggs) joe(_at_)joe(_dot_)com
>>or even
>>From: Joe (\(Joseph\)) Bloggs <joe(_at_)joe(_dot_)com>
>Or From: "Joe \(Joseph\) Bloggs" <joe(_at_)example>
>or From: Joe (Joseph) Bloggs <joe(_at_)example>

or From: Joe "\(Joseph\)" Bloggs <joe(_at_)example(_dot_)com>
or From: Joe "(Joseph)" Bloggs <joe(_at_)example(_dot_)com>

But Frank is exactly right: The original thing is not a legal phrase
in any of these standards.

Frank is also correct in point out that we don't have rules for how to deal
with such invalid syntax. And specifying such rules is a rathole I'd really
like to avoid.

More generally, there's a vast range of stuff that's close to being allowed but
isn't. I don't have a problem discussing how to handle such cases, but once it
is determined that the material is syntactically invalid it becomes an
operational issue, not a specification issue.

FWIW, I tried this specific example with our heuristic fixup parser and it
converted it to

   Joe Joseph Bloggs <joe(_at_)joe(_dot_)com>

which isn't bad but isn't what I would have guessed it would do. Guess I need
to go review the code since I've clearly forgotten some details.


<Prev in Thread] Current Thread [Next in Thread>