wales@UCLA-LOCUS.ARPA (Rich Wales) (09/04/84)
RFC822 says you can have "bare" CR's and LF's in header text. For
example, the definition of the lexical token "text" (Section 3.3, page
10) reads as follows:
text = <any CHAR, including bare ; => atoms, specials,
CR & bare LF, but NOT ; comments and
including CRLF> ; quoted-strings are
; NOT recognized.
Also, the discussion on quoting (Section 3.4.1, page 11) mentions that a
CR must be quoted (i.e., preceded by a backslash) if it occurs within a
quoted string, domain literal, or comment. This is, I suppose, implicit
evidence that someone thinks CR's or CRLF's could appear in addresses.
What I want to know is -- does ANYONE, ANYWHERE, use (or intend to use)
"bare" CR's or LF's in addresses, or indeed anywhere else in a header?
(I am, of course, not talking about the CRLF end-of-line delimiter.)
Rich Wales
UCLA Computer Science Department
3531 Boelter Hall // Los Angeles, CA 90024 // (213) 825-5683
ARPA: wales@UCLA-LOCUS.ARPA
UUCP: ...!{cepu,ihnp4,trwspp,ucbvax}!ucla-cs!walesRudy.Nedved@CMU-CS-A.ARPA (09/04/84)
Rich, CMU does not use the bare cr or bar lf BUT it handles cr, lf and crlf all the same...they are treated as end-of-line and are converted to whatever the local end-of-line convention is. "...be conservative in what you send and liberal in what you receive..." -Rudy
Craig.Everhart@CMU-CS-A.ARPA (09/04/84)
Even if your poll shows that nobody polled intended to use bare CRs or LFs in their addresses, would you feel justified in not providing support for them in your mail handling software? After all, if you fail to support that part of the standard, you can't claim that you follow it. Who knows what somebody will try to use as an address some day? For whatever reason? It sure sounds like interpreting quoted CRs and LFs as other than text (e.g., as a line break) is a bad idea, also.