Is "ISO8859-1" an acceptable variation/alias for "ISO-8859-1"

Question

I've got an application that sends an email notification. When the email is generated, it includes the following in the mime source:

Content-Type: text/plain;
      charset="ISO8859-1"
Content-Transfer-Encoding: quoted-printable

I've noticed that other email programs and open-source conversion tools (like iconv) don't support that specific spelling and instead require "ISO-8859-1".

I don't see "ISO8859-1" specifically listed on the IANA character set list: https://www.iana.org/assignments/character-sets/character-sets.xhtml

So my question is:

Is ISO8859-1 an acceptable variation name of ISO-8859-1 and is there some sort of RFC or standard available to definitively "prove" that one way or the other?

Answer 1

The IANA registry mentioned in the question cites RFC 2978, which in turn cites several RFCs which define how character encodings are to be specified in the Internet. Thus, since ISO8859-1 is not listed there, it is not correct to use it.

Programs may still accept it, as part of their error recovery, but they are not required to do so. Programs may do better error recovery, upon encountering an undefined character encoding name, by inspecting the actual content of text data and trying to make a guess on the encoding. Or they may simply fall back to some default encoding they use, and this may well be ISO-8859-1 (or, in fact, more often windows-1252).