But RFC 1468, which I quoted earlier, is also clear that the text of an
ISO-2022-JP message may contain:
single-byte-char = <any 7BIT, including bare CR & bare LF...
And that will include NUL as well.
The important question is: do iso-2022-jp documents, *in practice*,
contain occurances of 0x00, bare 0x0a or bare 0x0d? (Only the
documents that do use these codes would need to be encoded.)
Analogy: NUL is defined in the US-ASCII character set. This does not
prevent US-ASCII body parts that do not contain NUL, from using the
7bit content-transfer-encoding.
Keith