(4) "Content-type: text/plain; charset=ISO646-Bengali-Japanese
"Content-type: text/plain; charset=ISO646-Japanese-Bengali"
The order should be of Annex A of DIS 10646-1.2, where Devanagari
is numbered as 13 and CJK UNIFIED IDEOGRAPHS as 58. So,
(4) "Content-type: text/plain; charset=ISO646-Bengali-Japanese
is the right one.
I don't know about others, but I have pretty severe problems with
moving from "charset parameter is an unformatted string representing a
registration" to "charset parameters starting in ISO-10646- must be
further parsed and interpreted, starting with the second hyphen, to do
language differentiation".
That there are many "charset"s does not mean you must parse it, as long
as they are finite.
% grep ISO646 rfc1345.txt
&alias ISO646-GB
&alias ISO646-US
&alias ISO646-FI
&alias ISO646-SE
&alias ISO646-SE2
&alias ISO646-JP
&alias ISO646-IT
&alias ISO646-PT
&alias ISO646-ES
&alias ISO646-DE
&alias ISO646-FR1
&alias ISO646-CN
&alias ISO646-NO
&alias ISO646-NO2
&alias ISO646-FR
&alias ISO646-PT2
&alias ISO646-ES2
&alias ISO646-HU
&alias ISO646-JP-OCR-B
&alias ISO646-CA
&alias ISO646-CA2
&alias ISO646-YU
&alias ISO646-CU
&alias ISO646-DK
&alias ISO646-KR
Masataka Ohta