ISO-2022-JP-2: Multilingual Extension of ISO-2022-JP
A 7-bit text encoding that lets one email or news message mix Japanese, Chinese, Korean, Greek and Latin-1 text.
Ohta, M. and K. Handa, "ISO-2022-JP-2: Multilingual Extension of ISO-2022-JP", RFC 1554, DOI 10.17487/RFC1554, December 1993, <https://www.rfc-editor.org/info/rfc1554>.
@misc{rfc1554,
series = {Request for Comments},
number = 1554,
howpublished = {RFC 1554},
publisher = {RFC Editor},
doi = {10.17487/RFC1554},
url = {https://www.rfc-editor.org/info/rfc1554},
author = {M. Ohta and K. Handa},
title = {{ISO-2022-JP-2: Multilingual Extension of ISO-2022-JP}},
pagetotal = 6,
year = 1993,
month = dec,
} RFC text
Contents
Network Working Group M. Ohta
Request for Comments: 1554 Tokyo Institute of Technology
Category: Informational K. Handa
ETL
December 1993
ISO-2022-JP-2: Multilingual Extension of ISO-2022-JP
Status of this Memo
This memo provides information for the Internet community. This memo
does not specify an Internet standard of any kind. Distribution of
this memo is unlimited.
Introduction
This memo describes a text encoding scheme: "ISO-2022-JP-2", which is
used experimentally for electronic mail [RFC822] and network news
[RFC1036] messages in several Japanese networks. The encoding is a
multilingual extension of "ISO-2022-JP", the existing encoding for
Japanese [2022JP]. The encoding is supported by an Emacs based
multilingual text editor: MULE [MULE].
The name, "ISO-2022-JP-2", is intended to be used in the "charset"
parameter field of MIME headers (see [MIME1] and [MIME2]).
Description
The text with "ISO-2022-JP-2" starts in ASCII [ASCII], and switches
to other character sets of ISO 2022 [ISO2022] through limited
combinations of escape sequences. All the characters are encoded
with 7 bits only.
At the beginning of text, the existence of an announcer sequence:
"ESC 2/0 4/1 ESC 2/0 4/6 ESC 2/0 5/10" is (though omitted) assumed.
Thus, characters of 94 character sets are designated to G0 and
invoked as GL. C1 control characters are represented with 7 bits.
Characters of 96 character sets are designated to G2 and invoked with
SS2 (single shift two, "ESC 4/14" or "ESC N").
For example, the escape sequence "ESC 2/4 2/8 4/3" or "ESC $ ( C"
indicates that the bytes following the escape sequence are Korean KSC
characters, which are encoded in two bytes each. The escape sequence
"ESC 2/14 4/1" or "ESC . A" indicates that ISO 8859-1 is designated
to G2. After the designation, the single shifted sequence "ESC 4/14
4/1" or "ESC N A" is interpreted to represent a character "A with
acute".
The following table gives the escape sequences and the character sets
used in "ISO-2022-JP-2" messages. The reg# is the registration number
in ISO's registry [ISOREG].
94 character sets
reg# character set ESC sequence designated to
------------------------------------------------------------------
6 ASCII ESC 2/8 4/2 ESC ( B G0
42 JIS X 0208-1978 ESC 2/4 4/0 ESC $ @ G0
87 JIS X 0208-1983 ESC 2/4 4/2 ESC $ B G0
14 JIS X 0201-Roman ESC 2/8 4/10 ESC ( J G0
58 GB2312-1980 ESC 2/4 4/1 ESC $ A G0
149 KSC5601-1987 ESC 2/4 2/8 4/3 ESC $ ( C G0
159 JIS X 0212-1990 ESC 2/4 2/8 4/4 ESC $ ( D G0
96 character sets
reg# character set ESC sequence designated to
------------------------------------------------------------------
100 ISO8859-1 ESC 2/14 4/1 ESC . A G2
126 ISO8859-7(Greek) ESC 2/14 4/6 ESC . F G2
For further information about the character sets and the escape
sequences, see [ISO2022] and [ISOREG].
If there is any G0 designation in text, there must be a switch to
ASCII or to JIS X 0201-Roman before a space character (but not
necessarily before "ESC 4/14 2/0" or "ESC N ' '") or control
characters such as tab or CRLF. This means that the next line starts
in the character set that was switched to before the end of the
previous line. Though the designation to JIS X 0201-Roman is allowed
for backward compatibility to "ISO-2022-JP", its use is discouraged.
Applications such as pagers and editors which randomly seek within a
text file encoded with "ISO-2022-JP-2" may assume that all the lines
begin with ASCII, not with JIS X 0201-Roman.
At the beginning of a line, information on G2 designation of the
previous line is cleared. New designation must be given before a
character in 96 character sets is used in the line.
The text must end in ASCII designated to G0.
As the "ISO-2022-JP", and thus, "ISO-2022-JP-2", is designed to
represent English and modern Japanese, left-to-right directionality
is assumed if the text is displayed horizontally.
Users of "ISO-2022-JP-2" must be aware that some common transport
such as old Bnews can not relay a 7-bit value 7/15 (decimal 127),
which is used to encode, say, "y with diaeresis" of ISO 8859-1.
Other restrictions are given in the Formal Syntax section below.
Formal Syntax
The notational conventions used here are identical to those used in
STD 11, RFC 822 [RFC822].
The * (asterisk) convention is as follows:
l*m something
meaning at least l and at most m somethings, with l and m taking
default values of 0 and infinity, respectively.
message = headers 1*(CRLF text)
; see also [MIME1] "body-part"
; note: must end in ASCII
text = *(single-byte-char /
g2-desig-seq /
single-shift-char)
[*segment
reset-seq
*(single-byte-char /
g2-desig-seq /
single-shift-char ) ]
; note: g2-desig-seq must
; precede single-shift-char
headers = <see [RFC822] "fields" and [MIME1] "body-part">
segment = single-byte-segment / double-byte-segment
single-byte-segment = single-byte-seq
*(single-byte-char /
g2-desig-seq /
single-shift-char )
double-byte-segment = double-byte-seq
*((one-of-94 one-of-94) /
g2-desig-seq /
single-shift-char )
reset-seq = ESC "(" ( "B" / "J" )
single-byte-seq = ESC "(" ( "B" / "J" )
double-byte-seq = (ESC "$" ( "@" / "A" / "B" )) /
(ESC "$" "(" ( "C" / "D" ))
g2-desig-seq = ESC "." ( "A" / "F" )
single-shift-seq = ESC "N"
single-shift-char = single-shift-seq one-of-96
CRLF = CR LF
; ( Octal, Decimal.)
ESC = <ISO 2022 ESC, escape> ; ( 33, 27.)
SI = <ISO 2022 SI, shift-in> ; ( 17, 15.)
SO = <ISO 2022 SO, shift-out> ; ( 16, 14.)
CR = <ASCII CR, carriage return>; ( 15, 13.)
LF = <ASCII LF, linefeed> ; ( 12, 10.)
one-of-94 = <any one of 94 values> ; (41-176, 33.-126.)
one-of-96 = <any one of 96 values> ; (40-177, 32.-127.)
7BIT = <any 7-bit value> ; ( 0-177, 0.-127.)
single-byte-char = <any 7BIT, including bare CR & bare LF, but NOT
including CRLF, and not including ESC, SI, SO>
MIME Considerations
The name given to the character encoding is "ISO-2022-JP-2". This
name is intended to be used in MIME messages as follows:
Content-Type: text/plain; charset=iso-2022-jp-2
The "ISO-2022-JP-2" encoding is already in 7-bit form, so it is not
necessary to use a Content-Transfer-Encoding header. It should be
noted that applying the Base64 or Quoted-Printable encoding will
render the message unreadable in non-MIME-compliant software.
"ISO-2022-JP-2" may also be used in MIME headers. Both "B" and "Q"
encoding could be useful with "ISO-2022-JP-2" text.
References
[ASCII] American National Standards Institute, "Coded character set
-- 7-bit American national standard code for information
interchange", ANSI X3.4-1986.
[ISO2022] International Organization for Standardization (ISO),
"Information processing -- ISO 7-bit and 8-bit coded
character sets -- Code extension techniques",
International Standard, Ref. No. ISO 2022-1986 (E).
[ISOREG] International Organization for Standardization (ISO),
"International Register of Coded Character Sets To Be Used
With Escape Sequences".
[MIME1] Borenstein, N., and N. Freed, "MIME (Multipurpose Internet
Mail Extensions) Part One: Mechanisms for Specifying and
Describing the Format of Internet Message Bodies", RFC 1521,
September 1993.
[MIME2] Moore, K., "MIME (Multipurpose Internet Mail Extensions) Part
Two: Message Header Extensions for Non-ASCII Text", RFC 1522,
September 1993.
[RFC822] Crocker, D., "Standard for the Format of ARPA Internet Text
Messages", STD 11, RFC 1522, UDEL, August 1982.
[RFC1036] Horton M., and R. Adams, "Standard for Interchange of
USENET Messages", RFC 1036, AT&T Bell Laboratories, Center
for Seismic Studies, December 1987.
[2022JP] Murai, J., Crispin, M., and E. van der Poel, "Japanese
Character Encoding for Internet Messages", RFC 1468, June
1993.
[MULE] Nishikimi, M., Handa, K., and S. Tomura, "Mule: MULtilingual
Enhancement to GNU Emacs", Proc. of INET'93, August, 1993.
Acknowledgements
This memo is the result of discussion between various people in a
news group: fj.kanji and is reviewed by a mailing list: jp-msg
@iij.ad.jp. The Authors wish to thank in particular Prof. Eiichi
Wada for his suggestions based on profound knowledge in ISO 2022 and
related standards.
Security Considerations
Security issues are not discussed in this memo.
Authors' Addresses
Masataka Ohta
Tokyo Institute of Technology
2-12-1, O-okayama, Meguro-ku,
Tokyo 152, JAPAN
Phone: +81-3-5499-7084
Fax: +81-3-3729-1940
EMail: mohta@cc.titech.ac.jp
Ken'ichi Handa
Electrotechnical Laboratory
Umezono 1-1-4, Tsukuba,
Ibaraki 305, JAPAN
Phone: +81-298-58-5916
Fax: +81-298-58-5918
EMail: handa@etl.go.jp
AI ELI5
On this page
§1A Multilingual Extension
RFC 1554 describes ISO-2022-JP-2, a text encoding used experimentally for electronic mail and network news messages in several Japanese networks. It extends ISO-2022-JP, the existing Japanese encoding, and is supported by an Emacs-based multilingual editor called MULE.
The name is meant for the charset parameter of MIME headers.
It is an upgrade of the Japanese text code so one message can also hold other languages.
§2How the Encoding Works
Text starts in ASCII and switches to other ISO 2022 character sets using a limited set of escape sequences. All characters are encoded with 7 bits. 94-character sets are designated to G0 and invoked as GL. 96-character sets are designated to G2 and used one character at a time with the single shift ESC N.
For example, ESC $ ( C announces that the following bytes are Korean KSC characters, two bytes each. ESC . A designates ISO 8859-1 to G2, and then ESC N A stands for the character A with acute.
Escape sequences announce which set follows, much like switching a keyboard layout; a decoder reads them to know how to interpret each byte.
A secret marker says 'the next letters are Korean', and another marker switches back to plain English letters.
§3The Character Sets and Escape Sequences
The RFC lists seven 94-character sets designated to G0: ASCII, JIS X 0208-1978, JIS X 0208-1983, JIS X 0201-Roman, GB2312-1980, KSC5601-1987 and JIS X 0212-1990. It also lists two 96-character sets designated to G2: ISO 8859-1 and ISO 8859-7 (Greek).
Use of JIS X 0201-Roman is allowed for backward compatibility with ISO-2022-JP but is discouraged.
There is a short menu of alphabets you can switch to, and each has its own marker.
- ESC ( BASCII — Designated to G0 (registration 6).
- ESC $ @JIS X 0208-1978 — Designated to G0 (registration 42).
- ESC $ BJIS X 0208-1983 — Designated to G0 (registration 87).
- ESC ( JJIS X 0201-Roman — Designated to G0 (registration 14); discouraged.
- ESC $ AGB2312-1980 — Designated to G0 (registration 58).
- ESC $ ( CKSC5601-1987 — Designated to G0 (registration 149).
- ESC $ ( DJIS X 0212-1990 — Designated to G0 (registration 159).
- ESC . AISO8859-1 — Designated to G2 (registration 100).
- ESC . FISO8859-7 (Greek) — Designated to G2 (registration 126).
§4Line and Text Rules
If any G0 designation appears in the text, there must be a switch to ASCII or JIS X 0201-Roman before a space character or control characters such as tab or CRLF. So the next line starts in the character set switched to before the previous line ended. Programs that seek randomly in a file, such as pagers and editors, may assume lines begin with ASCII.
At the start of each line, the G2 designation of the previous line is cleared, so a new one must be given before any 96-character-set character is used. The text must end in ASCII designated to G0. Left-to-right direction is assumed for horizontal display.
The RFC also gives a formal syntax in the notation of RFC 822. It warns that some transports, such as old Bnews, cannot relay the 7-bit value 7/15 (decimal 127), which encodes, for example, y with diaeresis of ISO 8859-1.
Every line must start and end back in plain letters, so a reader can jump to any line and still make sense of it.
§5Use with MIME
In a MIME message the encoding is declared as Content-Type: text/plain; charset=iso-2022-jp-2. Because the encoding is already 7-bit, no Content-Transfer-Encoding header is needed, and applying Base64 or Quoted-Printable would make the message unreadable in software that is not MIME-compliant. The encoding may also be used in MIME headers, where both B and Q encoding could be useful.
The label on the envelope says which code the letter uses, and the letter needs no extra wrapping.