<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml"><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8" /><title>Compound Text Encoding</title><meta name="generator" content="DocBook XSL Stylesheets Vsnapshot_9276" /><style xmlns="" type="text/css">/*
* Copyright (c) 2011 Gaetan Nadon
* Copyright (c) 2010, Oracle and/or its affiliates. All rights reserved.
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice (including the next
* paragraph) shall be included in all copies or substantial portions of the
* Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
* DEALINGS IN THE SOFTWARE.
*/
/*
* Shared stylesheet for X.Org documentation translated to HTML format
* http://www.sagehill.net/docbookxsl/UsingCSS.html
* http://www.w3schools.com/css/default.asp
* https://addons.mozilla.org/en-US/firefox/addon/web-developer/developers
* https://addons.mozilla.org/en-US/firefox/addon/font-finder/
*/
/*
* The sans-serif fonts are considered more legible on a computer screen
* http://dry.sailingissues.com/linux-equivalents-verdana-arial.html
*
*/
body {
font-family: "Bitstream Vera Sans", "DejaVu Sans", Tahoma, Geneva, Arial, Sans-serif;
/* In support of using "em" font size unit, the w3c recommended method */
font-size: 100%;
}
/*
* Selection: all elements requiring mono spaced fonts.
*
* The family names attempt to match the proportionally spaced font
* family names such that the same font name is used for both.
* We'd like to use Bitstream, for example, in both proportionally and
* mono spaced font text.
*/
.command,
.errorcode,
.errorname,
.errortype,
.filename,
.funcsynopsis,
.function,
.parameter,
.programlisting,
.property,
.screen,
.structname,
.symbol,
.synopsis,
.type
{
font-family: "Bitstream Vera Sans Mono", "DejaVu Sans Mono", Courier, "Liberation Mono", Monospace;
}
/*
* Books have a title page, a preface, some chapters and appendices,
* a glossary, an index and a bibliography, in that order.
*
* An Article has no preface and no chapters. It has sections, appendices,
* a glossary, an index and a bibliography.
*/
/*
* Selection: book main title and subtitle
*/
div.book>div.titlepage h1.title,
div.book>div.titlepage h2.subtitle {
text-align: center;
}
/*
* Selection: article main title and subtitle
*/
div.article>div.titlepage h2.title,
div.article>div.titlepage h3.subtitle,
div.article>div.sect1>div.titlepage h2.title,
div.article>div.section>div.titlepage h2.title {
text-align: center;
}
/*
* Selection: various types of authors and collaborators, individuals or corporate
*
* These authors are not always contained inside an authorgroup.
* They can be contained inside a lot of different parent types where they might
* not be centered.
* Reducing the margin at the bottom makes a visual separation between authors
* We specify here the ones on the title page, others may be added based on merit.
*/
div.titlepage .authorgroup,
div.titlepage .author,
div.titlepage .collab,
div.titlepage .corpauthor,
div.titlepage .corpcredit,
div.titlepage .editor,
div.titlepage .othercredit {
text-align: center;
margin-bottom: 0.25em;
}
/*
* Selection: the affiliation of various types of authors and collaborators,
* individuals or corporate.
*/
div.titlepage .affiliation {
text-align: center;
}
/*
* Selection: product release information (X Version 11, Release 7)
*
* The releaseinfo element can be contained inside a lot of different parent
* types where it might not be centered.
* We specify here the one on the title page, others may be added based on merit.
*/
div.titlepage p.releaseinfo {
font-weight: bold;
text-align: center;
}
/*
* Selection: publishing date
*/
div.titlepage .pubdate {
text-align: center;
}
/*
* The legal notices are displayed in smaller sized fonts
* Justification is only supported in IE and therefore not requested.
*
*/
.legalnotice {
font-size: small;
font-style: italic;
}
/*
* For documentation having multiple licenses, the copyright and legalnotice
* elements sequence cannot instantiated multiple times.
* The copyright notice and license text are therefore coded inside a legalnotice
* element. The role attribute on the paragraph is used to allow styling of the
* copyright notice text which should not be italicized.
*/
p.multiLicensing {
font-style: normal;
font-size: medium;
}
/*
* Selection: book or article main ToC title
* A paragraph is generated for the title rather than a level 2 heading.
* We do not want to select chapters sub table of contents, only the main one
*/
div.book>div.toc>p,
div.article>div.toc>p {
font-size: 1.5em;
text-align: center;
}
/*
* Selection: major sections of a book or an article
*
* Unlike books, articles do not have a titlepage element for appendix.
* Using the selector "div.titlepage h2.title" would be too general.
*/
div.book>div.preface>div.titlepage h2.title,
div.book>div.chapter>div.titlepage h2.title,
div.article>div.sect1>div.titlepage h2.title,
div.article>div.section>div.titlepage h2.title,
div.book>div.appendix>div.titlepage h2.title,
div.article>div.appendix h2.title,
div.glossary>div.titlepage h2.title,
div.index>div.titlepage h2.title,
div.bibliography>div.titlepage h2.title {
/* Add a border top over the major parts, just like printed books */
/* The Gray color is already used for the ruler over the main ToC. */
border-top-style: solid;
border-top-width: 2px;
border-top-color: Gray;
/* Put some space between the border and the title */
padding-top: 0.2em;
text-align: center;
}
/*
* A Screen is a verbatim environment for displaying text that the user might
* see on a computer terminal. It is often used to display the results of a command.
*
* http://www.css3.info/preview/rounded-border/
*/
.screen {
background: #e0ffff;
border-width: 1px;
border-style: solid;
border-color: #B0C4DE;
border-radius: 1.0em;
/* Browser's vendor properties prior to CSS 3 */
-moz-border-radius: 1.0em;
-webkit-border-radius: 1.0em;
-khtml-border-radius: 1.0em;
margin-left: 1.0em;
margin-right: 1.0em;
padding: 0.5em;
}
/*
* Emphasis program listings with a light shade of gray similar to what
* DocBook XSL guide does: http://www.sagehill.net/docbookxsl/ProgramListings.html
* Found many C API docs on the web using like shades of gray.
*/
.programlisting {
background: #F4F4F4;
border-width: 1px;
border-style: solid;
border-color: Gray;
padding: 0.5em;
}
/*
* Emphasis functions synopsis using a darker shade of gray.
* Add a border such that it stands out more.
* Set the padding so the text does not touch the border.
*/
.funcsynopsis, .synopsis {
background: #e6e6fa;
border-width: 1px;
border-style: solid;
border-color: Gray;
clear: both;
margin: 0.5em;
padding: 0.25em;
}
/*
* Selection: paragraphs inside synopsis
*
* Removes the default browser margin, let the container set the padding.
* Paragraphs are not always used in synopsis
*/
.funcsynopsis p,
.synopsis p {
margin: 0;
padding: 0;
}
/*
* Selection: variable lists, informal tables and tables
*
* Note the parameter name "variablelist.as.table" in xorg-xhtml.xsl
* A table with rows and columns is constructed inside div.variablelist
*
* Set the left margin so it is indented to the right
* Display informal tables with single line borders
*/
table {
margin-left: 0.5em;
border-collapse: collapse;
}
/*
* Selection: paragraphs inside tables
*
* Removes the default browser margin, let the container set the padding.
* Paragraphs are not always used in tables
*/
td p {
margin: 0;
padding: 0;
}
/*
* Add some space between the left and right column.
* The vertical alignment helps the reader associate a term
* with a multi-line definition.
*/
td, th {
padding-left: 1.0em;
padding-right: 1.0em;
vertical-align: top;
}
.warning {
border: 1px solid red;
background: #FFFF66;
padding-left: 0.5em;
}
</style></head><body><div class="article"><div class="titlepage"><div><div><h2 class="title"><a id="ctext"></a>Compound Text Encoding</h2></div><div><h3 class="subtitle"><em>X Consortium Standard</em></h3></div><div><div class="authorgroup"><div class="author"><h3 class="author"><span class="firstname">Robert</span> <span class="othername">W.</span> <span class="surname">Scheifler</span></h3><div class="affiliation"><span class="orgname">X Consortium<br /></span></div></div></div></div><div><p class="releaseinfo">X Version 11, Release 7.7</p></div><div><p class="releaseinfo">Version 1.1</p></div><div><p class="copyright">Copyright © 1989 X Consortium</p></div><div><div class="legalnotice"><a id="id2525274"></a><p>
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
</p><p>
The above copyright notice and this permission notice shall be included in
all copies or substantial portions of the Software.
</p><p>
THE SOFTWARE IS PROVIDED “AS ISâ€, WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
X CONSORTIUM BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN
AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
</p><p>
Except as contained in this notice, the name of the X Consortium shall not be
used in advertising or otherwise to promote the sale, use or other dealings
in this Software without prior written authorization from the X Consortium.
</p><p>X Window System is a trademark of The Open Group.</p></div></div></div><hr /></div><div class="toc"><p><strong>Table of Contents</strong></p><dl><dt><span class="sect1"><a href="#Overview">Overview</a></span></dt><dt><span class="sect1"><a href="#Values">Values</a></span></dt><dt><span class="sect1"><a href="#Control_Characters">Control Characters</a></span></dt><dt><span class="sect1"><a href="#Standard_Character_Set_Encodings">Standard Character Set Encodings</a></span></dt><dt><span class="sect1"><a href="#Approved_Standard_Encodings">Approved Standard Encodings</a></span></dt><dt><span class="sect1"><a href="#Non_Standard_Character_Set_Encodings">Non-Standard Character Set Encodings</a></span></dt><dt><span class="sect1"><a href="#Directionality">Directionality</a></span></dt><dt><span class="sect1"><a href="#Resources">Resources</a></span></dt><dt><span class="sect1"><a href="#Font_Names">Font Names</a></span></dt><dt><span class="sect1"><a href="#Extensions">Extensions</a></span></dt><dt><span class="sect1"><a href="#Errors">Errors</a></span></dt></dl></div><div class="sect1"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="Overview"></a>Overview</h2></div></div></div><p>
Compound Text is a format for multiple character set data, such as
multi-lingual text. The format is based on ISO
standards for encoding and combining character sets. Compound Text is intended
to be used in three main contexts: inter-client communication using selections,
as defined in the
<span class="emphasis"><em>Inter-Client Communication Conventions Manual</em></span>
(ICCCM);
window properties (e.g., window manager hints as defined in the ICCCM);
and resources (e.g., as defined in Xlib and the Xt Intrinsics).
</p><p>
Compound Text is intended as an external representation, or interchange format,
not as an internal representation. It is expected (but not required) that
clients will convert Compound Text to some internal representation for
processing and rendering, and convert from that internal representation to
Compound Text when providing textual data to another client.
</p></div><div class="sect1"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="Values"></a>Values</h2></div></div></div><p>
The name of this encoding is "COMPOUND_TEXT". When text values are used in
the ICCCM-compliant selection mechanism or are stored as window properties in
the server, the type used should be the atom for "COMPOUND_TEXT".
</p><p>
Octet values are represented in this document as two decimal numbers in the
form col/row. This means the value (col * 16) + row. For example, 02/01 means
the value 33.
</p><p>
For our purposes, the octet encoding space is divided into four ranges:
</p><div class="informaltable"><table border="0"><colgroup><col align="left" class="c1" /><col align="left" class="c2" /></colgroup><tbody><tr><td align="left">C0</td><td align="left">octets from 00/00 to 01/15</td></tr><tr><td align="left">GL</td><td align="left">octets from 02/00 to 07/15</td></tr><tr><td align="left">C1</td><td align="left">octets from 08/00 to 09/15</td></tr><tr><td align="left">GR</td><td align="left">octets from 10/00 to 15/15</td></tr></tbody></table></div><p>
C0 and C1 are "control character" sets, while GL and GR are "graphic
character" sets. Only a subset of C0 and C1 octets are used in the encoding,
and depending on the character set encoding defined as GL or GR, a subset of
GL and GR octets may be used; see below for details. All octets (00/00 to
15/15) may appear inside the text of extended segments (defined below).
</p><p>
[For those familiar with ISO 2022, we will use only an 8-bit environment, and
we will always use G0 for GL and G1 for GR.]
</p></div><div class="sect1"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="Control_Characters"></a>Control Characters</h2></div></div></div><p>
In C0, only the following values will be used:
</p><div class="informaltable"><table border="0"><colgroup><col align="left" class="c1" /><col align="left" class="c2" /><col align="left" class="c3" /></colgroup><tbody><tr><td align="left">00/09</td><td align="left">HT</td><td align="left">HORIZONTAL TABULATION</td></tr><tr><td align="left">00/10</td><td align="left">NL</td><td align="left">NEW LINE</td></tr><tr><td align="left">01/11</td><td align="left">ESC</td><td align="left">(ESCAPE)</td></tr></tbody></table></div><p>
In C1, only the following value will be used:
</p><div class="informaltable"><table border="0"><colgroup><col align="left" class="c1" /><col align="left" class="c2" /><col align="left" class="c3" /></colgroup><tbody><tr><td align="left">09/11</td><td align="left">CSI</td><td align="left">CONTROL SEQUENCE INTRODUCER</td></tr></tbody></table></div><p>
[The alternate 7-bit CSI encoding 01/11 05/11 is not used in Compound Text.]
</p><p>
No control sequences are defined in Compound Text for changing the C0 and C1
sets.
</p><p>
A horizontal tab can be represented with the octet 00/09. Specification of
tabulation width settings is not part of Compound Text and must be obtained
from context (in an unspecified manner).
</p><p>
[Inclusion of horizontal tab is for consistency with the STRING type currently
defined in the ICCCM.]
</p><p>
A newline (line separator/terminator) can be represented with the octet 00/10.
</p><p>
[Note that 00/10 is normally LINEFEED, but is being interpreted as NEWLINE.
This can be thought of as using the (deprecated) NEW LINE mode, E.1.3, in ISO
6429. Use of this value instead of 08/05 (NEL, NEXT LINE) is for consistency
with the STRING type currently defined in the ICCCM.]
</p><p>
The remaining C0 and C1 values (01/11 and 09/11) are only used in the control
sequences defined below.
</p></div><div class="sect1"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="Standard_Character_Set_Encodings"></a>Standard Character Set Encodings</h2></div></div></div><p>
The default GL and GR sets in Compound Text correspond to the left and right
halves of ISO 8859-1 (Latin 1). As such, any legal instance of a STRING type
(as defined in the ICCCM) is also a legal instance of type COMPOUND_TEXT.
</p><p>
[The implied initial state in ISO 2022 is defined with the sequence:
01/11 02/00 04/03 GO and G1 in an 8-bit environment only. Designation also invokes.
01/11 02/00 04/07 In an 8-bit environment, C1 represented as 8-bits.
01/11 02/00 04/09 Graphic character sets can be 94 or 96.
01/11 02/00 04/11 8-bit code is used.
01/11 02/08 04/02 Designate ASCII into G0.
01/11 02/13 04/01 Designate right-hand part of ISO Latin-1 into G1.
]
</p><p>
To define one of the approved standard character set encodings to be
the GL set, one of the following control sequences is used:
</p><div class="informaltable"><table border="0"><colgroup><col align="left" class="c1" /><col align="left" class="c2" /><col align="left" class="c3" /><col align="left" class="c4" /></colgroup><tbody><tr><td align="left">01/11</td><td align="left">02/08</td><td align="left">{I} F</td><td align="left">94 character set</td></tr><tr><td align="left">01/11</td><td align="left">02/04</td><td align="left">02/08{I} F</td><td align="left">94<sup>N</sup> character set</td></tr></tbody></table></div><p>
To define one of the approved standard character set encodings to be
the GR set, one of the following control sequences is used:
</p><div class="informaltable"><table border="0"><colgroup><col align="left" class="c1" /><col align="left" class="c2" /><col align="left" class="c3" /><col align="left" class="c4" /></colgroup><tbody><tr><td align="left">01/11</td><td align="left">02/09</td><td align="left">{I} F</td><td align="left">94 character set</td></tr><tr><td align="left">01/11</td><td align="left">02/13</td><td align="left">{I} F</td><td align="left">96 character set</td></tr><tr><td align="left">01/11</td><td align="left">02/04</td><td align="left">02/09 {I} F</td><td align="left">94<sup>N</sup> character set</td></tr></tbody></table></div><p>
The "F"in the control sequences above stands for "Final character", which
is always in the range 04/00 to 07/14. The "{I}" stands for zero or more
"intermediate characters", which are always in the range 02/00 to 02/15, with
the first intermediate character always in the range 02/01 to 02/03. The
registration authority has defined an "{I} F" sequence for each registered
character set encoding.
</p><p>
[Final characters for private encodings (in the range 03/00 to 03/15) are not
permitted here in Compound Text.]
</p><p>
For GL, octet 02/00 is always defined as SPACE, and octet 07/15 (normally
DELETE) is never used. For a 94-character set defined as GR, octets 10/00 and
15/15 are never used.
</p><p>
[This is consistent with ISO 2022.]
</p><p>
A 94<sup>N</sup> character set uses N octets (N > 1) for each character.
The value of N is derived from the column value for F:
</p><div class="informaltable"><table border="0"><colgroup><col align="left" class="c1" /><col align="left" class="c2" /></colgroup><tbody><tr><td align="left">column 04 or 05</td><td align="left">2 octets</td></tr><tr><td align="left">column 06</td><td align="left">3 octets</td></tr><tr><td align="left">column 07</td><td align="left">4 or more octets</td></tr></tbody></table></div><p>
In a 94<sup>N</sup> encoding, the octet values 02/00 and 07/15 (in GL) and
10/00 and 15/15 (in GR) are never used.
</p><p>
[The column definitions come from ISO 2022.]
</p><p>
Once a GL or GR set has been defined, all further octets in that range (except
within control sequences and extended segments) are interpreted with respect to
that character set encoding, until the GL or GR set is redefined. GL and GR
sets can be defined independently, they do not have to be defined in pairs.
</p><p>
Note that when actually using a character set encoding as the GR set, you must
force the most significant bit (08/00) of each octet to be a one, so that it
falls in the range 10/00 to 15/15.
</p><p>
[Control sequences to specify character set encoding revisions (as in section
6.3.13 of ISO 2022) are not used in Compound Text. Revision indicators do not
appear to provide useful information in the context of Compound Text. The most
recent revision can always be assumed, since revisions are upward compatible.]
</p></div><div class="sect1"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="Approved_Standard_Encodings"></a>Approved Standard Encodings</h2></div></div></div><p>
The following are the approved standard encodings to be used with Compound
Text. Note that none have Intermediate characters; however, a good parser will
still deal with Intermediate characters in the event that additional encodings
are later added to this list.
</p><div class="informaltable"><table border="1"><colgroup><col align="left" class="c1" /><col align="left" class="c2" /><col align="left" class="c3" /></colgroup><thead><tr><th align="left">{I} F</th><th align="left">94/96</th><th align="left">Description</th></tr></thead><tbody><tr><td align="left">4/02</td><td align="left">94</td><td align="left">
7-bit ASCII graphics (ANSI X3.4-1968), Left half of ISO 8859 sets
</td></tr><tr><td align="left">04/09</td><td align="left">94</td><td align="left">
Right half of JIS X0201-1976 (reaffirmed 1984),
8-Bit Alphanumeric-Katakana Code
</td></tr><tr><td align="left">04/10</td><td align="left">94</td><td align="left">
Left half of JIS X0201-1976 (reaffirmed 1984),
8-Bit Alphanumeric-Katakana Code
</td></tr><tr><td align="left">04/01</td><td align="left">96</td><td align="left">Right half of ISO 8859-1, Latin alphabet No. 1</td></tr><tr><td align="left">04/02</td><td align="left">96</td><td align="left">Right half of ISO 8859-2, Latin alphabet No. 2</td></tr><tr><td align="left">04/03</td><td align="left">96</td><td align="left">Right half of ISO 8859-3, Latin alphabet No. 3</td></tr><tr><td align="left">04/04</td><td align="left">96</td><td align="left">Right half of ISO 8859-4, Latin alphabet No. 4</td></tr><tr><td align="left">04/06</td><td align="left">96</td><td align="left">Right half of ISO 8859-7, Latin/Greek alphabet</td></tr><tr><td align="left">04/07</td><td align="left">96</td><td align="left">Right half of ISO 8859-6, Latin/Arabic alphabet</td></tr><tr><td align="left">04/08</td><td align="left">96</td><td align="left">Right half of ISO 8859-8, Latin/Hebrew alphabet</td></tr><tr><td align="left">04/12</td><td align="left">96</td><td align="left">Right half of ISO 8859-5, Latin/Cyrillic alphabet</td></tr><tr><td align="left">04/13</td><td align="left">96</td><td align="left">Right half of ISO 8859-9, Latin alphabet No. 5</td></tr><tr><td align="left">04/01</td><td align="left">942</td><td align="left">GB2312-1980, China (PRC) Hanzi</td></tr><tr><td align="left">04/02</td><td align="left">942</td><td align="left">JIS X0208-1983, Japanese Graphic Character Set</td></tr><tr><td align="left">04/03</td><td align="left">942</td><td align="left">KS C5601-1987, Korean Graphic Character Set</td></tr></tbody></table></div><p>
The sets listed as "Left half of ..." should always be defined as GL. The
sets listed as "Right half of ..." should always be defined as GR. Other
sets can be defined either as GL or GR.
</p></div><div class="sect1"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="Non_Standard_Character_Set_Encodings"></a>Non-Standard Character Set Encodings</h2></div></div></div><p>
Character set encodings that are not in the list of approved standard
encodings can be included
using "extended segments". An extended segment begins with one of the
following sequences:
</p><div class="informaltable"><table border="0"><colgroup><col align="left" class="c1" /><col align="left" class="c2" /></colgroup><tbody><tr><td align="left">01/11 2/05 02/15 03/00 M L</td><td align="left">variable number of octets per character</td></tr><tr><td align="left">01/11 2/05 02/15 03/01 M L</td><td align="left">1 octet per character</td></tr><tr><td align="left">01/11 2/05 02/15 03/02 M L</td><td align="left">2 octet per character</td></tr><tr><td align="left">01/11 2/05 02/15 03/03 M L</td><td align="left">3 octet per character</td></tr><tr><td align="left">01/11 2/05 02/15 03/04 M L</td><td align="left">4 octet per character</td></tr></tbody></table></div><p>
[This uses the "other coding system" of ISO 2022, using private Final
characters.]
</p><p>
The "M" and "L" octets represent a 14-bit unsigned value giving the number
of octets that appear in the remainder of the segment. The number is computed
as ((M - 128) * 128) + (L - 128). The most significant bit M and L are always
set to one. The remainder of the segment consists of two parts, the name of
the character set encoding and the actual text. The name of the encoding comes
first and is separated from the text by the octet 00/02 (STX, START OF TEXT).
Note that the length defined by M and L includes the encoding name and
separator.
</p><p>
[The encoding of the length is chosen to avoid having zero octets in Compound
Text when possible, because embedded NUL values are problematic in many C
language routines. The use of zero octets cannot be ruled out entirely
however, since some octets in the actual text of the extended segment may have
to be zero.]
</p><p>
The name of the encoding should be registered with the X Consortium to avoid
conflicts and should when appropriate match the CharSet Registry and Encoding
registration used in the X Logical Font Description. The name itself should be
encoded using ISO 8859-1 (Latin 1), should not use question mark (03/15) or
asterisk (02/10), and should use hyphen (02/13) only in accordance with the X
Logical Font Description.
</p><p>
Extended segments are not to be used for any character set encoding that can
be constructed from a GL/GR pair of approved standard encodings. For
example, it is incorrect to use an extended segment for any of the ISO 8859
family of encodings.
</p><p>
It should be noted that the contents of an extended segment are arbitrary;
for example,
they may contain octets in the C0 and C1 ranges, including 00/00, and
octets comprising a given character may differ in their most significant bit.
</p><p>
[ISO-registered "other coding systems" are not used in Compound Text;
extended segments are the only mechanism for non-2022 encodings.]
</p></div><div class="sect1"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="Directionality"></a>Directionality</h2></div></div></div><p>
If desired, horizontal text direction can be indicated using the following
control sequences:
</p><div class="informaltable"><table border="0"><colgroup><col align="left" class="c1" /><col align="left" class="c2" /></colgroup><tbody><tr><td align="left">09/11 03/01 05/13</td><td align="left">begin left-to-right text</td></tr><tr><td align="left">09/11 03/02 05/13</td><td align="left">begin right-to-left text</td></tr><tr><td align="left">09/11 05/13</td><td align="left">end of string</td></tr></tbody></table></div><p>
[This is a subset of the SDS (START DIRECTED STRING) control in the Draft
Bidirectional Addendum to ISO 6429.]
</p><p>
Directionality can be nested. Logically, a stack of directions is maintained.
Each of the first two control sequences pushes a new direction on the stack,
and the third sequence (revert) pops a direction from the stack. The stack
starts out empty at the beginning of a Compound Text string. When the stack is
empty, the directionality of the text is unspecified.
</p><p>
Directionality applies to all subsequent text, whether in GL, GR, or an
extended segment. If the desired directionality of GL, GR, or extended
segments differs, then directionality control sequences must be inserted when
switching between them.
</p><p>
Note that definition of GL and GR sets is independent of directionality;
defining a new GL or GR set does not change the current directionality, and
pushing or popping a directionality does not change the current GL and GR
definitions.
</p><p>
Specification of directionality is entirely optional; text direction should be
clear from context in most cases. However, it must be the case that either
all characters in a Compound Text string have explicitly specified direction
or that all characters have unspecified direction. That is, if directionality
control sequences are used, the first such control sequence must precede the
first graphic character in a Compound Text string, and graphic characters are
not permitted whenever the directionality stack is empty.
</p></div><div class="sect1"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="Resources"></a>Resources</h2></div></div></div><p>
To use Compound Text in a resource, you can simply treat all octets as if they
were ASCII/Latin-1 and just replace all "\" octets (05/12) with the two
octets "\\", all newline octets (00/10) with the two octets "\n", and
all zero octets with the four octets "\000".
It is up to the client making use of the resource to interpret the data as
Compound Text; the policy by which this is ascertained is not constrained by
the Compound Text specification.
</p></div><div class="sect1"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="Font_Names"></a>Font Names</h2></div></div></div><p>
The following CharSet names for the standard character set encodings are
registered for use in font names under the X Logical Font Description:
</p><div class="informaltable"><table border="1"><colgroup><col align="left" class="c1" /><col align="left" class="c2" /><col align="left" class="c3" /></colgroup><thead><tr><th align="left">Name</th><th align="left">Encoding Standard</th><th align="left">Description</th></tr></thead><tbody><tr><td align="left">ISO8859-1</td><td align="left">ISO8859-1</td><td align="left">Latinalphabet No. 1</td></tr><tr><td align="left">ISO8859-2</td><td align="left">ISO8859-2</td><td align="left">Latinalphabet No. 2</td></tr><tr><td align="left">ISO8859-3</td><td align="left">ISO8859-3</td><td align="left">Latinalphabet No. 3</td></tr><tr><td align="left">ISO8859-4</td><td align="left">ISO8859-4</td><td align="left">Latinalphabet No. 4</td></tr><tr><td align="left">ISO8859-5</td><td align="left">ISO 8859-5</td><td align="left">Latin/Cyrillic alphabet</td></tr><tr><td align="left">ISO8859-6</td><td align="left">ISO 8859-6</td><td align="left">Latin/Arabic alphabet</td></tr><tr><td align="left">ISO8859-7</td><td align="left">ISO8859-7</td><td align="left">Latin/Greekalphabet</td></tr><tr><td align="left">ISO8859-8</td><td align="left">ISO8859-8</td><td align="left">Latin/Hebrew alphabet</td></tr><tr><td align="left">ISO8859-9</td><td align="left">ISO8859-9</td><td align="left">Latinalphabet No. 5</td></tr><tr><td align="left">JISX0201.1976-0</td><td align="left">JIS X0201-1976 (reaffirmed 1984)</td><td align="left">8-bit Alphanumeric-Katakana Code</td></tr><tr><td align="left">GB2312.1980-0</td><td align="left">GB2312-1980, GL encoding</td><td align="left">China (PRC) Hanzi</td></tr><tr><td align="left">JISX0208.1983-0</td><td align="left">JIS X0208-1983, GL encoding</td><td align="left">Japanese Graphic Character Set</td></tr><tr><td align="left">KSC5601.1987-0</td><td align="left">KS C5601-1987, GL encoding</td><td align="left">Korean Graphic Character Set</td></tr></tbody></table></div></div><div class="sect1"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="Extensions"></a>Extensions</h2></div></div></div><p>
There is no absolute requirement for a parser to deal with anything but the
particular encoding syntax defined in this specification. However, it is
possible that Compound Text may be extended in the future, and as such it may
be desirable to construct the parser to handle 2022/6429 syntax more generally.
</p><p>
There are two general formats covering all control sequences that are expected
to appear in extensions:
</p><p>
01/11 {I} F
</p><p>
For this format, I is always in the range 02/00 to 02/15, and F is always
in the range 03/00 to 07/14.
</p><p>
09/11 {P} {I} F
</p><p>
For this format, P is always in the range 03/00 to 03/15, I is always in
the range 02/00 to 02/15, and F is always in the range 04/00 to 07/14.
</p><p>
In addition, new (singleton) control characters (in the C0 and C1 ranges) might
be defined in the future.
</p><p>
Finally, new kinds of "segments" might be defined in the future using syntax
similar to extended segments:
</p><p>
01/11 02/05 02/15 F M L
</p><p>
For this format, F is in the range 03/05 to 3/15. M and L are as defined
in extended segments. Such a segment will always be followed by the number
of octets defined by M and L. These octets can have arbitrary values and
need not follow the internal structure defined for current extended
segments.
</p><p>
If extensions to this specification are defined in the future, then any string
incorporating instances of such extensions must start with one of the following
control sequences:
</p><div class="informaltable"><table border="0"><colgroup><col align="left" class="c1" /><col align="left" class="c2" /></colgroup><tbody><tr><td align="left">01/11 02/03 V 03/00</td><td align="left">ignoring extensions is OK</td></tr><tr><td align="left">01/11 02/03 V 03/01</td><td align="left">ignoring extensions is not OK</td></tr></tbody></table></div><p>
In either case, V is in the range 02/00 to 02/15 and indicates the major
version
minus one of the specification being used. These version control sequences are
for use by clients that implement earlier versions, but have implemented a
general parser. The first control sequence indicates that it is acceptable to
ignore all extension control sequences; no mandatory information will be lost
in the process. The second control sequence indicates that it is unacceptable
to ignore any extension control sequences; mandatory information would be lost
in the process. In general, it will be up to the client generating the
Compound Text to decide which control sequence to use.
</p></div><div class="sect1"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="Errors"></a>Errors</h2></div></div></div><p>
If a Compound Text string does not match the specification here (e.g., uses
undefined control characters, or undefined control sequences, or incorrectly
formatted extended segments), it is best to treat the entire string as invalid,
except as indicated by a version control sequence.
</p></div></div></body></html>