<!-- * *************************************************************** -->
<!-- * Commentary -->
<!-- * *************************************************************** -->
-<!-- * This file maps a small subset of Unicode characters (around 800 -->
-<!-- * or so) to corresponding roff escape sequences. Use it when you -->
-<!-- * need to convert Unicode (UTF-8 or UTF-16) encoded XML content -->
-<!-- * to roff (e.g., to generate "portable" man-page output). -->
<!-- * -->
-<!-- * Although the format follows the "character map" format -->
-<!-- * specified in the XSLT 2.0 working draft[1], it can be used with -->
-<!-- * an appropriate XSLT 1.0 stylesheet and any XSLT 1.0 processor. -->
+<!-- * This file maps a selected subset of Unicode symbols and special -->
+<!-- * characters (around 800 or so) to corresponding groff escape -->
+<!-- * sequences. Use it when you need to convert Unicode (UTF-8 or -->
+<!-- * UTF-16) encoded XML content to groff output for an environment -->
+<!-- * that uses a good, modern groff instead of nroff (for example, a -->
+<!-- * GNU/Linux system, FreeBSD system, or Cygwin environment). -->
+<!-- * -->
+<!-- * If, on the other hand, you aren't sure what OSes or -->
+<!-- * environments your man-page output might end up being viewed on, -->
+<!-- * and not sure what version of nroff/groff those environments -->
+<!-- * might have, you should probably instead use the accompanying -->
+<!-- * charmap.roff.min.xml file; it has a much smaller and "safer" -->
+<!-- * set of mappings (only about 40 characters). -->
+<!-- * -->
+<!-- * Although the format of this file follows the "character map" -->
+<!-- * format specified in the XSLT 2.0 working draft[1], the file can -->
+<!-- * also be used with an appropriate XSLT 1.0 stylesheet and any -->
+<!-- * XSLT 1.0 processor. -->
<!-- * -->
<!-- * [1] http://www.w3.org/TR/xslt20/#character-maps -->
<!-- * -->
unichar:name="DIVISION SLASH"
string="@esc@(f/"
/>
- <!-- * not in roff -->
- <!-- * <xsl:output-character -->
- <!-- * char="∖" -->
- <!-- * unichar:name="SET MINUS" -->
- <!-- * unichar:entity="setmn" -->
- <!-- * /> -->
+ <xsl:output-character
+ char="∖"
+ unichar:name="SET MINUS"
+ unichar:entity="setmn"
+ string="@esc@@esc@"
+ />
<xsl:output-character
char="∗"
unichar:name="ASTERISK OPERATOR"
--- /dev/null
+<?xml version="1.0" encoding="US-ASCII"?>
+<xsl:character-map xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
+ xmlns:unichar="http://docbook.sourceforge.net/xmlns/unichar/1.0">
+
+<!-- ********************************************************************
+ $Id$
+ ********************************************************************
+
+ This file is part of the XSL DocBook Stylesheet distribution.
+ See ../README or http://docbook.sf.net/release/xsl/current/ for
+ copyright and other information.
+
+ ******************************************************************** -->
+
+<!-- * *************************************************************** -->
+<!-- * Commentary -->
+<!-- * *************************************************************** -->
+<!-- * This file maps a minimal subset of Unicode characters to -->
+<!-- * corresponding roff escape sequences. The subset is limited to -->
+<!-- * less than 40 characters - basically, just a few common symbols, -->
+<!-- * special spaces and dashes, and "curly" quotation marks. -->
+<!-- * -->
+<!-- * The mappings are basically intended just for generating man -->
+<!-- * pages that will look acceptable on a TTY, without any concern -->
+<!-- * for how the same man pages will look if output to a different -->
+<!-- * device (to a postscript printer, for example). -->
+<!-- * -->
+<!-- * Use this file when you aren't sure what OSes or environments -->
+<!-- * your man-page output might end up being viewed on, and not sure -->
+<!-- * what version of nroff/groff those environments might have. -->
+<!-- * -->
+<!-- * If you're instead generating output for an environment that -->
+<!-- * uses a good, modern groff instead of nroff (for example, a -->
+<!-- * GNU/Linux system, FreeBSD system, or Cygwin environment), see -->
+<!-- * the accompanying charmap.groff.xml file; it has a much richer -->
+<!-- * set of mappings (it maps more than 800 characters). -->
+<!-- * -->
+<!-- * Although the format of this file follows the "character map" -->
+<!-- * format specified in the XSLT 2.0 working draft[1], the file can -->
+<!-- * also be used with an appropriate XSLT 1.0 stylesheet and any -->
+<!-- * XSLT 1.0 processor. -->
+<!-- * -->
+<!-- * [1] http://www.w3.org/TR/xslt20/#character-maps -->
+<!-- * -->
+<!-- * For more details, see the accompanying charmap.groff.xml file. -->
+<!-- * -->
+<!-- * Note: In place of the literal backlash "\" character used in roff -->
+<!-- * to indicate the start of an escape sequence, this file uses -->
+<!-- * "@esc@". So an additional string-substition step needs to be done -->
+<!-- * on any output generated using the data from this file, before -->
+<!-- * final roff output, to replace the "@esc@" instances with the -->
+<!-- * literal backslash characters that roff expects. -->
+
+<!-- * ################################################################# -->
+
+ <!-- * ***************************************************************** -->
+ <!-- * Begin: Latin-1/ISO-8859-1 -->
+ <!-- * x00a0 to x00ff -->
+ <!-- * ***************************************************************** -->
+
+ <xsl:output-character
+ char=" "
+ unichar:name="NO-BREAK SPACE"
+ unichar:entity="nbsp"
+ string="@esc@ "
+ />
+ <xsl:output-character
+ char="©"
+ unichar:name="COPYRIGHT SIGN"
+ unichar:entity="copy"
+ string="@esc@(co"
+ />
+ <xsl:output-character
+ char="®"
+ unichar:name="REGISTERED SIGN"
+ unichar:entity="reg"
+ string="@esc@(rg"
+ />
+ <xsl:output-character
+ char="×"
+ unichar:name="MULTIPLICATION SIGN"
+ unichar:entity="times"
+ string="@esc@(mu"
+ />
+ <xsl:output-character
+ char="÷"
+ unichar:name="DIVISION SIGN"
+ unichar:entity="divide"
+ string="@esc@(di"
+ />
+
+ <!-- * ***************************************************************** -->
+ <!-- * Begin: General Punctuation -->
+ <!-- * x2000 to x206f -->
+ <!-- * **************************************************************** -->
+
+ <!-- * first, spaces of various widths -->
+
+ <!-- * Note: There does not seem to be either a real em space or en space -->
+ <!-- * in roff; to approximate them, this character map assumes that in -->
+ <!-- * most fonts, an en space is about the same as the width of a digit -->
+ <!-- * (in roff, "\0"), so an em space (which by definition is -->
+ <!-- * equal to the width of two en spaces) is about the same as the width -->
+ <!-- * of two digits (thus, in roff, "\0\0") -->
+ <!-- * first, Spaces of various widths -->
+
+ <xsl:output-character
+ char=" "
+ unichar:name="EN QUAD"
+ string="@esc@0"
+ />
+ <xsl:output-character
+ char=" "
+ unichar:name="EM QUAD"
+ string="@esc@0@esc@0"
+ />
+ <xsl:output-character
+ char=" "
+ unichar:name="EN SPACE"
+ unichar:entity="ensp"
+ string="@esc@0"
+ />
+ <xsl:output-character
+ char=" "
+ unichar:name="EM SPACE"
+ unichar:entity="emsp"
+ string="@esc@0@esc@0"
+ />
+ <!-- * same as roff "digit" space -->
+ <xsl:output-character
+ char=" "
+ unichar:name="FIGURE SPACE"
+ unichar:entity="numsp"
+ string="@esc@0"
+ />
+ <!-- * punctuation space in most fonts is actually closer to a normal -->
+ <!-- * space than it is to a thin space -->
+ <xsl:output-character
+ char=" "
+ unichar:name="PUNCTUATION SPACE"
+ unichar:entity="puncsp"
+ string=" "
+ />
+ <!-- * Note: Not sure how best to deal with thin space, because the roff -->
+ <!-- * thin space, "\^", prints as a zero-width space in TTY -->
+ <!-- * output. However, it seems that, unlike a hair space, a thin space, -->
+ <!-- * at 1/12 of an em, is still recognizable to most people as a space, -->
+ <!-- * so treating it as zero-width seems wrong. So, for the sake of making -->
+ <!-- * TTY output look OK, just substitute with a normal space; but real -->
+ <!-- * roff escape is "\(\^" -->
+ <xsl:output-character
+ char=" "
+ unichar:name="THIN SPACE"
+ unichar:entity="thinsp"
+ string=" "
+ />
+ <!-- * I don't think there's a standard definition of what a hair -->
+ <!-- * space is; some guides just say it's "less than 1/5 of an em" or -->
+ <!-- * that it's "narrower than a thin space"; seems like in practice, -->
+ <!-- * it's *a lot* narrower than a thin space, to the point where -->
+ <!-- * it's close to being a non-space, so here it's substituted with -->
+ <!-- * roff equivalent of a zero-width no-break space -->
+ <xsl:output-character
+ char=" "
+ unichar:name="HAIR SPACE"
+ unichar:entity="hairsp"
+ string="@esc@&"
+ />
+ <!-- * map to roff "zero-width break point" -->
+ <xsl:output-character
+ char="​"
+ unichar:name="ZERO WIDTH SPACE"
+ string="@esc@:"
+ />
+ <!-- * .................................................... -->
+ <!-- * next, hyphens and various dashes, bars, underscores -->
+ <xsl:output-character
+ char="‐"
+ unichar:name="HYPHEN"
+ unichar:entity="hyphen"
+ string="@esc@(hy"
+ />
+ <xsl:output-character
+ char="‑"
+ unichar:name="NON-BREAKING HYPHEN"
+ string="@esc@&-@esc@&"
+ />
+ <!-- * roughly same width as en dash -->
+ <xsl:output-character
+ char="‒"
+ unichar:name="FIGURE DASH"
+ string="@esc@(en"
+ />
+ <xsl:output-character
+ char="–"
+ unichar:name="EN DASH"
+ unichar:entity="ndash"
+ string="@esc@(en"
+ />
+ <xsl:output-character
+ char="—"
+ unichar:name="EM DASH"
+ unichar:entity="mdash"
+ string="@esc@(em"
+ />
+ <!-- * seems roughly same width as em dash -->
+ <xsl:output-character
+ char="―"
+ unichar:name="HORIZONTAL BAR"
+ unichar:entity="horbar"
+ string="@esc@(em"
+ />
+
+ <!-- * .................................................... -->
+ <!-- * various quotation marks -->
+ <xsl:output-character
+ char="‘"
+ unichar:name="LEFT SINGLE QUOTATION MARK"
+ unichar:entity="lsquo"
+ string="@esc@(oq"
+ />
+ <xsl:output-character
+ char="’"
+ unichar:name="RIGHT SINGLE QUOTATION MARK"
+ unichar:entity="rsquo"
+ string="@esc@(cq"
+ />
+
+ <xsl:output-character
+ char="“"
+ unichar:name="LEFT DOUBLE QUOTATION MARK"
+ unichar:entity="ldquo"
+ string="@esc@(lq"
+ />
+ <xsl:output-character
+ char="”"
+ unichar:name="RIGHT DOUBLE QUOTATION MARK"
+ unichar:entity="rdquo"
+ string="@esc@(rq"
+ />
+
+ <!-- * .................................................... -->
+ <!-- * various symbols -->
+ <xsl:output-character
+ char="†"
+ unichar:name="DAGGER"
+ unichar:entity="dagger"
+ string="@esc@(dg"
+ />
+ <xsl:output-character
+ char="‡"
+ unichar:name="DOUBLE DAGGER"
+ unichar:entity="Dagger"
+ string="@esc@(dd"
+ />
+ <xsl:output-character
+ char="•"
+ unichar:name="BULLET"
+ unichar:entity="bull"
+ string="@esc@(bu"
+ />
+ <xsl:output-character
+ char="…"
+ unichar:name="HORIZONTAL ELLIPSIS"
+ unichar:entity="hellip"
+ string="@esc@&..."
+ />
+ <xsl:output-character
+ char="‧"
+ unichar:name="HYPHENATION POINT"
+ string="@esc@%"
+ />
+ <!-- * seems like "narrow" nbsp is basically the same as a no-break -->
+ <!-- * space -->
+ <xsl:output-character
+ char=" "
+ unichar:name="NARROW NO-BREAK SPACE"
+ string="@esc@ "
+ />
+ <xsl:output-character
+ char="′"
+ unichar:name="PRIME"
+ unichar:entity="prime"
+ string="@esc@(fm"
+ />
+ <xsl:output-character
+ char="″"
+ unichar:name="DOUBLE PRIME"
+ unichar:entity="Prime"
+ string="@esc@(sd"
+ />
+ <!-- * Regarding x2060 vs. xFEFF, the document "Unicode Standard Annex #14, -->
+ <!-- * Line Breaking Properties"[1] says: -->
+ <!-- * -->
+ <!-- * The word joiner character [x2060 a.k.a "WJ"] is the preferred -->
+ <!-- * choice for an invisible character to keep other characters -->
+ <!-- * together that would otherwise be split across the line at a direct -->
+ <!-- * break. The character FEFF has the same effect, but because it is -->
+ <!-- * also used in an unrelated way as a byte order mark, the use of the -->
+ <!-- * WJ as the preferred interword glue simplifies the handling of FEFF. -->
+ <!-- * -->
+ <!-- * [1] http://www.unicode.org/reports/tr14/ -->
+ <!-- * -->
+ <!-- * The groff docs seem ambiguous about whether \& is a joiner and -->
+ <!-- * prevents breaks, but, based on testing, seems like it does -->
+ <xsl:output-character
+ char="⁠"
+ unichar:name="WORD JOINER"
+ string="@esc@&"
+ />
+ <xsl:output-character
+ char="℠"
+ unichar:name="SERVICE MARK"
+ string="sm"
+ />
+ <xsl:output-character
+ char="™"
+ unichar:name="TRADE MARK SIGN"
+ unichar:entity="trade"
+ string="@esc@(tm"
+ />
+ <xsl:output-character
+ char="∖"
+ unichar:name="SET MINUS"
+ unichar:entity="setmn"
+ string="@esc@@esc@"
+ />
+ <xsl:output-character
+ char=""
+ unichar:name="ZERO WIDTH NO-BREAK SPACE"
+ string="@esc@&"
+ />
+</xsl:character-map>
projects / docbook-dsssl / commitdiff