#! /usr/bin/perl
-# $Cambridge: exim/doc/doc-docbook/Pre-xml,v 1.2 2005/11/10 12:30:13 ph10 Exp $
+# $Cambridge: exim/doc/doc-docbook/Pre-xml,v 1.3 2006/02/01 11:01:01 ph10 Exp $
# Script to pre-process XML input before processing it for various purposes.
# Options specify which transformations are to be done. Monospaced literal
# Changes:
-# -abstract: Remove the <abstract> element
-
-# -ascii: Replace &8230; (sic, no x) with ...
-# Replace ’ by '
-# Replace “ by "
-# Replace ” by "
-# Replace – by -
-# Replace † by *
-# Replace ‡ by **
-# Replace   by a space
-# Replace © by (c)
-# Put quotes round <literal> text
+# -ascii: Replace ’ by '
+# Replace © by (c)
+# Replace † by *
+# Replace ‡ by **
+# Replace by a space
+# Replace – by -
# Put quotes round <quote> text
#
+# -quoteliteral:
+# Put quotes round <literal> text
+#
# -bookinfo: Remove the <bookinfo> element from the file
#
# -fi: Replace "fi" by fi except when it is in an XML element, or
#
# -html: Certain things are done only for HTML output:
#
-# If <literallayout> is followed by optional # space and then a
+# If <literallayout> is followed by optional space and then a
# newline, the space and newline are removed, because otherwise you
# get a blank line in the HTML output.
#
# -noindex Remove the XML to generate a Concept and an Options index.
# -oneindex Ditto, but add XML to generate a single index.
+#
+# -optbreak Insert an optional line break (zero-width space, ​) after
+# every underscore in text within <option> and <variable> elements,
+# except when preceded by <entry> (i.e. not in tables). The same is
+# also done within a word of four or more upper-case letters (for
+# compile-time options).
-# The function that processes non-literal monospaced text
+# The function that processes non-literal, non-monospaced text
sub process()
{
$s =~ s/fi(?![^<>]*>)/fi/g if $ligatures;
+if ($optbreak)
+ {
+ $s =~ s%(?<!<entry>)(<option>|<varname>)([^<]+)%
+ my($x,$y) = ($1,$2); $y =~ s/_/_​/g; "$x"."$y"%gex;
+
+ $s =~ s?\b([A-Z_]{4,})\b?
+ my($x) = $1; $x =~ s/_/_​/g; "$x"?gex;
+ }
+
if ($ascii)
{
- $s =~ s/…/.../g;
$s =~ s/’/'/g;
- $s =~ s/“/"/g;
- $s =~ s/”/"/g;
- $s =~ s/–/-/g;
- $s =~ s/†/*/g;
- $s =~ s/‡/**/g;
- $s =~ s/ / /g;
- $s =~ s/©/(c)/g;
+ $s =~ s/©/(c)/g;
+ $s =~ s/†/*/g;
+ $s =~ s/‡/**/g;
+ $s =~ s/&nsbp;/ /g;
+ $s =~ s/–/-/g;
$s =~ s/<quote>/"/g;
$s =~ s/<\/quote>/"/g;
}
# The main program
-$abstract = 0;
$ascii = 0;
$bookinfo = 0;
$html = 0;
$madeindex = 0;
$noindex = 0;
$oneindex = 0;
+$optbreak = 0;
+$quoteliteral = 0;
foreach $arg (@ARGV)
{
if ($arg eq "-fi") { $ligatures = 1; }
- elsif ($arg eq "-abstract") { $abstract = 1; }
elsif ($arg eq "-ascii") { $ascii = 1; }
elsif ($arg eq "-bookinfo") { $bookinfo = 1; }
elsif ($arg eq "-html") { $html = 1; }
elsif ($arg eq "-noindex") { $noindex = 1; }
elsif ($arg eq "-oneindex") { $oneindex = 1; }
+ elsif ($arg eq "-optbreak") { $optbreak = 1; }
+ elsif ($arg eq "-quoteliteral") { $quoteliteral = 1; }
else { die "** Pre-xml: Unknown option \"$arg\"\n"; }
}
while (<STDIN>)
{
- # Remove <abstract> if required
-
- next if ($abstract && /^\s*<abstract>/);
-
# Remove <bookinfo> if required
if ($bookinfo && /^<bookinfo/)
if (/^(.*?)<\/literal>(?!<\/quote>)(.*)$/)
{
print $1;
- print "\"" if $ascii && !$inliterallayout;
+ print "\"" if $quoteliteral && !$inliterallayout;
print "</literal>";
$inliteral = 0;
$_ = "$2\n";
{
print &process($1);
print "<literal>";
- print "\"" if $ascii && !$inliterallayout;
+ print "\"" if $quoteliteral && !$inliterallayout;
$inliteral = 1;
$_ = "$2\n";
}