Kate syntax file generator

There have been things added to LSL since I released my syntax file, so I decided to solve that problem once and for all. This is quick and dirty, so feel free to comment and criticize. I tested it a bit, and it seems to work.

Usage:

./gensyntax.pl < lsl-pattern.xml > lsl.xml

Or:

./gensyntax.pl "$HOME/sl_path" < lsl-pattern.xml > lsl.xml

Now, the required files, which should all be executable and in the same directory:

lsl-pattern.xml:

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE language SYSTEM "language.dtd">
<language name="LSL" version="1.1" kateversion="2.4" section="Sources" extensions="*.lsl" mimetype="" priority="5" license="GPL" author="Dale Glass">
	<highlighting>
		<list name="keywords">
			<item> default </item>
			<item> state </item>
			<!--INI_EXTRACT: flow control keywords-->
		</list>
		<list name="types">
			<!--INI_EXTRACT: data types-->
		</list>
		<list name="functions">
			<!--GUIDE_EXTRACT-->
		</list>
		<list name="constants">
			<!--INI_EXTRACT: integer constants-->
			<!--INI_EXTRACT: string constants-->
			<!--INI_EXTRACT: float constants-->
			<!--INI_EXTRACT: compound constants-->
		</list>
		<list name="events">
			<!--INI_EXTRACT: events-->
		</list>
		<contexts>
			<context attribute="Normal Text" lineEndContext="#stay" name="Normal">
				<DetectSpaces />
				<RegExpr attribute="Preprocessor" context="Outscoped" String="#\s*if\s+0" beginRegion="Outscoped" firstNonSpace="true" />
				<DetectChar attribute="Preprocessor" context="Preprocessor" char="#" firstNonSpace="true" />
				<StringDetect attribute="Region Marker" context="Region Marker" String="//BEGIN" beginRegion="Region1" firstNonSpace="true" />
				<StringDetect attribute="Region Marker" context="Region Marker" String="//END" endRegion="Region1" firstNonSpace="true" />
				<keyword attribute="Keyword" context="#stay" String="keywords"/>
				<keyword attribute="Function" context="#stay" String="functions"/>
				<keyword attribute="Constant" context="#stay" String="constants"/>
				<keyword attribute="Data Type" context="#stay" String="types"/>
				<keyword attribute="Event" context="#stay" String="events"/>
				<DetectIdentifier />
				<DetectChar attribute="Symbol" context="#stay" char="{" beginRegion="Brace1" />
				<DetectChar attribute="Symbol" context="#stay" char="}" endRegion="Brace1" />
				<Float attribute="Float" context="#stay">
					<AnyChar String="fF" attribute="Float" context="#stay"/>
				</Float>
				<HlCOct attribute="Octal" context="#stay"/>
				<HlCHex attribute="Hex" context="#stay"/>
				<Int attribute="Decimal" context="#stay" >
					<StringDetect attribute="Decimal" context="#stay" String="ULL" insensitive="TRUE"/>
					<StringDetect attribute="Decimal" context="#stay" String="LUL" insensitive="TRUE"/>
					<StringDetect attribute="Decimal" context="#stay" String="LLU" insensitive="TRUE"/>
					<StringDetect attribute="Decimal" context="#stay" String="UL" insensitive="TRUE"/>
					<StringDetect attribute="Decimal" context="#stay" String="LU" insensitive="TRUE"/>
					<StringDetect attribute="Decimal" context="#stay" String="LL" insensitive="TRUE"/>
					<StringDetect attribute="Decimal" context="#stay" String="U" insensitive="TRUE"/>
					<StringDetect attribute="Decimal" context="#stay" String="L" insensitive="TRUE"/>
				</Int>
				<HlCChar attribute="Char" context="#stay"/>
				<DetectChar attribute="String" context="String" char="""/>
				<IncludeRules context="##Doxygen" />
				<Detect2Chars attribute="Comment" context="Commentar 1" char="/" char1="/"/>
				<Detect2Chars attribute="Comment" context="Commentar 2" char="/" char1="*" beginRegion="Comment"/>
				<AnyChar attribute="Symbol" context="#stay" String=":!%&()+,-/.*<=>?[]|~^;"/>
			</context>
			<context attribute="String" lineEndContext="#pop" name="String">
				<LineContinue attribute="String" context="#stay"/>
				<HlCStringChar attribute="String Char" context="#stay"/>
				<DetectChar attribute="String" context="#pop" char="""/>
			</context>
			<context attribute="Region Marker" lineEndContext="#pop" name="Region Marker">
			</context>
			<context attribute="Comment" lineEndContext="#pop" name="Commentar 1">
				<IncludeRules context="##Alerts" />
			</context>
			<context attribute="Comment" lineEndContext="#stay" name="Commentar 2">
				<Detect2Chars attribute="Comment" context="#pop" char="*" char1="/" endRegion="Comment"/>
				<IncludeRules context="##Alerts" />
			</context>
			<context attribute="Preprocessor" lineEndContext="#pop" name="Preprocessor">
				<LineContinue attribute="Preprocessor" context="#stay"/>
				<RegExpr attribute="Preprocessor" context="Define" String="define.*((?=\\))"/>
				<RegExpr attribute="Preprocessor" context="#stay" String="define.*"/>
				<RangeDetect attribute="Prep. Lib" context="#stay" char=""" char1="""/>
				<RangeDetect attribute="Prep. Lib" context="#stay" char="<" char1=">"/>
				<IncludeRules context="##Doxygen" />
				<Detect2Chars attribute="Comment" context="Commentar/Preprocessor" char="/" char1="*" beginRegion="Comment2"/>
			</context>
			<context attribute="Preprocessor" lineEndContext="#pop" name="Define">
				<LineContinue attribute="Preprocessor" context="#stay"/>
			</context>
			<context attribute="Comment" lineEndContext="#stay" name="Commentar/Preprocessor">
				<Detect2Chars attribute="Comment" context="#pop" char="*" char1="/" endRegion="Comment2" />
			</context>
			<context attribute="Comment" lineEndContext="#stay" name="Outscoped" >
				<DetectSpaces />
				<IncludeRules context="##Alerts" />
				<DetectIdentifier />
				<DetectChar attribute="String" context="String" char="""/>
				<IncludeRules context="##Doxygen" />
				<Detect2Chars attribute="Comment" context="Commentar 1" char="/" char1="/"/>
				<Detect2Chars attribute="Comment" context="Commentar 2" char="/" char1="*" beginRegion="Comment"/>
				<RegExpr attribute="Comment" context="Outscoped intern" String="#\s*if" beginRegion="Outscoped" firstNonSpace="true" />
				<RegExpr attribute="Preprocessor" context="#pop" String="#\s*(endif|else|elif)" endRegion="Outscoped" firstNonSpace="true" />
			</context>
			<context attribute="Comment" lineEndContext="#stay" name="Outscoped intern">
				<DetectSpaces />
				<IncludeRules context="##Alerts" />
				<DetectIdentifier />
				<DetectChar attribute="String" context="String" char="""/>
				<IncludeRules context="##Doxygen" />
				<Detect2Chars attribute="Comment" context="Commentar 1" char="/" char1="/"/>
				<Detect2Chars attribute="Comment" context="Commentar 2" char="/" char1="*" beginRegion="Comment"/>
				<RegExpr attribute="Comment" context="Outscoped intern" String="#\s*if" beginRegion="Outscoped" firstNonSpace="true" />
				<RegExpr attribute="Comment" context="#pop" String="#\s*endif" endRegion="Outscoped" firstNonSpace="true" />
			</context>
		</contexts>
		<itemDatas>
			<itemData name="Normal Text"  defStyleNum="dsNormal"/>
			<itemData name="Keyword"      defStyleNum="dsKeyword"/>
			<itemData name="Data Type"    defStyleNum="dsDataType"/>
			<itemData name="Function"     defStyleNum="dsFunction"/>
			<itemData name="Constant"     defStyleNum="dsConstant"/>
			<itemData name="Decimal"      defStyleNum="dsDecVal"/>
			<itemData name="Event"        defStyleNum="dsEvent"/>
			<itemData name="Octal"        defStyleNum="dsBaseN"/>
			<itemData name="Hex"          defStyleNum="dsBaseN"/>
			<itemData name="Float"        defStyleNum="dsFloat"/>
			<itemData name="Char"         defStyleNum="dsChar"/>
			<itemData name="String"       defStyleNum="dsString"/>
			<itemData name="String Char"  defStyleNum="dsChar"/>
			<itemData name="Comment"      defStyleNum="dsComment"/>
			<itemData name="Symbol"       defStyleNum="dsNormal"/>
			<itemData name="Preprocessor" defStyleNum="dsOthers"/>
			<itemData name="Prep. Lib"    defStyleNum="dsOthers"/> <!--,Qt::darkYellow,Qt::yellow,false,false)); -->
			<itemData name="Alert"        defStyleNum="dsAlert" />
			<itemData name="Region Marker" defStyleNum="dsRegionMarker" />
		</itemDatas>
	</highlighting>
	<general>
		<comments>
			<comment name="singleLine" start="//" />
		</comments>
		<keywords casesensitive="1" />
	</general>
</language>

extract_keywords.pl:

#!/usr/bin/perl -w
use strict;

my $section_found;
my $skip_next;
my $prefix         = "\t\t\t";
my $wanted_section = $ARGV[0] || die "Syntax: $0 <section>";


# Fetch whole thing, replace windows CRLF with unix format
undef $/;
my $data = <STDIN>;
$data =~ s/\r\n/\n/g;

foreach my $line (split(/\n/, $data)) {
	chomp $line;

	if ( $section_found && ! $skip_next ) {
		exit(0) if ( $line =~ /^\[/ ); # Exit if we found another section

		if ( $line !~ /^(\[|#)/ ) {
			my ($keyword) = split(/\s/, $line);
			print "${prefix}<item> ${keyword} </item>\n" if defined($keyword);
		}
	}

	undef $skip_next;
	if ( $line eq "# $wanted_section" ) {
		# Found the wanted section, ignore the next line, which begins with a [
		# so that we don't recognize it as the end of a section.
		$section_found = 1;
		$skip_next = 1;
	}

}

extract_guide.sh:

#!/bin/bash
grep -e '^>ll.*</A' | sed 's/^>/\t\t\t<item> /g; s/<\/A/ <\/item>/g;' | sort | uniq

gensyntax.pl:

#!/usr/bin/perl -w
use strict;
my $sl_path = $ARGV[0] || "$ENV{HOME}/.wine/drive_c/Program Files/SecondLife";

unless( -f "${sl_path}/app_settings/keywords.ini" ) {
	die "Can't find SL at ${sl_path}, please specify the correct path as a command line argument.";
}

while( my $line = <STDIN> ) {
	chomp $line;
	if ( $line =~ m/<!--INI_EXTRACT: (.*?)-->/ ) {
		print "\t\t\t<!-- Extracted from keywords.ini: $1 -->\n";
		system("./extract_keywords.pl \"$1\" <  \"${sl_path}/app_settings/keywords.ini\" | sort");
		#print "$1\n";
	} elsif ( $line =~ m/<!--GUIDE_EXTRACT-->/ ) {
		print "\t\t\t<!-- Extracted from LSL guide -->\n";
		system("./extract_guide.sh < \"${sl_path}/lsl_guide.html\" | sort");
	} else {
		print "$line\n";
	}
}