#!/usr/bin/env perl
#
#(c) Copyright 1998-2009 Apostolos Syropoulos
# asyropoulos@yahoo.com
#
# The LaTeX Project Public License (lppl)
# This software is copyright but you are granted a license which gives you,
# the "user" of the software, legal permission to copy, distribute, and/or
# modify the software. However, if you modify the software and then distribute
# it (even just locally) you must change the name of the software, or use other
# technical means to avoid confusion.
#
$makeindex = "makeindex"; #name of the index generation utility
sub g2e
{
$word = $_[0];
$tword = "";
foreach $_ (split(//, $word))
{
if (/^a|^A|^�|^�|^�|^\xA2|^\xB6/) { $tword .= "a";}
elsif (/^b|^B|^�|^�/) { $tword .= "b"; }
elsif (/^d|^D|^�|^�/) { $tword .= "d"; }
elsif (/^e|^E|^�|^�|^�|^�/) { $tword .= "e"; }
elsif (/^i|^I|^�|^�|^�|^�|^�/) { $tword .= "i"; }
elsif (/^o|^O|^�|^�|^�|^�/) { $tword .= "o"; }
elsif (/^p|^P|^�|^�/) { $tword .= "p"; }
elsif (/^g|^G|^�|^�/) { $tword .= "c"; }
elsif (/^z|^Z|^�|^�/) { $tword .= "f"; }
elsif (/^h|^H|^�|^�|^�|^�/) { $tword .= "g"; }
elsif (/^j|^J|^�|^�/) { $tword .= "h"; }
elsif (/^k|^K|^�|^�/) { $tword .= "j"; }
elsif (/^l|^L|^�|^�/) { $tword .= "k"; }
elsif (/^m|^M|^�|^�/) { $tword .= "l"; }
elsif (/^n|^N|^�|^�/) { $tword .= "m"; }
elsif (/^x|^X|^�|^�/) { $tword .= "n"; }
elsif (/^r|^R|^�|^�/) { $tword .= "q"; }
elsif (/^s|^c|^S|^�|^�|^�/) { $tword .= "r"; }
elsif (/^t|^T|^�|^�/) { $tword .= "s"; }
elsif (/^u|^U|^�|^�|^�|^�|^�/) { $tword .= "t"; }
elsif (/^f|^F|^�|^�/) { $tword .= "u"; }
elsif (/^q|^Q|^�|^�/) { $tword .= "v"; }
elsif (/^y|^Y|^�|^�/) { $tword .= "w"; }
elsif (/^w|^W|^�|^�|^�|^�/) { $tword .= "x"; }
elsif (/^'|^`|^~|^<|^>/) { }
else { $tword .= $_ }
}
return $tword;
}
$argc = @ARGV;
if ($argc == 0 || $argc > 4 ) # no command line arguments or more than 3
{ # arguments
die "Usage: mkgrkindex.pl [-s A|a] [-l] index\n";
}
else
{
print "This is mkgrkindex (previously known as mkindex) version 2.2\n";
$style = 0;
$is_latin = 0;
SWITCHES:
while($_ = $ARGV[0])
{
shift;
if (/^-s/)
{
if ($ARGV[0] eq "a")
{
$style = -1;
}
elsif ($ARGV[0] eq "A")
{
$style = 1;
}
else
{
die "$ARGV[0]: Illegal argument for switch -s\n";
}
shift;
}
elsif (/^-l/)
{
$is_latin = 1;
}
elsif (/^-\w+/)
{
die "$_: Illegal command line switch!\n";
}
else
{
$file = $_;
}
}
die "No index file name specified!\n" if $file eq "";
if (! (-e $file))
{
die "$file: no such file!\n" if $file =~ /.+\..+/;
die "$file.idx: no such file!\n" if (! (-e "$file.idx"));
$index_file = "$file.idx";
}
else
{
$index_file = $file;
$file = $1 if $index_file =~ /(.+)\..+/;
}
}
$old_file="$index_file.old";
if ($is_latin == 0)
{
rename $index_file, $old_file;
open(OLD, "<$old_file") || die "Can't open file $old_file\n";
open(NEW, ">$index_file") || die "Can't open file $index_file\n";
while (<OLD>)
{
#\x7B = {, \x7D = },\x40 = @, \x7C = |, \x21 = !, \x28 = (, \x29 = )
chomp;
#Nikos Platis has suggested the replacement of the original code with a better
#parsing method. The code that follows is essentially Nikos's code with
#some minnor modifications, as explained in the documentation.
$newentry = "";
/^\\indexentry\x7B(.+)\x7D(.+)/;
$fullentry = $1;
$page = $2;
($indexentry, $format) = split(/(?<!")\x7C/, $fullentry);
@entryparts = split(/(?<!")\x21/, $indexentry);
$numparts = @entryparts;
for ($i = 0; $i < $numparts; $i++) {
($x, $y) = split(/(?<!")\x40/, @entryparts[$i]);
if ($i > 0) {
$newentry .= "!"
}
$tx = g2e($x);
if ($y) {
$newentry .= "$tx\@$y"
}
else {
$newentry .= "$tx\@$x";
}
}
print NEW "\\indexentry{$newentry";
if ($format) {
print NEW "|$format"
}
print NEW "}$page\n"
}
close OLD;
close NEW;
unlink $old_file;
}
rename $index_file, $old_file;
open(OLD, "<$old_file") || die "Can't open file $old_file\n";
open(NEW, ">$index_file") || die "Can't open file $index_file\n";
while(<OLD>)
{
chomp($_);
if (/\\indexentry\x7B(.+)\x7D\x7B\\textlatin\s*?\x7B(\w+)\x7D\x7D/)
{
print NEW "\\indexentry{$1|textlatin}{$2}\n";
}
else
{
print NEW "$_\n";
}
}
close OLD;
close NEW;
unlink $old_file;
if ($style < 0)
{
system("$makeindex -s lowercase-headers.ist $file");
}
elsif ($style > 0)
{
system("$makeindex -s uppercase-headers.ist $file");
}
else
{
system("$makeindex $file");
}
if ($style != 0)
{
%e2g = (
'a' => '�', 'A' => '�',
'b' => '�', 'B' => '�',
'c' => '�', 'C' => '�',
'd' => '�', 'D' => '�',
'e' => '�', 'E' => '�',
'f' => '�', 'F' => '�',
'g' => '�', 'G' => '�',
'h' => '�', 'H' => '�',
'i' => '�', 'I' => '�',
'j' => '�', 'J' => '�',
'k' => '�', 'K' => '�',
'l' => '�', 'L' => '�',
'm' => '�', 'M' => '�',
'n' => '�', 'N' => '�',
'o' => '�', 'O' => '�',
'p' => '�', 'P' => '�',
'q' => '�', 'Q' => '�',
'r' => '�v', 'R' => '�',
's' => '�', 'S' => '�',
't' => '�', 'T' => '�',
'u' => '�', 'U' => '�',
'v' => '�', 'V' => '�',
'w' => '�', 'W' => '�',
'x' => '�', 'X' => '�'
);
$ind_file = "$file.ind";
$old_file="$ind_file.old";
if ($is_latin == 0)
{
rename $ind_file, $old_file;
open(OLD, "<$old_file") || die "Can't open file $old_file\n";
open(NEW, ">$ind_file") || die "Can't open file $ind_file\n";
while (<OLD>)
{
if (/^{\\hfil (\w?) \\hfil}/)
{
$lettergr = $e2g{$1};
print NEW "{\\hfil $lettergr \\hfil}\\nopagebreak\n";
}
else
{
print NEW ;
}
}
close OLD;
close NEW;
unlink $old_file;
}
rename $ind_file, $old_file;
open(OLD, "<$old_file") || die "Can't open file $old_file\n";
open(NEW, ">$ind_file") || die "Can't open file $ind_file\n";
while (<OLD>)
{
if (/^\x7B\\hfil (\w)ymbols \\hfil\x7D/)
{
if ($1 eq "s")
{
print NEW "{\\hfil \\textgreek{\\textbf{s'umbola}}";
print NEW "\\hfil}\\nopagebreak\n";
}
elsif ($1 eq "S")
{
print NEW "{\\hfil \\textgreek{\\textbf{S'umbola}}";
print NEW "\\hfil}\\nopagebreak\n";
}
else
{
die "Illegal header $1 in .ind file\n";
}
}
else
{
print NEW;
}
}
close OLD;
close NEW;
unlink $old_file;
}
__END__