sl@0: #
sl@0: # Copyright (c) 2008-2009 Nokia Corporation and/or its subsidiary(-ies).
sl@0: # All rights reserved.
sl@0: # This component and the accompanying materials are made available
sl@0: # under the terms of "Eclipse Public License v1.0"
sl@0: # which accompanies this distribution, and is available
sl@0: # at the URL "http://www.eclipse.org/legal/epl-v10.html".
sl@0: #
sl@0: # Initial Contributors:
sl@0: # Nokia Corporation - initial contribution.
sl@0: #
sl@0: # Contributors:
sl@0: #
sl@0: # Description: 
sl@0: # See line 111 of this file.
sl@0: #
sl@0: 
sl@0: if (@ARGV != 1 && @ARGV != 2)
sl@0: 	{
sl@0: 	print <<EOD;
sl@0: Usage: perl -w cp54936_4byte.pl cp54936_4byte.txt
sl@0: EOD
sl@0: 	exit(1);
sl@0: 	}
sl@0: 
sl@0: # Removes the extenstion from the filename
sl@0: $ARGV[0] =~ m/(.*)\..*/;
sl@0: my $root = $1;
sl@0: $root =~ m/.*[\\\/]([^\\\/]*)$/;
sl@0: my $header_to_include = $1;
sl@0: 
sl@0: if (@ARGV == 2)
sl@0: 	{
sl@0: 	$ARGV[1] =~ m/(.*)\..*/;
sl@0: 	$root = $1;
sl@0: 	}
sl@0: 
sl@0: open (IN, "<$ARGV[0]") or die ("Error: $ARGV[0] $!");
sl@0: 
sl@0: my $lineNumber = 0;
sl@0: my $acceptLineNumber = 0;
sl@0: my %lines;		# hash table of all characters in format with key=foreign(string) and value=unicode(string)
sl@0: while (!eof(IN))
sl@0: 	{
sl@0: 	my $line = <IN>;
sl@0: 	$lineNumber++;
sl@0: 	if ($line =~ /^(0[xX]8[1-4]3\d[\da-fA-F]{2}3\d)\s*(0[xX][\da-fA-F]{4}).*/)
sl@0: 		{
sl@0: 		# read a line like "0x81318133	0x060D"
sl@0: 		$acceptLineNumber++;
sl@0: 		my $foreign = $1;
sl@0: 		my $unicode = $2;
sl@0: 		$lines{$foreign} = $unicode;
sl@0: 		}
sl@0: 	else
sl@0: 		{
sl@0: 		#print "Ignore line: $line";
sl@0: 		}
sl@0: 	}
sl@0: close IN;
sl@0: print "Read $ARGV[0] done.\n";
sl@0: print "$acceptLineNumber of $lineNumber lines accepted.\n";
sl@0: 
sl@0: 
sl@0: # increase input cp54936 code by 1
sl@0: # param is a string like "0x81308439"
sl@0: # return a string like "0x81308530"
sl@0: sub IncreaseCP54936Code
sl@0: 	{
sl@0: 	my ($increaseme) = @_;
sl@0: 	$increaseme =~ /0[xX]([\da-fA-F]{2})([\da-fA-F]{2})([\da-fA-F]{2})([\da-fA-F]{2})/;
sl@0: 	($b1, $b2, $b3, $b4) = (hex($1), hex($2), hex($3), hex($4));
sl@0: 	$b4++;
sl@0: 	if ($b4 == 0x3A)
sl@0: 		{
sl@0: 		$b4 = 0x30;
sl@0: 		$b3++;
sl@0: 		if ($b3 == 0xFF)
sl@0: 			{
sl@0: 			$b3 = 0x81;
sl@0: 			$b2++;
sl@0: 			if ($b2 == 0x3A)
sl@0: 				{
sl@0: 				$b2 = 0x30;
sl@0: 				$b1++;
sl@0: 				}
sl@0: 			}
sl@0: 		}
sl@0: 	return sprintf("0x%02X%02X%02X%02X", $b1, $b2, $b3, $b4);
sl@0: 	}
sl@0: 
sl@0: # return the offset from 0x81308130 to input "0x8234A235"
sl@0: sub OffsetOfCP54936Code
sl@0: 	{
sl@0: 	$_[0] =~ /0[xX]([\da-fA-F]{2})([\da-fA-F]{2})([\da-fA-F]{2})([\da-fA-F]{2})/;
sl@0: 	($b1, $b2, $b3, $b4) = (hex($1), hex($2), hex($3), hex($4));
sl@0: 	return ($b1-0x81)*12600 + ($b2-0x30)*1260 + ($b3-0x81)*10 + ($b4-0x30);
sl@0: 	}
sl@0: 
sl@0: # return the last byte of input "0x8234A235"
sl@0: sub Byte4OfCP54936Code
sl@0: 	{
sl@0: 	$_[0] =~ /0[xX]([\da-fA-F]{2})([\da-fA-F]{2})([\da-fA-F]{2})([\da-fA-F]{2})/;
sl@0: 	return hex($4);
sl@0: 	}
sl@0: 
sl@0: 
sl@0: print "Write to $root.cpp...\n";
sl@0: open (CPP, ">$root.cpp") or die ("Error: $ARGV[0] Can't open cpp file");
sl@0: 
sl@0: print CPP <<EOD;
sl@0: // Copyright (c) 2008-2009 Nokia Corporation and/or its subsidiary(-ies).
sl@0: // All rights reserved.
sl@0: // This component and the accompanying materials are made available
sl@0: // under the terms of the License "Eclipse Public License v1.0"
sl@0: // which accompanies this distribution, and is available
sl@0: // at the URL "http://www.eclipse.org/legal/epl-v10.html".
sl@0: //
sl@0: // Initial Contributors:
sl@0: // Nokia Corporation - initial contribution.
sl@0: //
sl@0: // Contributors:
sl@0: //
sl@0: // Description:
sl@0: //
sl@0: // Auto-generated by the cp54936_4byte_tounicode.pl tool - Do not edit!!!
sl@0: //
sl@0: 
sl@0: #include <e32std.h>
sl@0: #include <e32def.h>
sl@0: #include "cp54936.h"
sl@0: 
sl@0: 
sl@0: // mapping table of: CP54936 4-byte in-BMP ---> Unicode
sl@0: // To calculate index: index=(b1-144)*12600+(b2-48)*1260+(b3-129)*10+(b4-48), in which,
sl@0: // b1,b2,b3,b4 is byte1,2,3,4 of CP54936 code.
sl@0: // For example, CP54936 code 0x8232EA38, the index=(0x82-144)*12600+(0x32-48)*1260+(0xEA-129)*10+(0x38-48)=16178
sl@0: // So we get the Unicode 0x42AB.
sl@0: // Generated with: \"perl -w ..\\group\\cp54936_4byte_tounicode.pl cp54936_4byte.txt cp54936_4byte_tounicode.cpp\".
sl@0: 
sl@0: EOD
sl@0: 
sl@0: my $bytecount = 0;
sl@0: my $expect = "0x81308130";
sl@0: my $last = "0x8431A439";
sl@0: my $totalCount = OffsetOfCP54936Code($last) + 1;
sl@0: 
sl@0: 
sl@0: print CPP "const TUint16 KMappingTable4ByteBmp2Unicode[$totalCount] =\n\t{\n\t";
sl@0: 
sl@0: my $outIndex = 0;	# to wrap every 10 items
sl@0: while (OffsetOfCP54936Code($expect) <= OffsetOfCP54936Code($last))
sl@0: 	{
sl@0: 	if (!exists($lines{$expect}))
sl@0: 		{
sl@0: 		print CPP "0xFFFD, ";
sl@0: 		}
sl@0: 	else
sl@0: 		{
sl@0: 		print CPP "$lines{$expect}, ";
sl@0: 		}
sl@0: 	$bytecount += 2;
sl@0: 	$outIndex++;
sl@0: 	if ($outIndex % 10 == 0)
sl@0: 		{
sl@0: 		print CPP "\t// $expect\n\t";
sl@0: 		}
sl@0: 	# to next foreign
sl@0: 	$expect = IncreaseCP54936Code($expect);
sl@0: 	}
sl@0: 
sl@0: print CPP "};\n";
sl@0: print CPP "// total byte count = $bytecount\n";
sl@0: print "\nTotal byte count: $bytecount.\n";
sl@0: close CPP;
sl@0: print "Done.\n";