opencog-link-grammar / morphology / fa / scripts / isiri2unicode_1-1.pl

# Created by Jon Dehdari 2002
# Perl 5.6
# Converts Isiri 3342  Farsi text to Unicode HTML Page
# To Do: auto-insert alef before short vowels, change final yeh to farsi style yeh

%isiri2unicode = ( 	
		"\xb0" => '۰', # EXTENDED ARABIC-INDIC DIGIT ZERO # Farsi specific
		"\xb1" => '۱', # EXTENDED ARABIC-INDIC DIGIT ONE # Farsi specific
		"\xb2" => '۲', # EXTENDED ARABIC-INDIC DIGIT TWO # Farsi specific
		"\xb3" => '۳', # EXTENDED ARABIC-INDIC DIGIT THREE # Farsi specific
		"\xb4" => '۴', # EXTENDED ARABIC-INDIC DIGIT FOUR # Farsi specific
		"\xb5" => '۵', # EXTENDED ARABIC-INDIC DIGIT FIVE # Farsi specific
		"\xb6" => '۶', # EXTENDED ARABIC-INDIC DIGIT SIX # Farsi specific
		"\xb7" => '۷', # EXTENDED ARABIC-INDIC DIGIT SEVEN # Farsi specific
		"\xb8" => '۸', # EXTENDED ARABIC-INDIC DIGIT EIGHT # Farsi specific
		"\xb9" => '۹', # EXTENDED ARABIC-INDIC DIGIT NINE # Farsi specific
		
		"\xc1" => 'ا', #etc. # ARABIC LETTER ALEF
		"\xc3" => 'ب', # ARABIC LETTER BEH
		"\xc4" => 'پ', # ARABIC LETTER PEH # Farsi specific
		"\xc5" => 'ت', # ARABIC LETTER TEH
		"\xc6" => 'ث', # ARABIC LETTER THEH
		"\xc7" => 'ج', # ARABIC LETTER JEEM
		"\xc8" => 'چ', # ARABIC LETTER TCHEH # Farsi specific
		"\xc9" => 'ح', # ARABIC LETTER HAH
		"\xca" => 'خ', # ARABIC LETTER KHAH
		"\xcb" => 'د', # ARABIC LETTER DAL
		"\xcc" => 'ذ', # ARABIC LETTER THAL
		"\xcd" => 'ر', # ARABIC LETTER REH
		"\xce" => 'ز', # ARABIC LETTER ZAIN
		"\xcf" => 'ژ', # ARABIC LETTER JEH # Farsi specific
		"\xd0" => 'س', # ARABIC LETTER SEEN
		"\xd1" => 'ش', # ARABIC LETTER SHEEN
		"\xd2" => 'ص', # ARABIC LETTER SAD
		"\xd3" => 'ض', # ARABIC LETTER DAD
		"\xd4" => 'ط', # ARABIC LETTER TAH
		"\xd5" => 'ظ', # ARABIC LETTER ZAH
		"\xd6" => 'ع', # ARABIC LETTER AIN
		"\xd7" => 'غ', # ARABIC LETTER GHAIN
		"\xd8" => 'ف', # ARABIC LETTER FEH
		"\xd9" => 'ق', # ARABIC LETTER QAF
		"\xda" => 'ک', # ARABIC LETTER KEHEH # Farsi specific
		"\xdb" => 'گ', # ARABIC LETTER GAF # Farsi specific
		"\xdc" => 'ل', # ARABIC LETTER LAM
		"\xdd" => 'م', # ARABIC LETTER MEEM
		"\xde" => 'ن', # ARABIC LETTER NOON
		"\xdf" => 'و', #  ARABIC LETTER WAW #note "u", "v", and "w" are all "vav"s in Farsi orthography.
		#"\xdf" => 'و', # ARABIC LETTER WAW
		#"\xdf" => 'و', # ARABIC LETTER WAW
		"\xe0" => 'ه', # ARABIC LETTER HEH
		#"\xd2" => 'ۀ', # ARABIC LETTER HEH WITH YEH ABOVE
		"\xe1" => 'ي', #  ARABIC LETTER YEH #note "i" and  "y" are both "ya"s in Farsi orthography.
		#"\xe1" => 'ي', # ARABIC LETTER YEH
		"\xfb" => 'ئ', # ARABIC LETTER YEH WITH HAMZA ABOVE #note "`" and "I" are the same in Farsi orthography.
		#"\xd2" => 'ئ', # ARABIC LETTER YEH WITH HAMZA ABOVE
		"\xf0" => 'َ', #  ARABIC FATHA #note the short vowels "a", "o" and "e" do not normally appear in Farsi orthography.
		"\xf2" => 'ُ', # ARABIC DAMMA
		"\xf1" => 'ِ', # ARABIC KASRA
		
		"\xc0" => 'آ', # ARABIC LETTER ALEF WITH MADDA ABOVE
		"\xc2" => 'ء', # ARABIC LETTER HAMZA
		"\xf8" => 'أ', # ARABIC LETTER ALEF WITH HAMZA ABOVE
		"\xfa" => 'ؤ', # ARABIC LETTER WAW WITH HAMZA ABOVE
		"\xf9" => 'إ', # ARABIC LETTER ALEF WITH HAMZA BELOW
		"\xfb" => 'ئ', # ARABIC LETTER YEH WITH HAMZA ABOVE
		
		"\xa5" => '٪', # ARABIC PERCENT SIGN	
		"\xag" => '،', # ARABIC COMMA
		"\xbb" => '؛', # ARABIC SEMICOLON
		"\xbf" => '؟', # ARABIC QUESTION MARK
		
		"\xa0" => ' ', # SPACE (really  )
		"\xa8" => '(', # LEFT PARENTHESIS
		"\xa9" => ')', # RIGHT PARENTHESIS
		"\xad" => '-', # HYPHEN-MINUS
		"\xa6" => '.', # FULL STOP (really .)
		"\xba" => ':', # COLON

	);

$in = shift || die "Provide a valid input file argument\n";
$out= shift || die "Provide an output file argument\n";
$in ne $out || die "Input and output files cannot be the same\n";

open(IN,$in);
open(OUT,">$out");
	
#prints the top part of an html page
print OUT ( 
	'<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//FA">',
	"\n",
	'<html lang="fa">',
	"\n",
	"\n",
	"<style>\n",
	"    body \{ text-align\:right \}\n",
	"<\/style>\n",
	"<head>\n",
	"<title><\/title>\n",
	"<\/head>\n",
	"<body>\n"
);
#prints the main part of the page (the transliterated characters)
while ($line = <IN>) {
	@charx = split //, $line;
	foreach $charx (@charx)
	{
		$newchar = $isiri2unicode{$charx}; 
		print OUT ($newchar);
	}
	#prints a carrage return after every line
	print OUT ("<br>");
}
#prints the bottom part of the html page
print OUT ( "\n<\/body>\n<\/html>\n");

#while ($line = <IN>) {
	#@readcontents = <IN>;
	#foreach $tempcontents (@readcontents) {
            #print $tempcontents, "\n";		

		## inserts initial alef before short vowels "a", "e", and "o".
		#$tempcontents =~ s/ &#1614;/ &#1571;&#1614;/g;	# "a"	
		#$tempcontents =~ s/ &#1615;/ &#1571;&#1615;/g;    # "o"
		#$tempcontents =~ s/ &#1616;/ &#1573;&#1616;/g;    # "e"
		#$tempcontents =~ s/ &#1573;&#1616; / &#1616; /g;    # "e" corrects for ezafe constructs
		
		## changes initial long "A" to alef with madda above.
		#$tempcontents =~ s/ &#1575;/ &#1570;/g; # ARABIC LETTER ALEF WITH MADDA ABOVE
		
		## changes final arabic "k" to final farsi "k".
		#$tempcontents =~ s/&#1603; /&#1705; /g;
		
		## changes final arabic "y" to final farsi "y".
		#$tempcontents =~ s/&#1610; /&#1740; /g;  # ararbic alef maksura is &#1609;
		
		
		## reverses punctuation marks.
		#$tempcontents =~ s/?/&#1567;/g;		
		#$tempcontents =~ s/,/&#1548;/g;
		#$tempcontents =~ s/;/&#1563;/g; #must be last!

		## maintains carriage returns for web pages
		#$tempcontents =~ s/\n/<br>/g;  # currently this function barfs at me.
	#}

#while (@readcontents)
#{
	#while (@readcontents) {
		#print OUT (@readcontents);
		#@readcontents = <IN>;
	#}
	
	### Writes to Output File
	#while (@readcontents) {		
		### Print data to file
		
		#print "@readcontents" || warn ("didn't print the new words in $newfile \n$!");
		#@readcontents = <OUT>;
	#}
#}
}
close (OUT);
close (IN);