src/backend/utils/mb/Unicode/UCS_to_UHC.pl

   1 #! /usr/bin/perl
   2 #
   3 # Copyright (c) 2007-2022, PostgreSQL Global Development Group
   4 #
   5 # src/backend/utils/mb/Unicode/UCS_to_GB18030.pl
   6 #
   7 # Generate UTF-8 <--> UHC code conversion tables from
   8 # "windows-949-2000.xml", obtained from
   9 # http://source.icu-project.org/repos/icu/data/trunk/charset/data/xml/
  10 #
  11 # The lines we care about in the source file look like
  12 #    <a u="009A" b="81 30 83 36"/>
  13 # where the "u" field is the Unicode code point in hex,
  14 # and the "b" field is the hex byte sequence for UHC
  15
  16 use strict;
  17 use warnings;
  18
  19 use convutils;
  20
  21 my $this_script = 'src/backend/utils/mb/Unicode/UCS_to_UHC.pl';
  22
  23 # Read the input
  24
  25 my $in_file = "windows-949-2000.xml";
  26
  27 open(my $in, '<', $in_file) || die("cannot open $in_file");
  28
  29 my @mapping;
  30
  31 while (<$in>)
  32 {
  33         next if (!m/<a u="([0-9A-F]+)" b="([0-9A-F ]+)"/);
  34         my ($u, $c) = ($1, $2);
  35         $c =~ s/ //g;
  36         my $ucs  = hex($u);
  37         my $code = hex($c);
  38
  39         next if ($code == 0x0080 || $code == 0x00FF);
  40
  41         if ($code >= 0x80 && $ucs >= 0x0080)
  42         {
  43                 push @mapping,
  44                   {
  45                         ucs       => $ucs,
  46                         code      => $code,
  47                         direction => BOTH,
  48                         f         => $in_file,
  49                         l         => $.
  50                   };
  51         }
  52 }
  53 close($in);
  54
  55 # One extra character that's not in the source file.
  56 push @mapping,
  57   {
  58         direction => BOTH,
  59         code      => 0xa2e8,
  60         ucs       => 0x327e,
  61         comment   => 'CIRCLED HANGUL IEUNG U',
  62         f         => $this_script,
  63         l         => __LINE__
  64   };
  65
  66 print_conversion_tables($this_script, "UHC", \@mapping);