parse-unidata.tcl

   1 #!/usr/bin/env tclsh
   2
   3 # Generate UTF-8 case mapping tables
   4 #
   5 # (c) 2010 Steve Bennett <steveb@workware.net.au>
   6 #
   7 # See LICENCE for licence details.
   8 #/
   9
  10 # Parse the unicode data from: http://unicode.org/Public/UNIDATA/UnicodeData.txt
  11 # to generate case mapping tables
  12 set map(lower) {}
  13 set map(upper) {}
  14 set map(title) {}
  15
  16 set f [open [lindex $argv 0]]
  17 while {[gets $f buf] >= 0} {
  18         set title ""
  19         set lower ""
  20         set upper ""
  21         foreach {code name class x x x x x x x x x upper lower title} [split $buf ";"] break
  22         set codex [string tolower 0x$code]
  23         if {$codex <= 0x7f} {
  24                 continue
  25         }
  26         if {$codex > 0xffff} {
  27                 break
  28         }
  29         if {![string match L* $class]} {
  30                 continue
  31         }
  32         if {$upper ne ""} {
  33                 lappend map(upper) $codex [string tolower 0x$upper]
  34         }
  35         if {$lower ne ""} {
  36                 lappend map(lower) $codex [string tolower 0x$lower]
  37         }
  38         if {$title ne "" && $title ne $upper} {
  39                 if {$title eq $code} {
  40                         set title 0
  41                 }
  42                 lappend map(title) $codex [string tolower 0x$title]
  43         }
  44 }
  45 close $f
  46
  47 foreach type {upper lower title} {
  48         puts "static const struct casemap unicode_case_mapping_$type\[\] = \{"
  49         foreach {code alt} $map($type) {
  50                 puts "\t{ $code, $alt },"
  51         }
  52         puts "\};\n"
  53 }