Update CJK converters and their generating scripts

scripts/sjis_gen.sh

 #!/bin/sh
 # Copyright 2014 The Chromium Authors. All rights reserved.
 # Use of this source code is governed by a BSD-style license that can be
 # found in the LICENSE file.
 
 # References:
 #   http://encoding.spec.whatwg.org/#shift_jis
 
 # Download the following file, run it in source/data/mappings directory
 # and save the result to euc-jp-html5.ucm
 #   http://encoding.spec.whatwg.org/index-jis0208.txt
 
 function preamble {
 cat <<PREAMBLE
 # ***************************************************************************
 # *
 # *   Copyright (C) 1995-2014, International Business Machines
 # *   Corporation and others.  All Rights Reserved.
 # *
 # *   Generated per the algorithm for Shift_JIS
 # *   described at http://encoding.spec.whatwg.org/#shift_jis
 # *
 # ***************************************************************************
 <code_set_name>               "shift_jis-html5"
 <char_name_mask>              "AXXXX"
 <mb_cur_max>                  2
 <mb_cur_min>                  1
 <uconv_class>                 "MBCS"
-<subchar>                     \xFC\xFC
-<subchar1>                    \x7F
+<subchar>                     \x3F
 <icu:charsetFamily>           "ASCII"
 
-<icu:state>                   0-80, 81-9f:1, a1-df, e0-fc:1
+<icu:state>                   0-80, 81-9f:1, a1-df, e0-fc:1, 82:3, 84:4, 85-86:2, 87:5, 88:2, 98:6, eb-ec:2, ef:2, f9:2, fc:7
+
 <icu:state>                   40-7e, 80-fc
+<icu:state>                   80-fc
+<icu:state>                   4f-7e, 80-fc, 59-5f.i, 7a-7e.i
+<icu:state>                   40-7e, 80-fc, 61-6f.i
+<icu:state>                   40-7e, 80-fc, 76-7d.i
+<icu:state>                   40-7e, 80-fc, 73-7e.i
+<icu:state>                   40-4b, 80-fc
+
 
 CHARMAP
 PREAMBLE
 }
 
 # The encoding spec for Shift_JIS says U+0080 has to be round-tripped with
 # 0x80. So, this is one character more than ASCII up to 128 (0x80).
 function ascii {
   for i in $(seq 0 128)
   do
@@ skipping 65 matching lines @@
 function unsorted_table {
   ascii
   half_width_kana
   jis208
   eudc "0x40" "0x7E" "0x40"
   eudc "0x80" "0xFC" "0x41"
   echo '<U00A5> \x5C |1'
   echo '<U203E> \x7E |1'
 }
 
+wget -N -r -nd http://encoding.spec.whatwg.org/index-jis0208.txt

[jsbell, 2015/03/18 21:38:31]

https

 preamble
 unsorted_table | sort  | uniq
 echo 'END CHARMAP'