ICU 56 step 6: Check in the pre-built ICU data

scripts/single_byte_gen.sh

 #!/bin/bash
 # 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.
 
 function preamble {
 
 encoding="$1"
 cat <<PREAMBLE
 # ***************************************************************************
@@ skipping 20 matching lines @@
 # The list of html5 encodings. Note that iso-8859-8-i is not listed here
 # because its mapping table is exactly the same as iso-8859-8. The difference
 # is BiDi handling (logical vs visual).
 encodings="ibm866 iso-8859-2 iso-8859-3 iso-8859-4 iso-8859-5 iso-8859-6\
            iso-8859-7 iso-8859-8 iso-8859-10 iso-8859-13 iso-8859-14\
            iso-8859-15 iso-8859-16 koi8-r koi8-u macintosh\
            windows-874 windows-1250 windows-1251 windows-1252 windows-1253\
            windows-1254 windows-1255 windows-1256 windows-1257 windows-1258\
            x-mac-cyrillic"
 
-ENCODING_DIR="$(dirname $0)/../source/data/mappings"
+ENCODING_DIR="$(dirname "$0")/../source/data/mappings"
 for e in ${encodings}
 do
   output="${ENCODING_DIR}/${e}-html.ucm"
   index="index-${e}.txt"
   indexurl="https://encoding.spec.whatwg.org/index-${e}.txt"
   curl -o ${index} "${indexurl}"
   preamble ${e} > ${output}
   awk 'BEGIN \
        { \
          for (i=0; i < 0x80; ++i) \
          { \
            printf("<U%04X> \\x%02X |0\n", i, i);} \
          } \
        !/^#/ && !/^$/ \
        {
          printf ("<U%4s> \\x%02X |0\n", substr($2, 3), $1 + 0x80); \
        }' ${index} | sort >> ${output}
   echo 'END CHARMAP' >> ${output}
   rm ${index}
 done