contrib/update-unicode/update_unicode.sh

   1 #!/bin/sh
   2 #See http://www.unicode.org/reports/tr44/
   3 #
   4 #Me Enclosing_Mark  an enclosing combining mark
   5 #Mn Nonspacing_Mark a nonspacing combining mark (zero advance width)
   6 #Cf Format          a format control character
   7 #
   8 cd "$(dirname "$0")"
   9 UNICODEWIDTH_H=$(git rev-parse --show-toplevel)/unicode_width.h
  10 (
  11         if ! test -f UnicodeData.txt; then
  12                 wget http://www.unicode.org/Public/UCD/latest/ucd/UnicodeData.txt
  13         fi &&
  14         if ! test -f EastAsianWidth.txt; then
  15                 wget http://www.unicode.org/Public/UCD/latest/ucd/EastAsianWidth.txt
  16         fi &&
  17         if ! test -d uniset; then
  18                 git clone https://github.com/depp/uniset.git
  19         fi &&
  20         (
  21                 cd uniset &&
  22                 if ! test -x uniset; then
  23                         autoreconf -i &&
  24                         ./configure --enable-warnings=-Werror CFLAGS='-O0 -ggdb'
  25                 fi &&
  26                 make
  27         ) &&
  28         UNICODE_DIR=. && export UNICODE_DIR &&
  29         cat >$UNICODEWIDTH_H <<-EOF
  30         static const struct interval zero_width[] = {
  31                 $(uniset/uniset --32 cat:Me,Mn,Cf + U+1160..U+11FF - U+00AD |
  32                   grep -v plane)
  33         };
  34         static const struct interval double_width[] = {
  35                 $(uniset/uniset --32 eaw:F,W)
  36         };
  37         EOF
  38 )