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 UNICODEWIDTH_H=../unicode_width.h
   9 if ! test -d unicode; then
  10         mkdir unicode
  11 fi &&
  12 ( cd unicode &&
  13         if ! test -f UnicodeData.txt; then
  14                 wget http://www.unicode.org/Public/UCD/latest/ucd/UnicodeData.txt
  15         fi &&
  16         if ! test -f EastAsianWidth.txt; then
  17                 wget http://www.unicode.org/Public/UCD/latest/ucd/EastAsianWidth.txt
  18         fi &&
  19         if ! test -d uniset; then
  20                 git clone https://github.com/depp/uniset.git
  21         fi &&
  22         (
  23                 cd uniset &&
  24                 if ! test -x uniset; then
  25                         autoreconf -i &&
  26                         ./configure --enable-warnings=-Werror CFLAGS='-O0 -ggdb'
  27                 fi &&
  28                 make
  29         ) &&
  30         UNICODE_DIR=. && export UNICODE_DIR &&
  31         cat >$UNICODEWIDTH_H <<-EOF
  32         static const struct interval zero_width[] = {
  33                 $(uniset/uniset --32 cat:Me,Mn,Cf + U+1160..U+11FF - U+00AD |
  34                   grep -v plane)
  35         };
  36         static const struct interval double_width[] = {
  37                 $(uniset/uniset --32 eaw:F,W)
  38         };
  39         EOF
  40 )