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#
8UNICODEWIDTH_H=../unicode_width.h
9if ! test -d unicode; then
10 mkdir unicode
11fi &&
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 echo "static const struct interval zero_width[] = {" >$UNICODEWIDTH_H &&
31 UNICODE_DIR=. ./uniset/uniset --32 cat:Me,Mn,Cf + U+1160..U+11FF - U+00AD |
32 grep -v plane >>$UNICODEWIDTH_H &&
33 echo "};" >>$UNICODEWIDTH_H &&
34 echo "static const struct interval double_width[] = {" >>$UNICODEWIDTH_H &&
35 UNICODE_DIR=. ./uniset/uniset --32 eaw:F,W >>$UNICODEWIDTH_H &&
36 echo "};" >>$UNICODEWIDTH_H
37)