1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
|
#!/bin/bash
# -*- sh -*-
# Takes on stdin a file formated like the official UnicodeData list;
# Writes on stdout a fallback table suitable for `consolechars -k'
# Note: we assume all diacrs match a non-diacr !
if [ $# != 3 ]
then
echo >&2 "Usage: $(basename $0) FULL_PATTERN REDUCTED_PATTERN TRANSLATION"
echo >&2 " < /usr/share/unicode/UnicodeData.txt >out.fallback"
exit 1
fi
FULL_GREP_PATTERN=$1
REDUCTED_PATTERN=$2
TRANSLATION=$3
FULL_SED_PATTERN=$(echo ${FULL_GREP_PATTERN} | sed -e 's/?/\\{0,1\\}/g' -e 's/\([()|]\)/\\\1/g')
FILE=/tmp/file.$$
cut -d\; -f1,2 >$FILE
cat << EOF
#
# Fallback file from chars matching:
# '${FULL_GREP_PATTERN}'
# to chars matching:
# '${REDUCTED_PATTERN}'
# using as a translation rule:
# '${TRANSLATION}'
#
# Auto-generated by $(basename $0) from UnicodeData list
#
# ** DO NOT EDIT MANUALLY **
#
EOF
cat < $FILE |
# select only targeted lines
egrep -f <(echo ";${FULL_GREP_PATTERN}$") |
# filter out those we want to reduct to
egrep -v -f <(echo ";${REDUCTED_PATTERN}$") |
# translate into wished reduction, and keep orig name as 3rd field
sed -f <(echo "s/;${FULL_SED_PATTERN}$/;${TRANSLATION}\0/") |
# translate into wished unicode
sed -f <(
# construct a sed filter that adds their unicodes to reduction names
egrep < $FILE ";${REDUCTED_PATTERN}$" |
awk -F\; '{print "s/;" $2 ";/;U+" $1 "\\0/"}'
) |
# filter out lines that did not give a reducted pattern
grep ';U+' |
# format lines in fallback-file syntax
awk -F\; '{print "# " $4 " : " $3 "\nU+" $1 " " $2}'
rm $FILE
|
