#!/bin/sh
+# Copyright (C) 2015 Free Software Foundation, Inc.
+
# Copyright (C) 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011
# National Institute of Advanced Industrial Science and Technology (AIST)
# Registration Number H13PRO009
# GLIBC-1 GLIBC-2 GLIBC-2-7 CZYBORRA IANA UNICODE UNICODE2 YASUOKA
# $4: awk script
-FILE="admin/charsets/$1"
-BASE=`basename $1 .gz`
+## So that eg [A-F] as used by KANJI-DATABASE branch below works as expected.
+## Otherwise with LANG=en_US.utf8, CNS-6.map was generated with a
+## bogus entry. By experiment, LC_COLLATE=C was not enough.
+export LC_ALL=C
+
+BASE=`expr "$1" : '.*/\(.*\)' '|' "$1"` # basename
+FILE="admin/charsets/mapfiles/$BASE"
+BASE=`expr "$BASE" : '\(.*\)\.gz$' '|' "$BASE"` # remove any .gz suffix
+AWK=${AWK:-awk}
case "$3" in
GLIBC*)
- FILE="$BASE in localedate/charmaps of glibc";
+ FILE="$BASE in localedata/charmaps of glibc";
SOURCE="";;
CZYBORRA)
BASE="$BASE.gz";
if [ -n "$4" ] ; then
if [ -f "$4" ] ; then
- AWKPROG="gawk -f $4"
+ AWKPROG="$AWK -f $4"
else
echo "Awk program does not exist: $4"
exit 1
if [ "$3" = "GLIBC-1" ] ; then
# Source format is:
# <UYYYY> /xXX
- zcat $1 | sed -n -e "$2 p" \
+ gunzip -c $1 | sed -n -e "${2}p" \
| sed -e 's,<U\([^>]*\)>[ ]*/x\(..\).*,0x\2 0x\1,' \
| sort | ${AWKPROG}
elif [ "$3" = "GLIBC-2" ] ; then
# Source format is:
# <UYYYY> /xXX/xZZ
- zcat $1 | sed -n -e "$2 p" \
+ gunzip -c $1 | sed -n -e "${2}p" \
| sed -e 's,<U\([^>]*\)>[ ]*/x\(..\)/x\(..\).*,0x\2\3 0x\1,' \
| sort | ${AWKPROG}
elif [ "$3" = "GLIBC-2-7" ] ; then
# Source format is:
# <UYYYY> /xXX/xZZ
# We must drop MSBs of XX and ZZ
- zcat $1 | sed -n -e "$2 p" \
+ gunzip -c $1 | sed -n -e "${2}p" \
| sed -e 's/xa/x2/g' -e 's/xb/x3/g' -e 's/xc/x4/g' \
-e 's/xd/x5/g' -e 's/xe/x6/g' -e 's/xf/x7/g' \
-e 's,<U\([^>]*\)>[ ]*/x\(..\)/x\(..\).*,0x\2\3 0x\1,' \
- | tee temp \
| sort | ${AWKPROG}
elif [ "$3" = "CZYBORRA" ] ; then
# Source format is:
# =XX U+YYYY
- zcat $1 | sed -n -e "$2 p" \
+ sed -n -e "${2}p" < $1 \
| sed -e 's/=\(..\)[^U]*U+\([0-9A-F]*\).*/0x\1 0x\2/' \
| sort | ${AWKPROG}
elif [ "$3" = "IANA" ] ; then
# Source format is:
# 0xXX 0xYYYY
- zcat $1 | sed -n -e "$2 p" \
+ sed -n -e "${2}p" < $1 \
| sed -e 's/\(0x[0-9A-Fa-f]*\)[^0]*\(0x[0-9A-Fa-f]*\).*/\1 \2/' \
| sort | ${AWKPROG}
elif [ "$3" = "UNICODE" ] ; then
# YYYY XX
# We perform reverse sort to prefer the first one in the
# duplicated mappings (e.g. 0x20->U+0020, 0x20->U+00A0).
- zcat $1 | sed -n -e "$2 p" \
+ sed -n -e "${2}p" < $1 \
| sed -e 's/\([0-9A-F]*\)[^0-9A-F]*\([0-9A-F]*\).*/0x\2 0x\1/' \
| sort -r
elif [ "$3" = "UNICODE2" ] ; then
# Source format is:
# 0xXXXX 0xYYYY # ...
- zcat $1 | sed -n -e "$2 p" \
+ sed -n -e "${2}p" < $1 \
| sed -e 's/\([0-9A-Fx]*\)[^0]*\([0-9A-Fx]*\).*/\1 \2/' \
| ${AWKPROG} | sort -n -k 4,4
elif [ "$3" = "YASUOKA" ] ; then
# Source format is:
# YYYY 0-XXXX (XXXX is a Kuten code)
- zcat $1 | sed -n -e "$2 p" \
+ sed -n -e "${2}p" < $1 \
| sed -e 's/\([0-9A-F]*\)[^0]*0-\([0-9]*\).*/0x\2 0x\1/' \
| sort | ${AWKPROG}
elif [ "$3" = "KANJI-DATABASE" ] ; then
# Source format is:
# C?-XXXX U+YYYYY .....
- zcat $1 | sed -n -e "$2 p" \
+ sed -n -e "${2}p" < $1 \
| sed -e 's/...\(....\) U+\([0-9A-F]*\).*/0x\1 0x\2/' \
| sort | ${AWKPROG}
else