summaryrefslogtreecommitdiff
path: root/admin/charsets/mapconv
diff options
context:
space:
mode:
authorGlenn Morris <rgm@gnu.org>2015-05-21 23:44:00 -0700
committerGlenn Morris <rgm@gnu.org>2015-05-21 23:44:00 -0700
commite7bc85db46679cc408f51d74e6a66734f1e32142 (patch)
tree250008ff1c624cb5cc39ceb2fcf7892cdb93c258 /admin/charsets/mapconv
parent9e41e0bc6a6fdb7a01841db62d8de42d67be534a (diff)
downloademacs-e7bc85db46679cc408f51d74e6a66734f1e32142.tar.gz
Generate admin/charsets Makefile via configure, and make more portable.
* configure.ac (SUBDIR_MAKEFILES): Add admin/charsets/Makefile. (admin/charsets/Makefile): Generate it. * admin/charsets/Makefile.in: Rename from Makefile. (AWK, srcdir, top_srcdir, AM_DEFAULT_VERBOSITY): New variables, set by configure. (charsetdir, lispintdir, mapfiledir, AM_V_GEN, am__v_GEN_) (am__v_GEN_0, am__v_GEN_1, AM_V_at, am__v_at_, am__v_at_0) (am__v_at_1, LOCAL, mapconv, run_mapconv, big5, compact, cp51932) (cp932, eucjp_ms, gb180302, gb180304, kuten): New variables. (TRANS_TABLE, CHARSETS): Add directory prefix to value. (all): Declare PHONY. (local): New PHONY target. (map_template): New template. Use to define short PHONY aliases. (*.map): Add directory prefixes to targets and prerequisites. Respect make verbosity. (JISC6226.map): Replace non-portable sed append without newline. (install): Remove rule. (clean): Only delete temporary sedscript. (bootstrap-clean, distclean, maintainer-clean, extraclean) (totalclean): New PHONY rules. * admin/charsets/mapconv (BASE): Replace basename with expr. (FILE): Add "mapfiles" subdirectory. (AWK): New variable. Use throughout in place of "awk". (main): Use "gunzip -c" in place of "zcat". Don't leave whitespace before "p", for older sed. * admin/charsets/mapfiles/PTCP154: Add final newline, to make older sed versions happy. ; * .gitignore: Ignore admin/charsets/Makefile.
Diffstat (limited to 'admin/charsets/mapconv')
-rwxr-xr-xadmin/charsets/mapconv28
1 files changed, 16 insertions, 12 deletions
diff --git a/admin/charsets/mapconv b/admin/charsets/mapconv
index 8433d222b8d..6fd13c61c4a 100755
--- a/admin/charsets/mapconv
+++ b/admin/charsets/mapconv
@@ -1,5 +1,7 @@
#!/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
@@ -33,8 +35,10 @@
# GLIBC-1 GLIBC-2 GLIBC-2-7 CZYBORRA IANA UNICODE UNICODE2 YASUOKA
# $4: awk script
-FILE="admin/charsets/$1"
-BASE=`basename $1 .gz`
+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*)
@@ -69,7 +73,7 @@ fi
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
@@ -81,20 +85,20 @@ fi
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,' \
@@ -102,13 +106,13 @@ elif [ "$3" = "GLIBC-2-7" ] ; then
elif [ "$3" = "CZYBORRA" ] ; then
# Source format is:
# =XX U+YYYY
- sed -n -e "$2 p" < $1 \
+ sed -n -e "${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
- sed -n -e "$2 p" < $1 \
+ sed -n -e "${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
@@ -116,25 +120,25 @@ 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).
- sed -n -e "$2 p" < $1 \
+ sed -n -e "${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 # ...
- sed -n -e "$2 p" < $1 \
+ sed -n -e "${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)
- sed -n -e "$2 p" < $1 \
+ sed -n -e "${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 .....
- sed -n -e "$2 p" < $1 \
+ sed -n -e "${2}p" < $1 \
| sed -e 's/...\(....\) U+\([0-9A-F]*\).*/0x\1 0x\2/' \
| sort | ${AWKPROG}
else