dictfmt_index2suffix.in

#!/bin/sh

#
# By Aleksey Cheusov <vle@gmx.net>
#

usage (){
   printf "\
Converts .index file from DICTD database to the index file .suffix\n\
usage: dictfmt_index2suffix [OPTIONS] [files...]\n\
OPTIONS:\n\
  --help    display this screen\n\
  --utf8    for creating utf8 .index file\n\
  all other -X and --XXX options are passed to dictfmt -I\n\
"
}

LC_ALL=C
export LC_ALL

arg_locale=C

# Processing arguments
while [ $# != 0 ]; do
    case $1 in
	--help)
	    usage
	    exit 0;;
	--utf8)
	    utf8_mode=1
	    args="$args $1";;
	-*)
	    args="$args $1";;
	*)
	    break;;
    esac
    shift
done

if test $BASH; then
	exit_="echo \${PIPESTATUS[@]} | egrep '^0( 0)*$' >/dev/null"
else
	exit_='exit $?'
fi

@AWK@ -v "utf8_mode=$utf8_mode" '
function charlen_utf8 (str){
	if (str == ""){
		return 0
	}else if (str ~ /^[\1-\177]/){
		return 1
	}else if (str ~ /^[\200-\277]/){
		return -1
	}else if (str ~ /^[\300-\337]/){
		return 2
	}else if (str ~ /^[\340-\357]/){
		return 3
	}else if (str ~ /^[\360-\367]/){
		return 4
	}else if (str ~ /^[\370-\373]/){
		return 5
	}else if (str ~ /^[\374-\375]/){
		return 6
	}else{
		return -1;
	}
}

BEGIN {
	FS = OFS = "\t"
}

{
	if (!utf8_mode){
		for (i = length($1); i >= 1; --i){
			printf "%s", substr($1, i, 1)
		}
	}else{
		i   = 1
		idx = 1
		while (i < length($1)){
			rest = substr($1, i)
			char_len = charlen_utf8(rest)
			if (char_len < 0){
				print "invalid UTF-8 input: `" rest "`" > "/dev/stderr"
				exit
			}
			inverse_char [idx] = substr($1, i, char_len)
			i += char_len
			++idx
		}
		while (idx--){
			printf "%s", inverse_char [idx]
		}
	}

	$1 = ""
	print $0
}' "$@" | dictfmt -I $args | uniq

eval $exit_