dxerr9: Use wide character string literals.

dlls/dxerr9/errors.awk

-BEGIN {
-	print "/* Machine generated. Do not edit. */"
-	print ""
-	lines = 0
-}
-{ 
-	split($0, array, FS)
-
-	if (NF > 0 && length(array[1]) > 0) {
-		lines++
-
-		# save the first word (or '&' separated list of words) in the names array
-		if (array[2] == "&") {
-			if (array[4] == "&") {
-				names[lines] = array[1] " " array[2] " " array[3] " " array[4] " " array[5]
-				start =  6
-			} else {
-				names[lines] = array[1] " " array[2] " " array[3]
-				start = 4 
-			}
-		} else {
-			names[lines] = array[1] 
-			start = 2
-		}
-
-		# create the WCHAR version of the name
-		printf "static const WCHAR name%dW[] = { ", lines
-		i = 1
-		len = length(names[lines]) + 1
-		while (i < len) {
-			printf "'%s',", substr(names[lines],i,1)
-			i++
-		}
-		print  "0 };"
-	
-		# create the CHAR version of the description
-		printf "static const CHAR description%dA[] = \"", lines
-		word = start 
-		while (word < (NF + 1)) {
-			printf "%s", array[word]
-			if (word < NF )
-				printf " "
-			word++
-		}
-		print  "\";"
-	
-		# create the WCHAR version of the description
-		printf "static const WCHAR description%dW[] = { ", lines
-		word = start
-		while (word < (NF + 1)) {
-			i = 1
-			len = length(array[word]) + 1
-			while (i < len) {
-				if (substr(array[word],i,1) == "'")
-					printf "'\\'',"
-				else
-					printf "'%s',", substr(array[word],i,1)
-				i++
-			}
-			if (word < NF )
-				printf "' ',"
-			word++
-		}
-		print  "0 };"
-	}
-}
-END {
-	print ""
-	print "static const error_info info[] = {"
-
-	i = 1 
-	while ( i <= lines) { 
-		split(names[i], words, FS)
-		printf "    { %s, \"%s\", name%dW, description%dA, description%dW },\n", 
-			words[1], names[i], i, i,i 
-		i++
-	}
-
-	print "};"
-}

dlls/dxerr9/make_errors

-#!/bin/sh
-exec awk -f errors.awk < errors.dat > errors.h