Xref: utzoo news.software.b:1671 news.admin:3697
Path: utzoo!utgpu!water!watmath!clyde!att!osu-cis!tut.cis.ohio-state.edu!bloom-beacon!mit-eddie!rutgers!bellcore!tness7!ninja!gdb
From: gdb@ninja.UUCP (David Butler)
Newsgroups: news.software.b,news.admin
Subject: Re: Here is my version of the microsoft fix
Keywords: white-space message-id header.c microsoft
Message-ID: <14916@ninja.UUCP>
Date: 13 Oct 88 15:57:33 GMT
References: <14849@ninja.UUCP>
Reply-To: gdb@ninja.UUCP (David Butler)
Organization: db Systems, Fort Worth, Texas
Lines: 43

Im sorry about this but I made a stupid mistake in my code.
I didn't catch spaces!!! :-) here is the line:

	if (*p2 < ' ' && *++p2 == '\0') {

It should be:

	if (*p2 <= ' ' && *++p2 == '\0') {
                 ^-- here is that bug

Please correct this line.  Also you may want to remove the #ifdef. Thanks.

Here is the whole correction.  In header.c about line 201 find:

		case MESSAGEID:
			getfield(hp->ident, sizeof(hp->ident));
			break;

Change it to:

		case MESSAGEID:
			getfield(hp->ident, sizeof(hp->ident));
			{
				register char *p1, *p2;
				
				for (p1 = p2 = hp->ident;
				    *p2 != '\0'; ++p1, ++p2) {
					if (*p2 <= ' ' && *++p2 == '\0') {
						break;
					}
					if (p1 != p2) {
						*p1 = *p2;
					}
				}
				if (p1 != p2) {
					*p1 = *p2;
				}
			}
			break;

====================|==========================================================
G. David Butler II  | "There is nothing good or bad, but thinking makes it so."
gdb@ninja.UUCP      | -Shakespeare