Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!utgpu!water!watmath!clyde!rutgers!nysernic!itsgw!batcomputer!cornell!uw-beaver!mit-eddie!ll-xn!ames!ucbcad!ucbvax!CORY.BERKELEY.EDU!dillon
From: dillon@CORY.BERKELEY.EDU.UUCP
Newsgroups: comp.sys.amiga
Subject: Re: Bug in Aztec 3.4b
Message-ID: <8710031936.AA05570@cory.Berkeley.EDU>
Date: Sat, 3-Oct-87 15:36:08 EDT
Article-I.D.: cory.8710031936.AA05570
Posted: Sat Oct  3 15:36:08 1987
Date-Received: Sun, 4-Oct-87 07:19:39 EDT
Sender: usenet@ucbvax.BERKELEY.EDU
Lines: 29

:typedef struct {
:unsigned char *TEST1;
:} OTHER_STRUCT;
:
:typedef struct {
:OTHER_STRUCT TEST2;
:} THIS_STRUCT;
:
:main()
:{
:THIS_STRUCT *TESTC;
:
:    if(TESTC->TEST2)
:	exit(21);
:}

	The conditional is not a valid C construct.  You cannot test a
structure.... it does not matter that the structure contains one element.
Therefore, the only fault that can be attributed to manx is that perhaps 
they do not give an error message.


>/* Don't mind what the program does.  The assember should not be fed
>garbage like tst.r (a0) no matter what */

	True, but the problem is not that the compiler produced bad code,
but that it produced code when it should not have.

				-Matt