Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!utgpu!water!watnot!watmath!clyde!rutgers!brl-adm!adm!Schauble@mit-multics.arpa
From: Schauble@mit-multics.arpa
Newsgroups: comp.lang.c
Subject: More on Microsoft C problem
Message-ID: <6375@brl-adm.ARPA>
Date: Mon, 30-Mar-87 04:26:19 EST
Article-I.D.: brl-adm.6375
Posted: Mon Mar 30 04:26:19 1987
Date-Received: Tue, 31-Mar-87 04:19:55 EST
Sender: news@brl-adm.ARPA
Lines: 58

This is a followup on my recent message about error checking in
Microsoft C V4.0.  The problem in that message was:

Within one file I have the function declaration

    void bgunlink_transaction_record (bgfile, client, transaction)
    BGFILE *bgfile;
    struct tclient_rec *client;
    struct ttransaction_rec *transaction;
    {
    ....
    }

The type BGFILE and the structures are correctly declared prior to this.
Later in the file I have the statement

          bgunlink_transaction_record (&bgfile, &client_rec, rec, &transaction);

Note that they don't match in number or types of arguments.  When
compiling with /W3 for maximum checking, the compiler SAYS NOT A WORD
ABOUT THIS.

On thinking about it more, this seemed even stranger.  The compiler does
process function declarations to get the return type.  That is, the
program fragment:

    int f(i,p)
    int i;
    char *p;
    {}

    void main()
    {
    int i;
    char *p, *z;
    z = f(p,5);
    }

produces a warning about the assignment to z (int assigned to char *)
but not about the argument mismatch.

Adding the line
    int f(char *, int *); causes a warning about the assignment, but
again not about the argument mismatch.  I believe that this is a
violation of the proposed ANSI standard in that this redefinition of f
is not allowed.  Would someone who has a copy of the standard please
check.

And finally, most curiously of all, deleting the function definition and
leaving only the prototype now produces the warning about argument
mismatch.

Note that the arguments were apparantly not checked even when the
prototype was present if the function was present.  Having the function
itself in the same compilation should certainly not suppress argument
checking.  There's something wrong here.

    Paul .