[Mimedefang] putc failed: Error 0 line 848

[Alberto.daSilva at rmb.co.za]

On 2002/11/14 11:57 AM, Steffen Kaiser wrote:

>On Thu, 14 Nov 2002, Rudolph Pereira wrote:
>
>On Wed, Nov 13, 2002 at 11:21:08AM -0500, David F. Skoll wrote:
>> On Wed, 13 Nov 2002 Alberto.daSilva at rmb.co.za wrote:
>>
>> > The problem is when *s = 0xff.
>> > putc returns -1 == EOF
>>
>> Then Solaris is broken, or gcc 3.2 on Solaris is.
>
>fwiw, running the included test program on sparc/solaris 2.8 machines
>compiled under gcc 3.1 and 3.2 produce the right results for me.
>
>For me as well. Otherwise I would have expected my test program (see mail
>from Nov 12) to fail as well.
>
>I've replaced the putc() by fputc() -- well, it shouldn't matter, but
>let's see -- no log entry since yesterday.

I've also tried the test program with gcc 3.2, and even some variations,
all worked correctly.

I got hold of a message that was failing and ran it thru a modified test.c,
still worked OK.

Replaced putc() with fputc() in mimedefang.c - that seems to work.
Test message went thru with no failures and no log entry since the
change to fputc().

I guess there must be something going wrong with the macro expansion of
putc().

Alberto