NOS 1.8d again

[Murray_RJ@cc.curtin.edu.au]

> Off the top of my head (an Americanism), look for improper use of stream
> positioning for that funny mbox_to() parsing.  When dealing with text
> mode streams (the reason for your problem with '\x1A', by the way), you
> should not use fseek() except on a value returned by ftell().  This is
> a frequently violated rule, and can cause exactly what you are seeing
> on a file that is open for both reading and writing.

   I must admit that I didn't know about this behaviour of fseek(), but I'll
keep it in mind. I, too, figured that the problem involved text-mode (hence
the $1A), but I thought it was more complex when I got one which seemed to
repeat half of the BID (i.e. something like 2345_v12345_vk6qqq instead of
just 12345_vk6qqq). It turns out that this isn't really a problem: he was
entering the "SB..." line manually into my mailbox, and edited the line
(presumably with left cursor rather than backspace) while he typed it. So
this example probably isn't the fault.

   Since neither I nor anybody else has experienced the $1A problem, I begin
to think it may not be a problem in NOS, but rather external to it. He has
apparently had bulletins forwarded to him which have a $1A near the front
of them (which apparently drives the mailbox crazy. Or maybe View; I forget
which), but the same bulletins forwarded to me from the same BBS were
perfectly ok. He's running a PC-100 board, so perhaps it or the SCC driver do
strange things now and then (although it's hard to see how). If it happens
again I'll add a 1A checker to his copy of NOS.

 .....Ron
                                 ***
 Ron Murray (vk6zjm)
 Internet: Murray_RJ@cc.curtin.edu.au
     Are we having fun yet?    -- Garfield