[EMAIL PROTECTED] wrote:
> This is really no big problem. The nice fact is that printf() is well defined
> and it really doesn't matter if the actual implementation differs in minor
> details from the official definition.
Well, obviously it does or we wouldn't have this thread.
> Then you don't see the problem with replacing stdio in general.
I think we do see it, and think it's a bad idea. If you called your
version something else, and avoided name clashes, this issue wouldn't
come up.
> There are global variables and hidden implementation details (specially
> related to MT safe code) that may be used directly from libc internal functions
> and cause problems. While you may solve these problems for a single OS, it
> is far too much complexity if you like to do it in general.
>
> If I add my own printf() there is no problem with global variables and no
> problem buffering of stdout ... also no problem with hidden internals.
So far I agree completely, you have a good reason to write your own, you
did, it works well, all good things.
> It makes no problem to have two printf() implementations at the same time.
> They will just cooperate fine.
That is less clear. It's obviously more work to keep them from stepping
on each other, and I don't see what you gain by having the same name as
the system routine (feel free to show something I miss). I don't doubt
you can do this, but I would think you wouldn't bother.
--
-bill davidsen ([EMAIL PROTECTED])
"The secret to procrastination is to put things off until the
last possible moment - but no longer" -me
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
