X-Authentication-Warning: delorie.com: mail set sender to djgpp-bounces using -f
From: "MikeC" <My_address AT end DOT of DOT post>
Newsgroups: comp.os.msdos.djgpp
References:  <B1EIi.52071$rr5 DOT 11613 AT newsfe1-win DOT ntli DOT net> <200709210020 DOT l8L0KG9G026658 AT envy DOT delorie DOT com>
Subject: Re: How to copy a file?
Lines: 35
X-Priority: 3
X-MSMail-Priority: Normal
X-Newsreader: Microsoft Outlook Express 6.00.2900.3138
X-RFC2646: Format=Flowed; Original
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.3138
Message-ID: <fhEIi.27$Qd2.7@newsfe6-win.ntli.net>
Date: Fri, 21 Sep 2007 00:31:07 GMT
NNTP-Posting-Host: 86.13.147.52
X-Complaints-To: http://netreport.virginmedia.com
X-Trace: newsfe6-win.ntli.net 1190334667 86.13.147.52 (Fri, 21 Sep 2007 01:31:07 BST)
NNTP-Posting-Date: Fri, 21 Sep 2007 01:31:07 BST
Organization: ntl Cablemodem News Service
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Reply-To: djgpp AT delorie DOT com

Thanks DJ.  Appreciated.

Is my function losing a lot of time with the fopen/fread/fwrite/fclose? 
Would that account for almost twice the time for the copy, when using a 64K 
buffer?

Another possibility, I suppose, is to delve into a DOS book and try it 
through a DOS interrupt.

Why can't I do it with ...

system(string);  /* string contains "copy <source path> <dest path>" */

... as I tried?  It compiles OK, and appears to execute when I single-step 
the program under RHIDE, but it takes an instant - not a minute or more - 
and the file doesn't get copied.  I don't know what I'm doing wrong.

Mike.

"DJ Delorie" <dj AT delorie DOT com> wrote in message 
news:200709210020 DOT l8L0KG9G026658 AT envy DOT delorie DOT com...
>
> I think you're doing it the right way, just keep in mind that DJGPP
> programs have to move all that data back and forth between DOS memory
> and DPMI memory.
>
> You might want to use open/read/write/close instead of
> fopen/fread/fwrite/fclose though.
>
> Or, even faster, _read and _write, although then you have to deal with
> all the odd DOS border cases.  That does bypass the CRLF code, even
> though you've disabled it there's still a check to see if it's enabled
> for each read/write.