Path: utzoo!utgpu!jarvis.csri.toronto.edu!clyde.concordia.ca!uunet!aplcen!uakari.primate.wisc.edu!brutus.cs.uiuc.edu!apple!voder!zok!wattres!steve
From: steve@wattres.UUCP (Steve Watt)
Newsgroups: comp.lang.c
Subject: Re: Novic C programmer needs help
Message-ID: <420@wattres.UUCP>
Date: 13 Feb 90 08:36:06 GMT
References: <1153@psc90.UUCP+ <1990Feb12.201406.215@wam.umd.edu>
Reply-To: steve@wattres.UUCP (Steve Watt)
Organization: Steven Watt, Consultant
Lines: 51

In article <1990Feb12.201406.215@wam.umd.edu+ dkim@wam.umd.edu (Daeshik Kim) writes:
+ In article <1153@psc90.UUCP> pyr201@.UUCP (d^2) writes:
++ char *form (va_alist)
  [ strange header ]
++  FILE b;
+
+	Why isn't this "File *b;" ?

  I believe this is because the FILE structure is being used to fake out the
putchar's that are called in doprnt.
+
++  va_start (pvar);
++  fmt_string = va_arg (pvar, char*);
++
++  b._flag = _IOWRT|_IOSTRG;
++  b._ptr = result;
++  b._cnt = LINELEN;
++
++--->  _doprnt(fmt_string, pvar, &b); <---
+	File b seems to be opended here.

File b is actually "opened" where b._flag =...  happens.  The thing that this
code is interested in is that b._ptr = result, the buffer to put the result
from doprnt into.

++  putc('\0', &b);
				/* Put a null terminator on result */
++
++  va_end (pvar);
++  return (result);
++}
+
+	Where File b is closed?
+	What is this file?
+	I have not seen the whole program but either '_doprnt()' does
+	something wrong or file 'b' better to be closed with 'fclose()'.
+	Just an opinion!
+--
+	Daeshik Kim	H: (301) 445-0475/2147 O: (703) 689-5878
+	SCHOOL:	dkim@wam.umd.edu, dskim@eng.umd.edu, mz518@umd5.umd.edu
+	WORK:	dkim@daffy.uu.net (uunet!daffy!dkim)

#if defined(FLAME) && FLAME >= 0

The entire piece of code, however, is *awful*!!!!!!  If there is a varargs
package around, there tends to be a vsprintf around!  vsprintf is the very
correct way of doing this sort of thing.  Depending on having _doprnt around
is NOT NICE!

#endif

--
Steve Watt
...!claris!wattres!steve        wattres!steve@claris.com is reported to work.
Is rm -rf a way to remove high-frequence noise?  (No names mentioned!  :-)