next up previous contents
Next: James Clough Up: Beta Correspondence Previous: Jim Alves-Foss

Gerald Atkinson

From gatkins@cs.uidaho.edu Fri Nov 22 19:16:12 PST 1996
Received: from waldrog.cs.uidaho.edu (waldrog.cs.uidaho.edu [129.101.100.23])
by dworshak.cs.uidaho.edu (8.7.5/1.1) with ESMTP id TAA14924; 
Fri, 22 Nov 1996 19:14:05 -0800 (PST)
Received: from localhost (gatkins@localhost) 
by waldrog.cs.uidaho.edu (8.7.5/1.0) with SMTP id TAA04321; 
Fri, 22 Nov 1996 19:14:04 -0800 (PST)
X-Authentication-Warning: waldrog.cs.uidaho.edu: 
gatkins owned process doing -bs
Date: Fri, 22 Nov 1996 19:14:04 -0800 (PST)
From: Gerald Atkinson <gatkins@cs.uidaho.edu>
To: jone9333@cs.uidaho.edu
cc: oman@gold.cs.uidaho.edu, tpearse@boi.hp.com
Subject: beta
Message-ID: <Pine.HPP.3.95.961122185512.1354E-100000@waldrog.cs.uidaho.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Status: RO


Here are the beta results that I've had so far:

I haven't had as much time as I would have liked, but hope to do a little
more before 3 DEC.

I'm using the default perl on the CS machines, 5.002 ? 

In general it seems correct and robust, I really like the graphical
output, especially the scalability of the output.  The "summary data" is
straight forward and useful.



I'm not sure all of the following observations are necessarily errors, but
they were unexpected by me:



------------------------ your readme file:

An example of this is:

 ./pmerge /usr/local/bin/perl cca   this does not work, but the 
				    following does: 

				    ./pmerge.pl /usr/local/bin/perl cca

				    so the directions prior to this are
				    correct


this was the only problem noticed with your readme file or with
installation.



------------------------ man pages:

seem to be generally helpful, simple, and correct....

... however I'm not crazy about the "usage statement," it takes a long
time to decipher the first time.



------------------------ basic tests:

blank lines are counted as lines of code???



------------------------ more basic tests:

the following 3 code fragments generate different output,
is that your intent???


#ifdef A 
  statement; 			shows 1 LOC in the graphical output
#endif



#ifdef A  statement; 		shows 0 LOC in the graphical output
#endif



#ifdef A  statement;  #endif	causes: Fatal Error - Missing #endif



------------------------ long file name:

handled file names in excess of 128 characters without a problem



------------------------- tile test:

handled horizontal and vertical tiling fine, though I noticed that 
during this test I ended up with A BLANK PAGE AT THE END of the 
file???



------------------------- vg2:

treats ALL of the following as acceptable vg2 indicators:


	#ifdef A && B

	#ifdef A &% B

	#ifdef A %& B

	#ifdef A & B

	#ifdef A || B

	#ifdef A | B

	#ifdef A %| B

        #ifdef A |% B

	(seems to accept any character in place of '%' above)

is this how it is supposed to behave???



------------------------- end


Sean, could you pass this on to the others, please?


Happy Holiday to all,

Jerry

***************************************************************************

From gatkins@cs.uidaho.edu Mon Dec  2 17:43:26 PST 1996
Received: from waldrog.cs.uidaho.edu (waldrog.cs.uidaho.edu [129.101.100.23])
by dworshak.cs.uidaho.edu (8.7.5/1.1) with ESMTP id RAA20919; 
Mon, 2 Dec 1996 17:42:07 -0800 (PST)
Received: from localhost (gatkins@localhost) 
by waldrog.cs.uidaho.edu (8.7.5/1.0) with SMTP id RAA15547; 
Mon, 2 Dec 1996 17:42:06 -0800 (PST)
X-Authentication-Warning: waldrog.cs.uidaho.edu: 
gatkins owned process doing -bs
Date: Mon, 2 Dec 1996 17:42:06 -0800 (PST)
From: Gerald Atkinson <gatkins@cs.uidaho.edu>
To: jone9333@cs.uidaho.edu
cc: oman@gold.cs.uidaho.edu, tpearse@boi.hp.com
Message-ID: <Pine.HPP.3.95.961202173516.13214H-100000@waldrog.cs.uidaho.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Status: RO


Team B:

Another beta observation:


Don't know if you've considered/found this, I ran across a bit about
"trigraphs" in K&R's "C: The Programming Language."

Apparently, where ascii is not supported, not a full character set, you
can substitued 

	??= 

for

	#

then compiling with -trigraphs or -ansi, the '??=' is replaced by '#'.


Anyway, code like this compiles, but cca does not seem to pick up on it by
noting a change in the V(g).

I think I am doing this correctly.

Don't know how often (legacy) code might use this....

Anyone ever heard of this?


Jerry