For the record, I was correct in the code. Hey, I do have a reputation, albeit
not yet much of one!
Brian Fundakowski Feldman _ __ ___ ____ ___ ___ ___
[EMAIL PROTECTED] _ __ ___ | _ ) __| \
FreeBSD: The Power to Serve! _ __ | _ \._ \ |) |
http://www.FreeBSD.org/ _ |___/___/___/
---------- Forwarded message ----------
Date: Sat, 3 Jul 1999 12:37:15 +0930
From: Greg Lehey <[EMAIL PROTECTED]>
To: Brian F. Feldman <[EMAIL PROTECTED]>
Subject: Re: Fixing other people's code (was: world broken in vinum (PATCH))
On Friday, 2 July 1999 at 23:06:17 -0400, Brian F. Feldman wrote:
> On Sat, 3 Jul 1999, Greg Lehey wrote:
>
>> On Friday, 2 July 1999 at 22:35:16 -0400, Brian F. Feldman wrote:
>>> On Sat, 3 Jul 1999, Greg Lehey wrote:
>>>
>>>> On Saturday, 3 July 1999 at 0:09:53 +0000, Ben Rosengart wrote:
>>>>> On Sat, 3 Jul 1999, Greg Lehey wrote:
>>>>>
>>>>> Was the fix that wasn't yours correct?
>>>>
>>>> No.
>>>>
>>>>> 'Cause I'd rather have code that doesn't compile than code that
>>>>> compiles but is subtly wrong.
>>>>
>>>> Well, you can now have code that compiles and is correct :-)
>>>>
>>>> Seriously, the only problem with Brian's fix was that it added a new
>>>> variable. It would have made the stats list functions not work. I
>>>
>>> Out of curiousity, what 'exactly' was wrong?
>>
>> I had a flag called 'stats', which was set by the -s flag. When I
>> introduced the 'mirror' command, I used the -s flag to mean "create a
>> striped plex". Obviously, the name 'stats' didn't fit this function,
>> so I changed it to 'sflag'. Your fix introduced an additional flag,
>> which wouldn't have been used. Hmmm, come to think of it, it might
>> have worked anyway... we did have the definition after all. OK, I
>> thought it wouldn't work.
>
> Hey, I did what I could do best: try to understand and fix it quickly!
> I _was_ pretty sure it was right tho :)
Right. I didn't check it, since I had the tested version waiting (all
night) to be checked in.
>> The changes for min() weren't needed; vext.h had a definition.
>
> I guessed that; however, you didn't use max() in the surrounding code, you
> used a broken-out form of what max() is usually defined to (or MAX() for
> that matter.)
Why be consistent when you can be confusing? Yes, maybe I should
reconsider that one.
Greg
--
See complete headers for address, home page and phone numbers
finger [EMAIL PROTECTED] for PGP public key
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message