httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Marc Slemko <>
Subject Re: cvs commit: apache-site bugdb.cgi
Date Tue, 28 Jul 1998 16:57:54 GMT
On Tue, 28 Jul 1998, Rodent of Unusual Size wrote:

> Marc Slemko wrote:
> > 
> > I can say the exact same thing for anything distributed with Apache, and
> > the answer is the same: check the CVS logs and use diff.  Unless you are
> > hiding something from me, everyone has trivial access to the CVS history
> > for any file in multiple ways.
> bugdb.cgi and aren't distributed with Apache - they live only
> in apache-site.  I suppose interested parties can check the webcvs stuff;

Or anoncvs, or probably cvsup, etc.  Just like the rest of the source.

> that's a good point.  But what's the harm in the inline comments?  It's
> only one file.  In the environments where I've worked, it's appropriate
> to put mod histories both into the source and the CM logs - you never
> know when a snapshot is going to be taken or the CM system decommissioned
> or broken.

The harm is that it is only one file, it is completely different than
everything else that is done for no good resaon, and it looks a bit silly
to add more to the file in changelog than the tiny little change that was

If you are truly worried about CVS magically going away over night, then
you have a lot bigger problems that adding a change history to one file
isn't going to solve.  

> The CVS logs only help if the log messages actually make sense; there
> have been times when that's been problematical - such as blank lines
> or that silliness earlier this year with weather reports.  The latter
> was a special case, though. :-)

Erm... if there is nothing sensible to put as a log message then there is
no harm in not putting a useful message.  No one had better try such a
thing with a code change.

That is a bogus excuse.  "I need to enter change history in the file
because I can make bogus log messages when I commit."

View raw message