bug-inetutils
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: [bug-inetutils] Generated files in CVS?


From: Alain Magloire
Subject: RE: [bug-inetutils] Generated files in CVS?
Date: Fri, 20 Jan 2006 09:26:37 -0500


> -----Original Message-----
> From: Alfred M. Szmidt [mailto:address@hidden
> Sent: Sunday, January 15, 2006 5:49 PM
> To: Sergey Poznyakoff
> Cc: address@hidden; address@hidden
> Subject: Re: [bug-inetutils] Generated files in CVS?
> 
>    Thanks for the tip, this is really handy. However, unbelievable as
>    it is, there are people that don't use Emacs :) As far as I know,
>    Alain is one of them.
> 
> Then we should get Alain to switch. ;-)
>

NEVER !!!
But then again, I say never to not have an SUV (A big monstrous car feeding
on gaz like crazy), to not live in the suburban, to no watch TV, to not eat
at McDonalds, to exercise everyday etc ..
 
>    Besides, even if unwanted noise is easy to delete, it is still
>    produced. And the time and bandwidth needed to obtain the diff
>    should not, in my opinion, be ignored.
> 
> I think the concern about bandwidth is moot.  gnulib takes up more
> bandwidth than any regenerated file would ever over a 2 year period.
> 
>    The problem is not that these are rarely generated. The problem is
>    that it will suffice for one of us to use slightly different
>    versions of autotools (and this happens quite often) to make his
>    copies of Makefile.in's and/or configure differ from those in the
>    repository, and, consequently to obtain superfluous output from cvs
>    diff.
> 
> I was going to say that this was a really valid point, but it occured
> to me that this won't happen in practise.  The only time
> configure/Makefile.in is regenerated if we commit them to the CVS
> tree, is when one modifes configure.ac or Makefile.am.  And never
> during a normal run.
> 
> And if you do modify a Makefile.am (or configure.ac), then only that
> file will be regenerated by automake/autoconf anyway, so you won't end
> up with a dozen new Makefile.in's that need to be commited.

I think, I'm leaning toward Sergey 8-).  I'm not really convince with your
arguments, especially when regenerating the file is a one liner
"./autogen.sh".  And secondly, I have this bad habit of changing the
Makefile{,.in} for a quick test followed by a "cvs commit ." ... which is a
world of trouble if I do not pay attention.
 




reply via email to

[Prev in Thread] Current Thread [Next in Thread]