savannah-hackers
[Top][All Lists]
Advanced

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

Re: address@hidden: Re: [Jeff Bailey <address@hidden>] Re: [Savannah-hac


From: Joel N. Weber II
Subject: Re: address@hidden: Re: [Jeff Bailey <address@hidden>] Re: [Savannah-hackers] savannah.gnu.org: submission of GNU and FSF Organizational Files]
Date: Sun, 11 Mar 2001 15:55:21 -0500

   I would very much like gnuorg to go in CVS, but I haven't the time to make
   the Makefile DTRT.  I hope someone can work on this, because I'd hate to
   have the import delayed waiting for someone to work on that.

   (BTW, I do agree with devnull that issue needs to be handled, and that it's
    a good idea to wait a week no matter what.)

   Here's an idea for the makefile:

      Have it still run in /gd/gnuorg, and add all it's derived objects to
      .cvsignore.

   Would that solution work?

Sigh.  Is it the case that the people who think they have the time to
put this into CVS don't have the time to actually read
/gd/gnuorg/Makefile, or am I just confused?

Anyway, most of what /gd/gnuorg/Makefile does is to help certain files
find their way onto the web and ftp servers.  (There's also something
that makes a tarball that just gets stored in /gd/gnuorg, which is
refered to probably by maintain.* but I'm not convinced that anyone
actually ever retrieves that file, and there's also a cron job to
create tasks-cod-std, which I suspect gvc has never sent out in the
last five years due to the current preferences of the gvc folk.  I
suspect that we could delete those generated files, and comment them
out in the Makefile, and do something about them in the unlikely event
that anyone ever complains.)  If we're going to be using CVS, it might
not be a bad idea to just do CVS checkouts on the ftp and web
servers.

We could decide to not check in any of the generated files, and run
the Makefile on ftp and www, with some sort of logic so that if the
generated files happen to have had syntax errors, mail will be sent
somewhere appropriate, and the old version will remain in place.  This
is acceptable as long as we figure we can keep tex and texinfo
properly installed on ftp and www.

Otherwise, we should give some machine write bits on the repository,
and let it check in the generated files, perhaps.



reply via email to

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