httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ryan Bloom <...@ntrnet.net>
Subject Re: Tagging .31 soon
Date Thu, 24 Jan 2002 00:29:28 GMT

BTW, this stream of messages is one of the reasons that we moved to the
new tagging scheme.  People should feel free to tag when they believe that
the code is stable.  If you announce your intent to tag on the list, you
are encouraging people to commit code at the last minute, just to get it
in the new release, and thus destabilize the code.

If you are going to make a comment like "I will tag in 24 hours, assuming
nothing destabilizes the tree", don't make the comment, just tag the
tree.  You can send a message saying "I just tagged, but I have no time to
finish the process, I'll release tarballs tomorrow."

Ryan

On Wed, 23 Jan 2002, Bill Stoddard wrote:

> FWIW, I am not opposed to Ian tagging the 2.0 tree for his purposes.
> 
> Bill
> 
> ----- Original Message ----- 
> From: "William A. Rowe, Jr." <wrowe@covalent.net>
> To: <dev@httpd.apache.org>
> Sent: Wednesday, January 23, 2002 1:21 PM
> Subject: Re: Tagging .31 soon
> 
> 
> > From: "Justin Erenkrantz" <jerenkrantz@ebuilt.com>
> > Sent: Wednesday, January 23, 2002 11:29 AM
> > 
> > 
> > > On Wed, Jan 23, 2002 at 09:08:48AM -0800, Ian Holsman wrote:
> > > > probably tomorrow if no major problems/objections occur.
> > > 
> > > Can we please wait until Monday?  I'd like to do the following
> > > things:
> > > 
> > > - Remove mod_auth_db
> > > - Check in ap_rgetline rewrite (Ryan Morgan's had the only comment so
> > >   far)
> > > - Change the ap_get_brigade prototype once more to remove *readbytes
> > 
> > From: "Ian Holsman" <ianh@apache.org>
> > Sent: Wednesday, January 23, 2002 11:50 AM
> > %
> > 
> > % no offense justin,
> > % but I'd like to Tag BEFORE these patches go in. they have the possibilty
> > % of de-stablising the entire tree, which would could shoot the friday/monday.
> > 
> > I very much disagree.  DO NOT FOIST AN API WE INTEND TO CHANGE ONTO 
> > DEVELOPERS.  If we are going to break the API, we break it.  But no fair 
> > releasing when we know _exactly_ what the changes are.
> > 
> > > I also know Aaron is intending to look at the shmem code in the 
> > > next few days to resolve his hack he checked in last night.
> > 
> > And I have about three things or so to drop into .31
> > 
> >  - fix Win32 scoreboard fault [unnamed malloc-style fallback case]
> >  - fix Win32 anon shm allocation [converse of Aarons' headaches :-]
> >  - fix apr_dup2's prototype [goes to what I screamed above]
> > 
> > Of course, fix the "Win32 is broke" bug
> > 
> > > FWIW, if you don't intend on releasing .31 - which *I* won't
> > > vote for without these changes, then you can simply tag it as
> > > iholsman_cool_tag or some other nonsense.  Apparently, rumour
> > > has it that this is what we used to do in the 1.3 series.  -- justin
> > 
> > No, I'll veto release on the points we raised above - wait for Monday
> > and save the tag.
> > 
> 

-- 

_______________________________________________________________________________
Ryan Bloom                        	rbb@apache.org
550 Jean St
Oakland CA 94610
-------------------------------------------------------------------------------


Mime
View raw message