New releases for packages in CVS (was: Version Naming)

This is mostly the case if leads are waiting for some other fixes or small new features to be introduced with the next version. However, sometimes a new release is simply forgotten. And some fixes might in the future also be done in CVS by PEAR QA without the package-lead being actually involved (e.g. maybe because he didn't react for a certain time). If you notice such packages, please feel free to send a short email to pear-qa@lists.php.net. We'll take care of those issues, if a new release is urgent or if the package-lead wasn't available for a longer time. Stefan On 27 Feb 2004 at 10:08, Marcos Neves wrote: > I agree. > > I would like to know if exist a rule to bring a cvs version to pear > package. I mean, many packages has bugfixes at cvs but didn't update > at pear. why? > > At 14:04 27/2/2004 +0100, Klaus Guenther wrote: > >From: "Lukas Smith" <smith@backendmedia.com> > > > > > Marcos Neves wrote: > > > > > > > Excelent!!! > > > > What about an expire date to every package be update. > > > > I think one month is enouth. > > > > > > I have heard other people also say that they would prefer to have > > > deadlines by which a given proposal must be applied to a given > > > package. I dont think however that this is really necessary. What > > > do other people think? > > > >One thing that would be rather important would be to determine a > >proper api version method. Iirc, current methods usually rely on a > >float, but it should probably be an array. I think it would be > >helpful to require an apiVersion method for all packages, which > >would, of course, require them to be updated pretty soon. There could > >be a deadline set for that. But I don't think we need a deadline for > >version changes -- current packages still work, even if there are > >benefits to the new system. > > > >That said, it would be a good idea to wait until the handling for > >ex-maintainers has been implemented before forcing new releases, so > >that it can all be done in one swoop.