[Development] Adding support for version number comparisons

Oswald Buddenhagen oswald.buddenhagen at digia.com
Tue May 13 17:47:59 CEST 2014

On Tue, May 13, 2014 at 07:43:18AM -0700, Thiago Macieira wrote:
> Em ter 13 maio 2014, às 11:58:15, Oswald Buddenhagen escreveu:
> > what is the added value of hard-coding arbitrary policies (and
> > thereby restricting possible use cases) instead of providing a
> > minimalistic solution (or two, one for semver and one for strings) and
> > putting a few recipes for common schemes into the documentation?
> What's the point of providing a minimalistic solution for cookies which just 
> respects the original Netscape spec and the RFC? 
> Both have the same answer: it depends on how well it deals with the real 
> world.
the analogy is entirely bogus. the thresholds for usefulness and the
user's ability to manipulate the input into something the qt code can
work with are entirely different.

> > policy in the classes was always considered a very un-qt thing. the one
> > counterexample i can come up with is qlibraryinfo, and it is in fact a
> > constant pita.
> QLibraryInfo is not about policy, either. It's just reporting what Qt's 
> settings are.
it's reporting our installation directory layout policy. unfortunately,
that policy (which is basically FHS) doesn't match the ideal layout of
mac os bundles at all.
also, the implementation sucks. but that's a different discussion.

More information about the Development mailing list