[Dbix-class] Re: SQL::Translator conversion from Class::Base to Moo

[Peter Rabbitson]

On Wed, Aug 29, 2012 at 10:53:12AM +0100, Dagfinn Ilmari Mannsåker wrote:
> Peter Rabbitson <rabbit+dbic at rabbit.us> writes:
> 
> > On Wed, Aug 29, 2012 at 01:15:50AM +0100, Dagfinn Ilmari Mannsåker wrote:
> >> ilmari at ilmari.org (Dagfinn Ilmari Mannsåker) writes:
> >> 
> >> > The code passes the test suites of DBIx::Class (with
> >> > DBICTEST_SQLT_DEPLOY=1), ::DeploymentHandler and ::Migration, but I
> >> > would like some testing and review by actual users before merging it to
> >> > trunk.
> >> 
> >> Thanks to frew and jnap for reviewing and testing this.  Unless anyone
> >> objects I'll merge it in the next few days so we can get a dev release
> >> out for smoking.
> >> 
> >
> > A cursory test of non-DBIC based SQLT dependents[1] turned up
> > Metabase::Backend::SQLite[2] with a bunch of readonly accessors being called
> > as setters. Now, Metabase::Backend::SQLite fails one test on currently
> > released SQLT anyway (please bug xdg), but nevertheless this is might be
> > an indication that you were too eager on applying 'ro'.
> 
> I only applied 'ro' to the already read-only accessors.  The difference
> is that the old ones didn't actually check if they were being mistakenly
> called with an argument.  I guess I'll have to apply an 'around' that
> ignores arguments.  *grmble*
> 

An around that warns makes sense then, yes.