On Thu, May 08, 2008 at 12:41:19PM +0100, Spiros Denaxas wrote: > On Thu, May 8, 2008 at 12:35 PM, Aristotle Pagaltzis <pagaltzis@gmx.de> wrote: > > > > A Perl::Critic policy has been submitted for this, but it seems > > to me that this is a potential bug worthy of attention from the > > core. Now, I am hesitant on adding new warnings but I think in > > this case it really is worthwhile; it could, after all, be > > disabled easily enough. (What was the consensus on making > > `use warnings` mean something other than `use warnings 'all'`?) > > > > Regards, > > -- > > Aristotle Pagaltzis // <http://plasmasturm.org/> > > > > Hullo, > > I agree that this can potentially save people a fair amount of time when trying > to track down bugs in their code but I am somewhat hesitant of adding > a new warning just for it. Could this potentially be integrated in the > 'diagnostics' > pragma and be explicitly printed only if the pragma is being used? Me too. Two points I like to say. First, a general point. IMO, such issues should be addressed by a lint like program (for instance, Perl::Critic), and not by 'use warnings'. As for this particular case, the number of people that use 'BEGIN:', don't spot it quickly, and actually use 'use warnings' doesn't strike me as high. Doesn't strike me as a worthwhile trade off. OTOH, I doubt there will be much code that would trigger a false warning if such a warning would be added. And any obfuscated code wouldn't use warnings any way. AbigailThread Previous | Thread Next