Front page | perl.perl5.porters |
Postings from August 2012
On Aug 1, 2012, at 1:11 PM, Ricardo Signes wrote: > * Nicholas Clark <nick@ccl4.org> [2012-07-26T05:50:06] >> On Wed, Jul 25, 2012 at 09:24:13PM -0000, Father Chrysostomos wrote: >>> rjbs wrote: >>>> The smoke-me/require branch contains a fix for this bug, and will land on >>>> blead shortly assuming further testing finds no problems. >>> >>> That fix does not look correct to me. >>> >>> Why should require ::foo be treated as require foo, but not ::::foo or >>> main::foo? >> >> As to ::foo vs ::::foo, because they are not the same: >> [...] > > The answer to the disagreement here is not obvious to me without more time, and > we've classified this problem as only a bug fix, rather than a critical > security issue, so I am tempted to get 5.16.1 shipped for the CLEAR bug fixes > rather than wait much longer for this. The change that makes ‘require ::foo’ mean ‘require foo’ is an incompatible change. Such a change should be supported by sound reasoning. The reasoning here is that ::foo and foo refer to the same package, therefore they should load the same module. But that conflates packages and modules, which are not the same. main::foo and main::main::main::foo also refer to the same package as ::foo, so should require main::foo change? I hope not. Therefore I don’t consider the reasoning sound. Hence, since we are changing its behaviour, require ::foo should be made an error, for consistency with require ::::foo. (BTW, foo::bar and foo::::bar refer to different packages, but the same module.)Thread Previous | Thread Next