Bug #86940 for Moose: 'extends' sometimes tries to load modules

Mon Jul 15 06:07:49 2013 MAUKE [...] cpan.org - Ticket created

Subject:

'extends' sometimes tries to load modules

The documentation for 'extends' says: | extends (@superclasses) | | This function will set the superclass(es) for the current class. What it doesn't say: In addition to setting up inheritance, in some cases[1] it tries to load corresponding modules from disk first. I find this behavior surprising and it doesn't seem to be documented anywhere. I think this behavior should be documented (or alternatively removed). [1] exact logic hard to describe; see sub _is_class_loaded in Class::Load::PP

Mon Jul 15 09:14:02 2013 doy [...] tozt.net - Correspondence added

Subject:	Re: [rt.cpan.org #86940] 'extends' sometimes tries to load modules
Date:	Mon, 15 Jul 2013 09:13:46 -0400
To:	Lukas Mai via RT <bug-Moose [...] rt.cpan.org>
From:	Jesse Luehrs <doy [...] tozt.net>

On Mon, Jul 15, 2013 at 06:07:50AM -0400, Lukas Mai via RT wrote: Show quoted text

> Mon Jul 15 06:07:49 2013: Request 86940 was acted upon. > Transaction: Ticket created by MAUKE > Queue: Moose > Subject: 'extends' sometimes tries to load modules > Broken in: 2.1001 > Severity: (no value) > Owner: Nobody > Requestors: MAUKE@cpan.org > Status: new > Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=86940 > > > > The documentation for 'extends' says: > > | extends (@superclasses) > | > | This function will set the superclass(es) for the current class. > > What it doesn't say: In addition to setting up inheritance, in some > cases[1] it tries to load corresponding modules from disk first. I > find this behavior surprising and it doesn't seem to be documented > anywhere. I think this behavior should be documented (or alternatively > removed). > > [1] exact logic hard to describe; see sub _is_class_loaded in > Class::Load::PP

Why is this surprising? 'use base' and 'use parent' both work the same way. Noting this in the documentation probably wouldn't be a bad idea though. -doy

Mon Jul 15 09:14:03 2013 The RT System itself - Status changed from 'new' to 'open'

Mon Jul 15 12:35:50 2013 MAUKE [...] cpan.org - Correspondence added

On Mon Jul 15 09:14:02 2013, doy@tozt.net wrote: Show quoted text

> On Mon, Jul 15, 2013 at 06:07:50AM -0400, Lukas Mai via RT wrote:

> > The documentation for 'extends' says: > > > > | extends (@superclasses) > > | > > | This function will set the superclass(es) for the current class. > > > > What it doesn't say: In addition to setting up inheritance, in some > > cases[1] it tries to load corresponding modules from disk first. I > > find this behavior surprising and it doesn't seem to be documented > > anywhere. I think this behavior should be documented (or alternatively > > removed). > > > > [1] exact logic hard to describe; see sub _is_class_loaded in > > Class::Load::PP

> > Why is this surprising? 'use base' and 'use parent' both work the same > way. Noting this in the documentation probably wouldn't be a bad idea > though.

It's surprising because 'extends' is not described as being like base/parent; it's described as setting superclasses, which I understand to be like '@ISA = ...'. The documentation for base/parent says: "Allows you to both load one or more modules, while setting up inheritance from those modules at the same time." ... i.e. it emphasizes the module loading part that 'extends' doesn't even mention.

Mon Jul 22 12:02:52 2013 DROLSKY [...] cpan.org - Correspondence added

This is patched in HEAD.

Mon Jul 22 12:02:54 2013 DROLSKY [...] cpan.org - Status changed from 'open' to 'resolved'