These are generally excellent suggestions. If you would like to supply an ssh key someone can get you commit access to update the relevant POD.
Regarding Lazy, no it would simply cache the value and supply it on demand rather than at attribute definition time. -Chris On Jul 29, 2010, at 2:50 PM, Offer Kaye <[email protected]> wrote: > Hi, > New to Moose here, trying to read through the docs and get a basic > grasp of this rather complex beast... As I was reading I came across > some things which bothered me and thought I would ask about them and > perhaps offer some suggestions. So here goes :) > > 1. In the "Moose::Manual::Attributes" documentation - at one place it says: > "This is a dumb example" > > I didn't understand why the given example is dumb. I think that's a > rather harsh word to use. Maybe a better word would be "trivial"? > Personally I thought that the given example of using a subroutine as > the default value was rather good, neither trivial nor dumb. But > probably that's just me. > > 2. Again in the "Moose::Manual::Attributes" documentation - in the > explanation for the default value is says: > > ######################################################## > If you want to use a reference of any sort as the default value, you > must return it from a subroutine. This is necessary because otherwise > Perl would instantiate the reference exactly once, and it would be > shared by all objects: > > has 'mapping' => ( > is => 'ro', > default => {}, # wrong! > ); > Moose will throw an error if you pass a bare non-subroutine reference > as the default. > > If Moose allowed this then the default mapping attribute could easily > end up shared across many objects. Instead, wrap it in a subroutine > reference: > > has 'mapping' => ( > is => 'ro', > default => sub { {} }, # right! > ); > This is a bit awkward, but it's just the way Perl works. > ######################################################## > > I'm sorry but it seems as if the parts saying "This is necessary > because otherwise Perl would instantiate the reference exactly once, > and it would be shared by all objects" and "If Moose allowed this then > the default mapping attribute could easily end up shared across many > objects" are basically saying the same thing. Maybe this could be > re-worded? Maybe something like: > > ######################################################## > If you want to use a reference of any sort as the default value, you > must return it from a subroutine. This is necessary because otherwise > Perl would instantiate the reference exactly once, and it would be > shared by all objects: > > has 'mapping' => ( > is => 'ro', > default => {}, # wrong! > ); > > has 'mapping' => ( > is => 'ro', > default => sub { {} }, # right! > ); > > Moose will throw an error if you pass a bare non-subroutine reference > as the default. > ######################################################## > > As you can see I also got rid of the last sentence - I'm sorry but to > me it just sounded too apologetic. > > Frankly I think it would be better to perhaps have some code inside > Moose that if a bare non-subroutine reference is given, it is replaced > by a subroutine reference - then you could remove all of this section > :) > > On a related note, reading further, I reached the "lazy" explanations. > Probably I misunderstood but wouldn't using "lazy => 1" solve the > above problem and allow people to use a bare non-subroutine reference? > > 3. Again in the "Moose::Manual::Attributes" documentation - in the > documentation for "trigger" it says: > > This differs from an after method modifier in two ways > > However the "after method modifier" is not explained or mentioned > anywhere else in this POD. If it is explained elsewhere, maybe you > could create a link to that POD? > > Regards, > -- > Offer Kaye
