[Catalyst] Re: Why does $c->stats require -Debug flag?

[Zbigniew Lukasiak]

On Fri, Apr 25, 2008 at 1:04 AM, Aristotle Pagaltzis <pagaltzis at gmx.de> wrote:
> Hi Wade.Stuart,
>
>  * Wade.Stuart at fallon.com <Wade.Stuart at fallon.com> [2008-04-25 00:30]:
>
>
> > Aristotle Pagaltzis <pagaltzis at gmx.de> wrote on 04/24/2008 12:32:12 PM:
>  >> * Jonathan Rockway <jon at jrock.us> [2008-04-24 19:10]:
>  >>> * On Thu, Apr 24 2008, Aristotle Pagaltzis wrote:
>  >>>> * Jonathan Rockway <jon at jrock.us> [2008-04-24 11:25]:
>  >>>>> * On Thu, Apr 24 2008, Jon Schutz wrote:
>  >>>>>> No problems, if that's what the Catalyst standard says; I
>  >>>>>> must have missed it. Where is it? I'd like to consult it
>  >>>>>> on a number of matters... please post the link.
>  >>>>>
>  >>>>> Basically it's more of a "zeitgeist" than an actual
>  >>>>> document. There are some things that the community has
>  >>>>> decided and "just do".
>  >>>>
>  >>>> That?s the sort of feel-good bollocks I?d expect to read on
>  >>>> a Rails hype blog, not here. Unspoken rules and gut feel are
>  >>>> no way to run a community. Catalyst suffers from this in
>  >>>> general: way too little is written down, much less in any
>  >>>> systematic fashion.
>  >>>
>  >>> Nobody has time to run a bureaucracy.  We just want to write
>  >>> code.
>  >>
>  >> Yes, backcompat code. And I suppose the time to run a
>  >> deprecation cycle bureaucracy will find itself. File under
>  >> ?false laziness.?
>  >
>  > If you expect behavior over cycles, write test code.  If
>  > changes happen that make that test fail it will prompt
>  > discussion and offset the depreciation cycle to the closest
>  > change set.  If you want to document down to the dot,  without
>  > test code -- you are in for a world of outdated documentation.
>
>  I am not even talking about code docs here, I am talking about a
>  short note about the project's conventions. Something like the
>  `Documentation/CodingStyle` document in the Linux kernel source,
>  except appropriately scaled down since Catalyst is so much
>  smaller in terms of both SLoC and contributor count.
>
>  It doesn't have anything to do with bureaucracy, I'm talking
>  about one, maybe two pages of bullet points about the most
>  important guidelines of the project. "Methods without docs are to
>  be considered internal," "leading underscore in a documented
>  method name means it's OK for subclasses to touch but not
>  outsiders," "all patches must be accompanied by tests" etc etc,
>  that sort of thing. Not a law code, just a summary of how things
>  are done 'round here. In fact it *mustn't* be too elaborate and
>  intricate, else it won't get read.

I've put those points at http://catwiki.toeat.com/source_code/Coding_conventions

-- 
Zbigniew Lukasiak
http://perlalchemy.blogspot.com/