Grokbase
x

Brian Cassidy (brian.ca...@nald.ca)

Profile | Posts (65)

User Information

Display Name:Brian Cassidy
Partial Email Address:brian.ca...@nald.ca
Posts:
65 total
21 in Catalyst Framework Development
21 in Catalyst Framework
1 in catalyst@lists.scsys.co.uk
5 in DBIx::Class
19 in HTML::FormFu

5 Most Recent

All Posts
1) Brian Cassidy [Catalyst] [PATCH] Catalyst::Plugin::ConfigLoader - documentation
| +1 vote
A slightly altered patch has been checked into svn revision 7926. Thanks, -Brian
catalyst@lists.scsys.co.uk
[ Profile | Reply to group ] [ Flat  Thread  Threaded ]
Sergio Salvi wrote:
> After hitting my head against a wall, I figured out how to pass
> options to Config::General using C::P::ConfigLoader.
>
> Here is a documentation patch that would have saved my head :)
>
> Patched against r7925 from http://dev.catalyst.perl.org/repos/Catalyst.

A slightly altered patch has been checked into svn revision 7926.

Thanks,

-Brian
2) Brian Cassidy [Catalyst] [RFC] refactoring of component resolution methods + warnings
| +1 vote
Hey All, Further to my message from the 27th of May about component resolution, I have created a...
catalyst@lists.scsys.co.uk
[ Profile | Reply to group ] [ Flat  Thread  Threaded ]
Hey All,

Further to my message from the 27th of May about component resolution, I
have created a branch [1] to refactor how it works behind the scenes.

Here is the excerpt from my updates to the Changes file:

- Refactored component resolution (component(), models(), model(), et
al). We now
   throw warnings for two reasons:
   1) model() or view() was called with no arguments, and two results 
are returned
      -- set default_(model|view), current_(model|view) or 
current_(model|view)_instance
      instead
   2) you call a component resolution method with a string, and it 
resorts to a regexp
      fallback wherein a result is returned -- if you really want to 
search, call the
      method with a regex as the argument

In detail, I've gutted most of the private methods that were used for
this task. In their place are 3 methods:

(1) _filter_component() -- this hasn't changed. it handles components
that ACCEPT_CONTEXT.
(2) _comp_names() -- a slight implementation change. used, for example,
when you need to grab all of the valid values for controllers.
(3) _comp_search_prefixes() -- this combines the old _comp_singular,
_comp_explicit and _comp_search methods in to one method.

Although combining the 3 methods in to 1 sounds bad, the result is 32
lines long -- 19 if you take out blank lines and comments.

Besides the refactoring, there is now an explicit regexp behavior for
component searching. All of model(), view(), controller() and
component() will accept a regexp and return (in list context) the matches.

Example from the docs:

# find all views that start with Foo
my @foo_views = $c->view(qr{^Foo});

FYI: all of the existing tests pass, plus I've added a number of new
tests for the regexp behavior.

Please take a look and pass along your comments.

-Brian

[1]
http://dev.catalyst.perl.org/repos/Catalyst/Catalyst-Runtime/5.70/branches/compres/
3) Brian Cassidy [Catalyst] Re: new catalyst version breaks test in Catalyst-Model-Adaptor-0.02 and Catalyst-Plugin-Unicode-0.8
| +1 vote
This was a regression I created when trying to solve the Path(0) bug. Essentially "Path" by itself...
catalyst@lists.scsys.co.uk
[ Profile | Reply to group ] [ Flat  Thread  Threaded ]
John Goulah wrote:
> On Mon, May 19, 2008 at 5:35 PM, John Goulah <jgoulah@gmail.com> wrote:
>> Hi folks,
>>
>> I've narrowed down that the new version (5.7013) breaks tests in both
>> of these modules, but I'm not sure why (they work against the last
>> version of Catalyst 5.7012, so there has been some change thats
>> affected this)
>
> mst thinks this is almost certainly an error screen, anyone care to comment?

This was a regression I created when trying to solve the Path(0) bug.
Essentially "Path" by itself means Path('') which is defined but of 0
length.

Fixed in svn r7778 (+ test).

-Brian
4) Brian Cassidy [Catalyst] FYI: Contracting possibilities
| +1 vote
Not sure at all. In the interest of full disclosure here is the private message i was sent (on vox)...
catalyst@lists.scsys.co.uk
[ Profile | Reply to group ] [ Flat  Thread  Threaded ]
J. Shirley wrote:
> Brian, are you sure this is a legit gig? It seems moderately
> 419-esque in the vocabulary and some other things.

Not sure at all. In the interest of full disclosure here is the private
message i was sent (on vox) -- there isn't much private about it, so i'm
sure he wont' mind.

---

Catalyst help required
Hi Brian,

I am looking for Catalyst help.  Do you know of any hired guns out there 
looking to do some fairly easy projects?  We are in Silicon Valley but 
can work with anyone.


thanks,
Michael

---

-Brian
5) Brian Cassidy [Catalyst] FYI: Contracting possibilities
| +1 vote
Hey All, I was notified via private message that a start-up in Silicon Valley is looking for some...
catalyst@lists.scsys.co.uk
[ Profile | Reply to group ] [ Flat  Thread  Threaded ]
Hey All,

I was notified via private message that a start-up in Silicon Valley is
looking for some Catalyst help. I figured I'd forward on the message.

Please read the relevant post here:

http://zenshadow.vox.com/library/post/catalyst-help-please.html

-Brian

NB: I know nothing about this person or their company. I'm just passing
along the info.

spacer
Profile | Posts (65)
Home > People > Brian Cassidy