Mailing List Archive: Catalyst: Dev

Re: [Catalyst] Config-only components

Index | Next | Previous | View Threaded

jon at jrock

Feb 12, 2008, 5:07 PM

Post #1 of 2 (267 views)
Permalink

Re: [Catalyst] Config-only components

* On Tue, Feb 12 2008, Dagfinn Ilmari Mannsåker wrote:
> Hi all,
>
> Castaway mentioned the idea of having components defined entirely by
> config entries, without the need for any actual class files on disk, and
> I was bored, so I went ahead and implemented it.

Hi ilmari,

I applied this to my private repository last night, played with it a
bit, and quite like it. I've thought about it a bit and don't consider
it to be particularly invasive, although it's definitely not a bugfix.

Anyone else have an opinion on the worthiness of this patch for
inclusion in 5.70? I'm thinking it would be possible to make this a
plugin with only a small cleanliness sacrifice. I'll see if I can make
that work tonight...

Regards,
Jonathan Rockway

_______________________________________________
Catalyst-dev mailing list
Catalyst-dev[at]lists.scsys.co.uk
http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst-dev

pagaltzis at gmx

Feb 13, 2008, 1:08 AM

Post #2 of 2 (250 views)
Permalink

Re: Config-only components [In reply to]

* Jonathan Rockway <jon[at]jrock.us> [2008-02-13 02:10]:
> Anyone else have an opinion on the worthiness of this patch for
> inclusion in 5.70?

It worries me faintly that this sort of thing has the potential
to conceal issues in rare edge cases, where a component is
missing or somehow failed to load without crashing the app. At
the very least the fact that a component was created implicitly
should be logged. Possibly there should also be a strict mode
that will

a) only create components implicitly if they have been declared
(eg. with an `implicit_components` key in the config for
Catalyst::Component::Implicit or something), and

b) possibly die screaming if it finds that a component declared
already explicitly.

The latter might be too much. The former would be just a small
bit of extra code though, and would provide an escape hatch for
debugging deployment issues, which isn’t a lot of fun even under
the best of circumstances.

Regards,
--
Aristotle Pagaltzis // <http://plasmasturm.org/>

_______________________________________________
Catalyst-dev mailing list
Catalyst-dev[at]lists.scsys.co.uk
http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst-dev

Index | Next | Previous | View Threaded

Interested in having your list archived? Contact lists@gossamer-threads.com