JCachePage::get() nonsense

[Paladin]

OK here's something weird. Code like the following occurs both in
JCachePage::get and JCachePage::store and I don't get it:

$this->_id = $id;
$this->_group = $group;

Seems simple and straightforward, right? Except _id and _group are
*not* properties of JCachePage nor of any parent object. What *are*
they?

I'd add the properties, but I have no idea where they belong or what
they're about. (I'm just getting started writing tests for the JCache
family, so I don't have a good mental model yet of what's supposed to
be going on here.)

[Ian MacLennan]

Well...  not sure what they are...  but to warn you...  the caching stuff is a little bit messy with a mix of half implemented and poorly implemented stuff.  It is certainly something that needs to be reworked at some point.

Ian

[Paladin]

On Nov 24, 10:26 pm, Ian MacLennan <ian.maclen...@joomla.org> wrote:
> Well...  not sure what they are...  but to warn you...  the caching stuff is
> a little bit messy with a mix of half implemented and poorly implemented
> stuff.  It is certainly something that needs to be reworked at some point.

I'm noticing more than a few fiddly bits about JCache while I'm
testing. Just as an uninformed guess, the memcache and eaccelerator
are the badly implemented bits? Those and the view and callback
handlers gave me the most issues during writing some basic tests for
JCache, anyway. I'm checking in a set of rudimentary tests for JCache;
shouldn't break the unittest tree and maybe they'll prove useful for
someone while debugging. I'm going to walk on through this family, and
probably amend some of the tests as I go, as my understanding of the
code improves. This first batch of tests are just a few steps away
from shooting in the dark.

And, as every ex-SAC vet knows, shooting in the dark only is reliable
if you nuke 'em 'til they glow, first.

(Unofficial Strategic Air Command motto: 'We nuke 'em 'til they glow,
so you can shoot 'em in the dark!')

Just think of unit tests as the developer's SAC. ;{>}

[klas berlič]

> Just as an uninformed guess, the memcache and eaccelerator
are the badly implemented bits?

That would be all caching methods except file caching.

2009/11/25 Paladin <arlen....@gmail.com>

--
------------------------------------------------------------------
http://www.bzzzz.biz  - Creating images
Marketing and web communication agency

http://www.belmondo.si - vacation packages,
travel, flight tickets & hotels

------------------------------------------------------------------

[Paladin]

On Nov 25, 4:03 am, klas berlič <klas.ber...@gmail.com> wrote:
> > Just as an uninformed guess, the memcache and eaccelerator
>
> are the badly implemented bits?
>
> That would be all caching methods except file caching.

And, after having opened up JCacheStorage for the first time, all I
can say is:

I see.