Bug #78227 for CHI: Could it be redone without Moose?

CHI gets a ton of benefits from Moose. I tried to use Mouse in an old version, but I kept running into missing or broken features. What do you mean by "really huge"? Do you mean amount of memory, or amount of disk space, or number of dependencies, or it just feels big conceptually? Are you on a hosting provider with a very restricted amount of memory/disk space? On Thu Jul 05 14:04:13 2012, DMUEY wrote:

And let me assure you, I'm not trying to start a Moose war, I just love the concept of CHI (I enjoyed your talk @ YAPC about it BTW) and have seen the same basic idea implemented wonderfully without moose (e.g. Log::Dispatch) yes, process RSS goes up almost 12MB just using it from a quick and dirty check: [dmuey@multivac ~]$ perl -e 'print `ps u -p $$`;' USER PID %CPU %MEM VSZ RSS TT STAT STARTED TIME COMMAND dmuey 39922 0.0 0.0 2437344 1204 s000 R+ 1:46PM 0:00.04 perl -e print `ps u -p $$`; [dmuey@multivac ~]$ perl -MMoose -e 'print `ps u -p $$`;' USER PID %CPU %MEM VSZ RSS TT STAT STARTED TIME COMMAND dmuey 39928 46.9 0.2 2444940 13124 s000 R+ 1:46PM 0:00.35 perl -MMoose -e print `ps u -p $$`; [dmuey@multivac ~]$ by comparison Log::Dispatch, that I noted earlier as a similar concept but for logging instead of caching, only adds about 2 MB to RSS even with quite a dependency chain itself. not so much that maybe a bit: it takes time to find all of them and load them and compile them, for what benefit exactly? a few less lines of code and MOPiness? Doesn't seem worth it to me for an abstract caching mechanism I suppose that fits a little, MOP is cool, I get that, its a great idea. Just not when its cost outweighs the benefit. no, I just don't want my personal servers bogged down for no reason and our customers demand reasonable memory use, thus Moose is out the door along w/ Java :)

ok, thanks for your time and quick response.

On Thu Jul 05 16:34:54 2012, swartz@pobox.com wrote: I echo Daniel's comments, especially those quoted above. With the recent buzz about Moo, [http://blogs.perl.org/users/sawyer_x/2012/07/anymoose-is-out-moo-is-in.html] I wonder if you would consider moving in that direction? What benefits does Moose provide that Moo doesn't and that prevent doing that implementation easily? (Wish I had more experience with both, I'd answer that question myself!) I am in a position where it will be some time before my company can migrate key applications to Moose (much as I would like to). And until MOP is in the core, I really would like to have the option of moving caching to CHI in the meantime. With the full Moose dependency, I simply don't. DLS

On Jul 27, 2012, at 11:12 AM, http://dmaestro12.myopenid.com/ via RT wrote: Moo is certainly more likely than plain-old-objects, I'll consider it. I'd have to think about the feature gap. Likewise, can I ask what is preventing your company from using Moose? Is it memory usage? Startup time?

On Fri Jul 27 18:23:56 2012, swartz@pobox.com wrote: Also echoing support for $Moo++; $Moose--

I have a work in progress port of CHI to Moo. I have it passing all of its tests, and all of dependent CPAN modules still work with no changes. I still have a bit of cleanup I'd like to do, and I'd like to get some people testing it in the real world, but everything is looking pretty good so far.

Huzzah! On Mon Jun 24 17:25:22 2013, haarg wrote:

On Mon Jun 24 17:25:22 2013, haarg wrote: It looks like CHI now uses Moo as of Monday. Shouldn't this ticket be closed now?