[Catalyst] re catinabox -- have you ever actually done this on a non-root server? Re: Catalyst Documentation

Thu Jun 1 13:05:51 CEST 2006

have you actually tried this on a shared (non root) server?

A few months ago, I did, with pretty disappointing results. I
attempted to document my modest success, but mainly failure, in

http://dev.catalyst.perl.org/wiki/Dreamhost

Maybe because I was sort of feeling sheepish, this sort of trails off
without really admitting that I wasn't able to accomplish what I
wanted to do, or warning off others who might dare to tread where I
had failed.

IIRC, by using catinabox, I was able to get a basic cat install going
okay. Just catalyst, not Bundle::Catalyst, or whatever the modern
equivelant is thiese days -- which doesn't let you do very much.

When I tried adding some of the goodies from bundle catalyst and
elsewhere, which are necessary to get any of the sample apps from the
wiki up to play with (including the one packaged up with catinabox
iirc ?), I couldn't. CPAN kept saying, can't do this without sudo
this, and asking me if I should install such and so dependency (even
though I had follow turned on in my cpan config). Dependency hell.
Supposedly cpanplus improves things significantly, if you have the
right linux kernel, if you have the right version of perl... if, if
if. Eventually I just gave up.

By contrast, I was able to get the shadowcat install and related
goodies pretty painlessly on a freshly installed suse box where I had
root.

Maybe if my cpan/unix skills were better, I would have been able to
get it to work in a non-root environment. I am thinking of giving it a
try again in the next few days, since my cpan/unix skills *are* in a
fact a good deal better than they were several months ago when I first
tried this.

My attempt to install on dreamhost was a tremendous learning
experience, however, it made me feel like a luzer while I was going
about it, because I was led to believe this should be relatively easy
by catalyst's propaganda on shared hosting.

http://www.catalystframework.org/calendar/2005/10 :

[Some hand waving... and...] "http://mydomain.com/ should now Just
Work. Congratulations, now you can tell your friends about your new
website (or in our case, tell the client it's time to pay the invoice
:)"

This was *not true* for me. Sorry, but it didn't "just work." And
anyway, I didn't really *want* a learning experience, I wanted the
instant gratification of installing the latest cool new framework. And
I would bet I'm not the only one.

Like I said, after having given this a several months breather, I plan
on giving this another try over the next couple of days, so maybe I
will pop back here and sheepishly admit I am making a mountain out of
a molehill.

I'd like to hear from others who agree or disagree with my experience.

If getting catalyst installed as non root is non trivial, I believe
this needs to be said much more explicitly, and we should be steering
people towards XEN/UML, NOT dreamhost.

Thomas.

2006/6/1, Kieren Diment <diment at gmail.com>:
> well, there's catinabox:
>
> http://mfrost.typepad.com/photos/uncategorized/catboxjpg.jpg
>
> oops, I mean
> http://dev.catalyst.perl.org/wiki/Guides/Install/CatInABox
>
> or if you want to do it from cpan,
>
> http://www.shadowcatsystems.co.uk/static/cat-install
>
> For paravirtualiasation, people who know™ recommend xen, but I use uml as it
> fits my budget better.   Catalyst is very well supported under Debian by the
> way.
>
>
> On 01/06/06, Thomas Hartman <thomashartman1 at googlemail.com> wrote:
> > "Mason seems to require custom systems in place, and iirc from when I
> > looked into mason, a long time ago, also requires mod_perl, a problem
> > in a shared hosting environment. "
> >
> > When I tried catalyst on a shared environment, it croaked in
> > dependency hell. Have things improved any?
> >
> > I'm gearing up for doing a web site on a paravirtualized server, just
> > trying to decide between uml and xen...
> >
> > thomas.
> >
> > 2006/6/1, Kieren Diment < diment at gmail.com>:
> > > Apologies for the arrogance :)
> > >
> > > I think that the point I was making was that Mason seems to require
> custom
> > > systems in place, and iirc from when I looked into mason, a long time
> ago,
> > > also requires mod_perl, a problem in a shared hosting environment.
> Perin
> > > expressed this much more clearly than me, clearly because he has
> significant
> > > experience with Mason, which I do not.
> > >
> > > Technology goes through iterations - as in the progression from
> mechanical
> > > pianos, to wax cylinders to discs and so on.  Mason is an excellent
> example
> > > of high quality stuff from the previous iteration of web development
> > > methodology.  Because of my circumstances I wouldn't be able to be
> involved
> > > in serious web development without something like catalyst to take out
> the
> > > repetetive strain.
> > >
> > > As mentioned previously, the catalyst documentaton:
> > >
> > > svn co
> > >
> http://dev.catalyst.perl.org/repos/Catalyst/branches/Catalyst-docs
> > > currently in the docs branch of the svn repos is essential reading.
> > >
> > > svn co
> > >
> http://dev.catalyst.perl.org/repos/Catalyst/branches/Catalyst-docs
> > >
> > > On 01/06/06, Paul Wallingford < paul at cybergestalt.net> wrote:
> > > > Perrin Harkins wrote:
> > > > > On Wed, 2006-05-31 at 07:47 +1000, Kieren Diment wrote:
> > > > >
> > > > >> Mason is a templating system with a few application-development
> type
> > > > >> features included. Catalyst is a full blown application development
> > > > >> environment.  Mason makes a serviceable Catalyst View, although I
> > > > >> perfer Template Toolkit for a couple of reasons.
> > > > >
> > > > > I think you're exaggerating the differences here.  Although they go
> > > > > about it in totally different ways, Mason and Catalyst offer very
> > > > > similar functionality.  For example:
> > > > >
> > > > > - Flexible ways to map URIs to code.
> > > > > - Uniform API for access to request data and passing parameters.
> > > > > - Abstraction of runtime environment so that mod_perl, CGI, and
> FastCGI
> > > > > all work the same.
> > > > > - Error handling with helpful debug screens.
> > > > > - Plugins for sessions and other extras.
> > > > >
> > > > > You could say that Mason has MORE ground to cover in the docs, since
> it
> > > > > actually includes a templating system and a cache, while Catalyst
> just
> > > > > provides glue code.  (The cache is provided by Cache::Cache, but is
> > > > > covered nicely in the Mason docs.)
> > > > >
> > > > > My point is not to belittle Catalyst, but rather to say that Mason
> is a
> > > > > good example of a complex Perl project with very readable docs, and
> it's
> > > > > worth looking at.  The quality of documentation was one of the
> things
> > > > > that set Mason apart from the very beginning, and I think it was a
> key
> > > > > reason for Mason's success.  It came with guides for developers and
> > > > > administrators and had a more professional feel to it than most of
> the
> > > > > other tools at that time.
> > > >
> > > > Thanks for the reply Perrin.  I am very familiar with Mason and have
> > > > built a lot of stuff that does what Catalyst does.  However, I am
> > > > looking for something to speed development - and a framework like
> > > > Catalyst just might do the trick.  Building custom systems each time
> > > > takes too long and I don't have time to build my own framework.
> > > > Catalyst *seems* to fit the bill and I want to evaluate it, but the
> > > > documentation seems lacking.  Maybe I am just spoiled with Mason.
> > > >
> > > > That said, obviously I feel that Catalyst has a lot to offer,
> otherwise
> > > > I would not even consider wasting valuable time evaluating it (and
> there
> > > > are a lot of other frameworks that I have discounted right off the
> bat).
> > > >   This says a lot about my opinion of Catalyst.  Plus a lot of people
> > > > use Catalyst and Mason together, which is encouraging.
> > > >
> > > > I was a bit turned off by the arrogant reply by Kieren, however.  I
> just
> > > > wanted a little guidance, not some overblown sales pitch.
> > > >
> > > > Thanks again.
> > > >
> > > > Paul Wallingford
> > > >
> > > >
> > > > _______________________________________________
> > > > Catalyst mailing list
> > > > Catalyst at lists.rawmode.org
> > > > http://lists.rawmode.org/mailman/listinfo/catalyst
> > > >
> > >
> > >
> > > _______________________________________________
> > > Catalyst mailing list
> > > Catalyst at lists.rawmode.org
> > > http://lists.rawmode.org/mailman/listinfo/catalyst
> > >
> > >
> > >
> >
> > _______________________________________________
> > Catalyst mailing list
> > Catalyst at lists.rawmode.org
> > http://lists.rawmode.org/mailman/listinfo/catalyst
> >
>
>
> _______________________________________________
> Catalyst mailing list
> Catalyst at lists.rawmode.org
> http://lists.rawmode.org/mailman/listinfo/catalyst
>
>
>