[Catalyst] catalyst++

Joe Landman landman at scalableinformatics.com
Fri Sep 29 18:01:25 CEST 2006

Garrett Goebel wrote:
> On Sep 29, 2006, at 8:39 AM, Jonathan Rockway wrote:
>>>   When I looked at InstantCrud, it only worked with TT, and I want to
>>> use it with Mason.  Is there a way to do this?
>> Welcome to the world of software. :)  If you want flexibility, you're
>> going to have to code it yourself.  If you want ready-to-go magic,
>> you're going to have to compromise somewhere.

Heh ... of course  (though your introduction is about 20 years late, but 

We are using Catalyst for building our future web products to deliver to 
customers.  Jifty did look neat, but I had a sense of a less clean 
separation between my layers, and we largely had something "jifty-like" 
in place for our development before.

What I like is that I just implemented one of our queue-status (Linux 
cluster job queue) with state, significant analysis, and other bits, 
quite quickly without fighting over sessions, cookies, and other bits, 
with displays in Mason.  This is what we want to continue to do.

What I would like is to add something like

	myapp_create.pl authentication source
	myapp_create.pl authorization dsn

but I didn't want to re-invent the wheel, if it was in flight.  For the 
moment I am doing this bit by hand (largely with cut and paste from the 
examples, and then tweaking from there).

> I think Zbigniew Lukasiak has recently mentioned plans to modularize  
> forms generation. You might want to contact and work with him on  
> providing the capability to generate alternate Mason views.

Unfortunately we don't have much in the way of time/manpower to work on 
more than the core apps that we are working on now.  I would like to, 
but we are not there yet.

If we get to the point were I can afford to commit such resources, we 
will examine where it makes the most sense to contribute.

>> Also, Jifty probably does exactly what you want (Lots of magic,  
>> lots of
>> mason).  Unfortunately it's a bit too clever for its own good --  
>> syntax
>> like:
>> use MyWeblog::Record schema {
>>   column title =>
>>         type is 'text',
>>         label is 'Title',
>>         default is 'Untitled post';
>>   column body =>
>>         type is 'text',
>>         label is 'Content',
>>         render_as 'Textarea';
>> }
>> has to be confusing to people who are new to perl.  (Why he couldn't
>> have just done

Hmmm... I think that syntax is not too clever for its own good.  My 
DBIx::SimplePerl is all about hiding simple SQL from people who frankly 
don't want to deal with writing any SQL and have really simple tasks to 
perform.  This is in line with it.  Manipulate the database by 
manipulating simple Perl structures.

Matt Trout's DBIx modules are this done right in a big way (with lots of 
extra things that I don't normally care about).  If I want to create a 
table for setup of a really simple app, I should be able to do this 
*easily*.  Same thing if I want to search/update a simple database, not 
a complex database (most of our apps use very simple databases).


Joseph Landman, Ph.D
Founder and CEO
Scalable Informatics LLC,
email: landman at scalableinformatics.com
web  : http://www.scalableinformatics.com
phone: +1 734 786 8423
fax  : +1 734 786 8452 or +1 866 888 3112
cell : +1 734 612 4615

More information about the Catalyst mailing list