Good afternoon,
I have been developing in mod_perl for many years, and my livelihood is
based on
being productive with it. I’ve spent the past week reading and testing
RoR and
it does look good. But before I take the plunge for my next project, I
was
hoping to hear from other mod_perl switchers.
One of the main reasons I like mod_perl is code flexibility. I like
being able
to insert code into any of Apache’s request phases. I like the code base
available through CPAN. I like the execution speed of registry scripts
and
mod_perl handlers. And maybe most importantly, I like my comfort in
working with
a ‘known’ entity and refined coding style (iow, I’ve been using it for
so many
years it has become an old & dear friend). (For small apps, I generally
use
CGI::Application and HTML::Template along with DBI/DBD::mysql. For
larger
projects, I use whatever I feel is appropriate, probably not CGI::App.)
I’m concerned I might lose to much of what I’m used to in switching to
RoR. I
can see what I would gain by switching, but what would I lose?
For example, I haven’t yet worked out the flow from page to another (in
RoR)
when there are validation errors. I have a well-tested methodology when
using
CGI::App’s run modes (along with Data::FormValidator). RoR also seems to
make a
lot of DBI calls when creating (& populating) ActiveRecord objects. And
that
seems to be compounded when multiple components are called in one
request cycle.
Eg. if I have a LoginUser object; it seems to get re-created when I
redirect
from one page/component to another.
Any comments or pointers to documentation for switchers would be greatly
appreciated. I have looked at alternative languages many times over the
years,
and I’ve never gotten even close to switching from mod_perl. So getting
this
close with RoR is sort of exciting.
Thanks,
Charlie