SPUG: Best practice? DBI connection parameters *not* in Perl code

C.J. Adams-Collier cjac at colliertech.org
Thu Apr 8 15:04:46 PDT 2010


Ye'hear that, Ingy?  I think he said your mother was a hamster.

On Thu, 2010-04-08 at 11:26 -0700, Brian E. Lozier wrote:
> Regarding YAML -- in my experience, the "YAML" module on CPAN is buggy
> and can't properly handle complex structures.  This bit us and we
> moved to JSON (this wasn't for configuration).  It should be fine for
> simple config data (but may be a bit heavy weight for that).  See the
> bug list: https://rt.cpan.org/Public/Dist/Display.html?Name=YAML
> 
> It's been suggested to use YAML::Syck, but that's compiling its own
> huge list of bugs as well.  Depending on your deployment targets, you
> may want to stick with a pure-perl solution, which, I believe,
> YAML::Syck isn't.
> 
> I've used Config::Fast and Config::General and have been happy with both.
> 
> Brian
> 
> On Thu, Apr 8, 2010 at 10:53 AM, Skylos <skylos at gmail.com> wrote:
> > I'm not entirely sure if its best practice, but my feeling is that I would
> > prefer a config file because its a more obvious artifact that can be
> > deployed with the application.  As for format, YAML of course.  And here's a
> > module especial for that.
> >
> > http://search.cpan.org/~moconnor/YAML-AppConfig-0.16/lib/YAML/AppConfig.pm
> >
> > Skylos
> >
> >
> > On Thu, Apr 8, 2010 at 10:45 AM, Michael R. Wolf <MichaelRWolf at att.net>
> > wrote:
> >>
> >> As some DBI-based code is being ported to other machines, I'm having to
> >> pass command-line parameters to too many administrative and test programs.
> >>  I'd like to change this *once* external to *all* programs and have it apply
> >> to them all.  This will have the added benefit of articulating a cleaner
> >> differentiation between environments (development, test, production).
> >>
> >> What's more popular to do this?
> >>  - environment variables
> >>  - config file
> >>
> >> In code, and in *theory*, it's a trivial difference...
> >>
> >>    $db_user = $ENV{db_user} || 'root';
> >>        versus
> >>    $db_user = $some_config_ref->get_attribute('db_user') || 'root';
> >>
> >> I'm more interested in how this works in *practice*.  That is, when
> >> fielded, which seems to be groked by more end users?  And if the answer is
> >> 'config', which of the 2 score and twenty config modules seems to be groked
> >> by more end users?
> >>
> >> Thanks,
> >> Michael
> >>
> >> --
> >
> >
> > --
> > "If only I could get rid of hunger by rubbing my belly" - Diogenes
> >
> > _____________________________________________________________
> > Seattle Perl Users Group Mailing List
> >     POST TO: spug-list at pm.org
> > SUBSCRIPTION: http://mail.pm.org/mailman/listinfo/spug-list
> >    MEETINGS: 3rd Tuesdays
> >    WEB PAGE: http://seattleperl.org/
> >
> _____________________________________________________________
> Seattle Perl Users Group Mailing List
>      POST TO: spug-list at pm.org
> SUBSCRIPTION: http://mail.pm.org/mailman/listinfo/spug-list
>     MEETINGS: 3rd Tuesdays
>     WEB PAGE: http://seattleperl.org/

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: This is a digitally signed message part
URL: <http://mail.pm.org/pipermail/spug-list/attachments/20100408/53b3f2a6/attachment.bin>


More information about the spug-list mailing list