I am a Perl developer and have gravitated towards a specific suite of modules that I use for almost everything. I primarily build GIS and database oriented web applications for reporting and data entry and the like.
I'm curious what groups of modules other Perl devs have settled on using regularly.
Mine:
I use DBIx::Simple instead of plain DBI, the interface is, well, simpler and more regular. Also, of course, XML::Twig for XML processing. Then Getopt::Std for (simple) options, and YAML::Syck for temporary storage that doesn't require a DB, as well as debugging, instead of Data::Dumper. And, like you, Text::Template for most complex output.
I haven't used Perl in a while (sorry, my employer uses a mix of Java and C#), but I recall using the following quite a bit (in alphabetical order):
(You might notice a focus on non-framework webapp modules in there...)
My most common modules are probably
I'm not sure if it counts as a "tool" or not, but I tend to write my programs in a mostly functional style. I try to only assign to each variable once, and I try to avoid subroutines that modify their arguments. I generally prefer "grep" or "map" over a loop. The biggest exception to this rule is when I am building up a return value. I may push things onto the end of an array or add items to a hash in a loop.
Checking over recent stuff, here's what I see most often:
strftimeHere are three good sources to look at:
Off the top of my head:
I'm probably missing some obvious ones.
Not counting all the module development and test toolchain stuff, my toolbox for getting work done has included:
Nuts and bolts:
Hand tools:
Power tools:
I'm surprised nobody has mentioned Data::Dumper, I tend to leave the use declaration in my scripts because I know it will be useful for maintenance debugging.
I have used these each quite often, many of them going back 10 years.
CPAN module is, of course, one of the best that we sometimes forget about because it fades into the background. It is a meta-module, so to speak.
Top group below appears in output of our newperl script which provides a working example of best practices. the resulting perl script implements some basic requirements and hints at basic debug, so that perl novices can develop good habits using a real-life design. we have an analogous newperlmod as well.
The first two are more like pragmas in that they really hook onto the nature of the language built-ins.
Data gathering toolbox -- more specialized
tie with G/DBM family: for fast, scalable persisent hashesA recent discovery that is a boon for maintenance. Useful for output of config and batch files, e.g.
Net::FTP. Much of my job involves pulling files off of one system, manipulating them, and pushing them onto another system.
Without Perl6, the first one on the above list makes modules infinitely more readable (IMHO).
I develop web apps and desktop apps with these.
I notice a lot of plain CGI.pm fans on here, though I like CGI::Application, especially combined with C::A::D, a lot. It's still pretty light-weight, has a reasonable amount of plugins, a decent website and isn't as slow or heavy as Catalyst.