use Perl

I'm glad that Yves asked me to help him on Date::Simple. Since I don't know much XS, I thought it would be a nice way of learning it a bit.

After I got co-maintainer status of this project, I already fixed some bugs and I'm planning to release a new version this weeks with some bugs fixed.

If you are reading this, and you posted some bug I took on rt, please check if your bug is solved on the latest developer release (3.03_03) and send me some feedback.

I updated last week DBIx::Placeholder::Named to version 0.07, which is more compliant with DBI - accepting PlaceholderSuffix and PlaceholderPrefix keys - after I presented it on Amsterdam.pm meeting. I can say that I was surprised that nobody yelled with some module doing the exact same thing...

And continuing the things-I-would-like-to-see-on-DBI I wrote DBIx::Dictionary, which tries to add, in the same clean way DBIx::Placeholder::Named does with named placeholders, query dictionary functionality.

It integrates well with DBIx::Placeholder::Named, and currently only supports Config::General configuration files. I'm planning to add support for some SQL compliant configuration file.

I'd like to hear from you what you think about those modules!

After asking at Perlmonks, I chose DBIx::Placeholder::Named for the DBI subclass I wrote inspired on Tao::DBI. It was released yesterday, but may be not available on some CPAN mirrors.

I already identified it needs some improvement on documentation, so I plan to fix and release it soon.

Hope you like it.

I've been working last week on SQL::Tokenizer, which was used to, erm, tokenize a SQL query correctly to extract named placeholders for a DBI subclass I'm working on. It is actually using the known-and-loved Perl's regular expression engine, but maybe I can write a version using Pugs::Compiler::Rule, let's see.

About the DBI subclass I mentioned above, Adriano wrote the Tao::DBI, which is a proxy to DBI that implements the named placeholders concept. I liked that and re-implemented the Tao::DBI as a subclass of DBI, but since it is not backward compatible with Adriano's one, I have to found another name for it before release.

Any suggestion?

I've been testing Catalyst for a while, and it seems to be a very nice framework for web applications. I think it has the same power and robustness other widely used frameworks have, and it is written in Perl.

It is powerful enough for developing a big project, and simple enough to write a simple application. Its plugin architeture Just Workstm and there's a plenty of already written plugins around.

Kudos to Catalyst development team! Keep the good work!

My vacations started yesterday, so I think I'll have a good time ten days from now. As I don't need to bother about my job, I can learn Bricolage and other things that are needed for a job I'm looking for.

Talking about Bricolage, as it needs a lot of Perl modules, it took me a good effort to make things work. Now I have to figure out how do work with it, before looking its internals. If you know a good HOWTO or manual about Bricolage, aside its own documentation, please let me now :)

YAPC::Brasil::2006 was a big success. Randal Schwartz did a lecture about communication and inside out objects, and David Fetter about DBI-Link and Pg/Perl. Er Galvão, Flávio Glock and Marco Lima also did a lecture about The Monastery Gates, Perl6 and CPAN. I'd never seen a Perl lecture with a full room like that :)

I hope this year's YAPC::SouthAmerica will be another success for South America Perl Users.

Nice, two days left to YAPC::Brasil::2006. This year Randal Schwartz and David Fetter will be here talking about cool things about Perl. I think a new year for FOSS and especially Perl in Brasil will begin.

This week I've written a lot of code and documentation on logspy.pl. I'm planning to release it on SourceForge in about two weeks, after writing some input and output examples modules.

logspy.pl is a rewrite of a project I wrote at work, for parsing Postfix' logfiles on the fly. I figured out that it could be used to parse other logs, so I wrote a simple main engine and a plugin architeture for modular parsing, i.e. you will be able to use an input module like Logspy::Input::Messages to parse syslog messages file (/var/log/messages) and outputing it to a MySQL database using Logspy::Input::Messages::MySQLDump, for example.

You can write your own modules for anything you want. At work, for example we use a similar system that reads the /var/log/maillog file and then count messages sent from authenticated users and messages received from external IP addresses. Those datum is used to generate iptables rules for spam prevention.

I'll make sure you'll know when I release it :)

Update: An english typo :)