I've completed the cpan(1) fastest mirror enhancement I described in Perl in Catincan, an open source crowd funding proposal. I wanted to test this new crowd-funding system on something small. Everything worked out and "Perl" is the first project to have a fully funded feature implemented. There were some bumps because I was the first, but once I sent in my bank details, I had the money in a day. The code stuff is done, and once CPAN.pm releases a few things in my pull request, cpan(1) will have the new feature.
I specifically choose to work on my cpan script because it's part of core Perl, and I'd like to see other people who contribute directly to core try out some of their own features under "Perl". For those who aren't a contributor, I'm also interested in what you'd pay and how much you'd pay for a feature to be improved, implemented or whatever. Maybe you'd only pay $10, but maybe 1,000 other people would pay $10 too.
I'm also thinking of what I'd like to try next. I have a list of things I'd like to do, but I don't have any idea of the things that many other people would like me to do.
I'll be doing a live webcast about Pinto with Randal Schwartz for FLOSS Weekly. Tune in to http://live.twit.tv next Wednesday, March 27 at 08:30 (Pacific Time) to the see the show. Your can send in your questions in real-time via the #twitlive channel on irc.twit.tv.
Pinto is a robust tool for creating custom CPAN-like repositories of Perl modules. You can fill your repository with any combination of private and public modules, and then build/test/install them using the standard tools (e.g. cpan, cpanm, cpanp). Since you control the repository, you'll get exactly the same modules every time. Pinto also has some novel tools for tracking and managing changes, so you can upgrade modules with confidence and control.
The Milwaukee Perl Mongers just held their second meeting and had another great turnout. JT from the MadMongers came to visit and presented two fantastic talks.
To help everyone know where to go when they arrived last night, we laid down some raptor tracks to follow :)
It's a snow day and time to play. I have a cow orker who prefers to write all the business logic in SQL because he hasn't moved on from what he learnt in the 80's. Nice guy, but I suspect he has a hearing deficiency because he never listens to what I have to say (or he's got a nasty case of NIH). Yesterday, he happened to be getting on my last nerve, but that no longer matters for I will have sweet, sweet Revenge.
OK. After a bit of pain I've set up an empty database for the MusicBrainz database.
In particular, where they say 'sudo cpan Carton' you probably want 'cpanm Carton'.
This exercise was because, astonishingly, one of the users of my module GraphViz2 had doubts about it handling 149 tables spread over 4 schema.
The version of dbi.schema.pl shipped with GraphViz2 V 2.07 needed a patch to handle multiple schema, but after fixing that, you can view the SVG in all its ghastly glory.
Firstly, set up you env vars thusly:
env | grep DBI
DBI_PASS=musicbrainz
DBI_DSN=dbi:Pg:dbname=musicbrainz_db
DBI_SCHEMA=musicbrainz,cover_art_archive,report,statistics
DBI_USER=musicbrainz
Then the command is:
dbi.schema.pl svg musicbrainz.svg > musicbrainz.log
Further, you can download this set of files:
o musicbrainz.svg (size: 257,156 bytes)
o musicbrainz.log (output of dbi.schema.pl, containing the DOT file)
o dbi.schema.pl (patched)
o tables.log (the list of 149 tables as output by Pg's psql) here.
The influential
Design Patterns
book
lays out 23 patterns for programming.
One of them, the Interpreter Pattern, is rarely used.
Steve Yegge puts it a bit more strikingly -- he says
that the book contains
22
patterns and a practical joke.
That sounds (and in fact is) negative, but
elsewhere
Yegge says that
"[t]ragically, the only [Go4] pattern that can help code get smaller
(Interpreter) is utterly ignored by programmers".
(The
Design Patterns
book has four authors,
and is often called the Gang of Four book, or Go4.)
In fact, under various names and definitions, the
Interpreter Pattern and its close relatives and/or identical twins
are widely cited,
much argued and highly praised [1].
I adopted PDF::Report a couple of weeks ago after contacting the authors, who were happy to provide me with the privs I needed on pause, since then I've created a github repo but not had a chance to do any of the work I've been planning.
Before I start messing around with PDF::Report too much though, I'm doing housekeeping.
- so first task is to apply the patches in the current rt.cpan.org bug queue and close them (2 down, 1 to go).
- also, now that it's on github I've added the metadata for that and the perl license to the Makefile.PL so that it's a bit nicer.
- I've also updated the pod so that it's got more of what you expect - a SYNOPSIS and more detailed pod will follow in the next release (and/or possibly the next one)
Often when deploying application one may face the risk of divergence between testing and production environment.
Even though you have a stage servers `like production one', it'd be reasonable to check distributive in
production environment before release is happened. I would call it `early` testing. Yes, of course, some
subtle bugs will arise only in runtime phase, and unit test cannot cover it all, I say here about
prerequisite unmet issues. In perl world unit tests and prerequisites checks are executed in standart way. One follows standard procedure, when installing things.
I put here
example for Module::Build based project, but with ExtUtils::MakeMaker it's almost the same:
perl Build.PL # check dependencies and generate build file
./Build
./Build test # run unit tests
So, why not to automate this process in continues integration approach, like for example Jenkins does?
Note: this post assumes you have Test::Class::Moose version 0.06 or higher (on its way to the CPAN now).
By now you may have heard of Test::Class::Moose. I wrote this to solve a need that many people have: they want the awesomeness of Test::Class, but the modern OO facilities of Moose.
Test::Class::Moose isn't for testing Moose classes, it's for testing anything you would have previously used Test::Class for, except that now you get Moosy (Moosee? Moosey?) goodness to go with it. I'll be attending YAPC::NA 2013 in Austin and I've pitched a Test::Class::Moose talk and, even if it doesn't get accepted, I figured I should at least write the slides. One of my most common uses cases (and and itch I always rescratch whenever I use Test::Class, Test::Class::Most or, now, Test::Class::Moose) caused one of my slides to have too much code.
So now it's fixed, released to the CPAN, and available for everyone to tell me it's "too magical" (a complaint I've heard in the past). Here's the problem and how I solved it.
Perlnews.org has been going for about 2 years now and keeps going well, but we want to ramp it up... a bit... (we want a couple of stories a week really). We are now feeds www.perl.org homepage as well.
So if you have a big event in your Perl project, new major release, or you hear of Perl being used for something that warrants publicity - please let us know.
http://perlnews.org/about/ lets you know about our rough submission policy - but if in doubt please submit and we'll let you know from there.
We are also on the look out for editors - let us know if you are interested.
Slowly, but the Perl Maven site is gaining popularity among people looking for information about Perl. I'll write about the stats later.
Last week Felipe Leprevost asked me if I'd agree for him to translate the Perl Tutorial to Portuguese. I was happy to say yes as I believe having a decent introduction to Perl in your native language can make a big difference, even if later on you'll need to improve your English.
He made the translation and we already made the first article public: Instalando o Perl, imprimindo "Olá Mundo".
There were also several people who offered their help. This is awesome!
Thanks Felipe for your initiative!
The source of Perl Maven on Github
In order to make it easier for us to collaborate I set up Github repository for the Brazilian Portuguese files, and I also made the source of Perl 5 Maven available to make it easier for him to fetch the original English sources. The format is the horrific result of combining PODish with XMLish.
One of the most challenging aspects of website design is presenting information to users in such a way that appeals to the majority. But when advertisers are also part of the equation, a balance needs to be made between the end user's experience and the needs of advertisers. Because of this, we decided to use Perl for our image rotation needs.
Although in the past we used Perl on a variety of websites to rotate images, this task was a bit more complex. We needed a lightweight solution to serve rotated banners to tens of thousands of visitors daily. With an overhaul of our previous rotation code, now written for 5.16.3, we were able to develop a server friendly image rotator that works well in a shared hosting environment.
Again a lovely bike ride to the venue, this time with some air pressure trouble in the front wheel. Then a great day with talks, chatting and again excellent food.
chris fedde asks How do we know when a CORE module is deprecated?. I don't know what was there originally, but it set off a lot of fingerpointing and posturing, and nobody answered the Perl question for the rest of the universe that finds it that post through Google!
As of Perl 5.12, deprecated features and modules warn the user as they're used.
There's the deprecate module which provides this warning. A module is deprecated for at least one cycle before it's actually removed. That doesn't help much if you skip over several perl versions.
And, as one commenter, mentioned, the fallback is for people to read the perldelta pages to find the changes between the version they are using and the version they want to upgrade too.
The information is out there, and as programmers, we have everything we need to play with it however we like. :)
I have mentioned before how much I like CPANtesters! Here is another story.
Yesterday I got an email from them listing a number of failures from Galileo, my CMS. I had recently pushed some bugfix releases, but it had some new, and as yet unused code and tests for that code in it. The tests passed on all my Linux systems, so I wasn’t worried about the release. Yet the failures came in. Some on Linux, some on other platforms, but not all the tests were failures and I couldn’t figure out a pattern. CPANtesters put me on to a problem but for this I needed faster results.
I've spent the last year or so of my daily commute building a social bookmarking site that for me has now replaced my use of delicious, and I think offers more than other services such as diigo, Google Bookmarks et al.
The original site at bkmrx.com was mainly built in PHP & MySQL, however since finishing that site I've taught myself how to use Mojolicious, and did a fairly comprehensive rewrite of the website into a Mojolicious and MongoDB stack. As such it's not a full replication of the features available on bkmrx.com (see the about page for a comparison).
Now with a new job on the horizon, I want to spend less time on building out a better bookmarking service, but at the same time don't want it to stagnate.
To that end, I've released the code for bkmrx.org on Github, and uploaded a live version of the site at bkmrx.org.
This is the first time I've released a relatively big project onto Github and since I'm by no means a full time developer, go easy on me :) However I'm hoping there are others in the Perl community who might be interested in contributing to, forking or otherwise using the code.
You can read more details on my blog if you're interested.
