Perl Maven - September 2014

With September the new school-year has started which gave me more time to write articles and to record screencasts. This extra free time (to work :), show in much better productivitiy. Albeit I am still frustrated on days when I feel I did not make enough progress.

As I can see I published 18 new Perl Maven posts and managed to some other stuff.

Perl Maven Pro

I've recorded screencasts for several earlier articles:

• Logging with Log4perl the easy way
• Logging in modules with Log4perl the easy way
• What is the status of the current test script?

and went on with the Test automation series resulting in these articles and screencasts:

• Test without a Plan
• Separating test data from test code
• prove, the harness
• Moving over to Test::More
• Test diagnostic messages using diag, note, and explain
• TODO - testing a bug or a future feature
• Using like to test without exact values
• Testing timeout with cmp_ok

Perl Maven free

I had the time to publish a number of articles related to MetaCPAN and some generic Perl articles:

• Check several regexes on many strings
• Some MetaCPAN advanced search tricks using prefixes
• What you need to know about CPAN
• How to get the index of specific element (value) of an array?
• How to sort a hash of hashes by value?
• Using Travis-CI and installing Geo::IP on Linux and OSX
• Cloud automation at Digital Ocean using Perl
• DWIM Perl for Linux; $^X vs $Config{perlpath}; Relocatable Perl; Test::Differences
• CPAN, mcpan, MetaCPAN
• Short-circuit in boolean expressions

EduMaven

I kept publishing pages on the EduMaven site albeit at a much slower rate. By now there are 399 Programming Perl pages, 389 Programming Python pages, 324 Test Automation Using Perl pages, and a bunch of other slides. A total of 1390 pages.

Perl Weekly

Nothing special happened. The the weekly growth rate of the subscribers is les than 10.

Padre, the Perl IDE

It wasn't my doing, but the source code repository of Padre has moved to GitHub. This is awesome. I used the opportunity to factor out some code into a separate module, and some other people have also started to do stuff. This might provide an opportunity to restart development.

Once thing I have been planning for a while alrady was moving many of the wiki pages to be part of the real Padre site where they belong. I've started to do that but got frustrated by the manual process. Became the maintainer of Text::Trac, but did not have time to improve it to be able to conevrt the pages automatically. I found a few other, more interesting projects.

Code::Maven

I long time ago I started a project called CPAN::Digger, that was supposed to analyze the source code of all the CPAN modules and even Perl code that is not on CPAN. Later I started another project called PyDigger, written in Python, that had the same purpose, but to analyzes Python source code. Both were stopped due to lack of time.

This time I hope it will be different. The goal is also different. The Code::Maven site was launched on 16th September, and the few days after it I spent improving it a lot, but have not got to the real interesting part yet. Then I got distracted. Specifically I got fed-up that I have to brew perl on every server I use and decided to restart building DWIM Perl for Linux. Now that DWIM Perl is quite ready I should go back to working on the Code::Maven project.

DWIM Perl for Linux

I spent almost a week building DWIM Perl for Linux. So far I've release 9 revision of it with a growing number of modules and the 10th should be ready soon. The project went quite well, but I am a quite frustrated as the feedback was really minimal. I think I've heard only from 3 or 4 people who have tried it or liked the idea. Even though I wrote about it on blogs.perl.org, got it onto Perl News, and posted on Reddit too.

I know I should show it in use more, but first I need to make sure I can use it on all of my servers.

Open Source

I kept patching MetaCPAN. It is fun. I learn quite a few things, and once my pull-requests are accepted this can have an immediate impact to the 160,000 people visiting the MetaCPAN site every month. As I think people need to be introduced to the new features of MetaCPAN I even wrote about it. Oh and I became one of the top contributors of MetaCPAN.

In August I started to send pull-requests to CPAN authors who did not have link to VCS. At the beginning of the month I kept doing it, but after a while I got tired of this. I think there are still a few pull-requests I sent that are waiting for people to integrate.