CPAN Testers is only made possible with the support of our sponsors.
For more information on sponsoring, please visit the I CPAN Testers website.

Upgrade Notice

The CPAN Testers Blog site has been upgraded since you last accessed the site. Please press the F5 key or CTRL-R to refresh your browser cache to use the latest javascript and CSS files.

News & Views

This month's summary has quite a bit of news and several thank yous.

Firstly, last month we had both the QA Hackathon in Lancaster and the Satellite event in Tokyo. A big thank you to all the sponsors (cPanel, Dijkmat, Dyn, Eligo, Evozon, $foo, Shadowcat Systems Limited, Enlightened Perl Organisation and Mongueurs de Perl). Both events were a big success for many reasons, amd thanks to all the attendees who have improved, released and talked about code, testing and the toolchain during the event. CPAN Testers is a small recipient to all the efforts, but they do help everyone working in and on Perl.

For CPAN Testers specifically, the weekend prove quite productive. The report parser from the Metabase to the cpanstats database was improved, the CPAN Testers Reports website builder was improved and as a result reports are added to distro pages now typically within a few hours for popular distros. The authors pages are now built less frequently, as these are both viewed less frequently, and more subject to change over the course of a few hours. The end result is that we are speeding up the processing. You can read more of David Golden's, Ricardo Signes and my own experiences on our blogs, together with the Lancaster Consensus.

As mentioned previously, the SQLite database is a big strain on resources and now seems to be downloaded only by web crawlers. As a result, the maintenance of the SQLite database for the cpanstats database has been halted. It may be reinstated once we have another server to build it on, but for now we would encourage you to investigate using the Reports API to update your own copy, rather than downloading the full DB every time you need it. If this proves a problem for anyone, please let me know and I'll see what we can do.

Garu was very productive during the QA Hackathon, and worked further on both a CPAN Testers client for cpanm, but also push his work on a common client tool, that he began at the QA Hackathon in Paris last year. The common client, allows all the smoker clients to use the same reporting mechanism, regardless of the installer, but perhaps more importantly is the start of a true structured data report, rather than the simple plain text version we have been using for all these years. The idea of moving to the Metabase was so that we could store reports in such a way that specific elements of a report could be singled out and investigated and searched for. Once the Common Client is integrated with all the smoker clients, this will mean we can finally start looking at better analysis tools and allow authors, testers and users to interrogate reports for the criteria that interest them. The fruits of Garu's work were final minutes of the QA Hackathon, when got his report pushed all the way through the CPAN Testers eco-system and the report appeared on the Reports website. A landmark moment. The following week, Garu's App::cpanminus::reporter was also the weekly winner of the MetaCPAN up-vote summary.

Just after the QA Hackathon we finally topped 30 million reports. It is quite a phenomenal achievement really, and it will be a long time before any other language library repository ever comes close. I would be very interested to see efforts in Python or Ruby to make something like CPAN Testers. On occasion I've even thought about attempting to make something myself, but I'll have to wait until I have considerably more time on my hands than I do currently. However, if anyone does feel the desire to create an equivalent to CPAN Testers, please let me know.

On the mailing list, Diab Jerius asked whether 'Would it be possible to provide an estimate for turn-around times at'. David Golden explained some of the problems we're currently experiencing with Amazon's SimpleDB and how we're planning to move to MongoDB soon. We have some servers on the way, but it will take us a while before they're fully operational.

I did mean to mention a blog post in the last summary, but accidentally missed it in my notes. So to make amends, thanks to Joel Berger who posted I love CPANtesters and Travis-CI. Travis-CI is a system that close resembles the PITA idea that Adam Kennedy had many years ago. More recently Alceu Rodrigues de Freitas Junior posted The CPAN Testers game , where he highlighted how easy it can be to set yourself up as a CPAN Testers. Regarding the title of the post, he moves to look at the way that the Statistics site has helped to add a little healthy competition to being a CPAN Tester. Although BinGOs will take some beating as a whole, there are still plenty of platforms Chris has yet to make a dent with.

As always a big thank you to everyone who has help to make CPAN Testers what it is, and many thanks to those who have contributed to The CPAN Testers Fund. More updates about contributors coming soon.

<< June 2013 (1) April 2013 (3) >>