Wednesday 12 December 2007

Didn't make it to FooMongers today - was having plenty of fun getting the stuff I have written for the NPD web app ready for deployment. Tests must work! Excel spreadsheets must be able to download! Refactoring must take place when complexity scores go above 22.

I can see Test::Perl::Critic being a particualr favourite of mine, not! I have to admit that having now started to write with standards in mind, the stuff I produce is much more readable, and maintainable, but those C-style for loops are something that I am going to have to get out of writing.

Anyway, we are ready to deploy tomorrow with the Sample Summary. This was nice to do, and a good starting place for me, as it got me working with both the views. templates and model. Wrapping up the business end of determining Good/Bad samples within the model itself so this exports a method to determine if it is Good or Bad - believe it or not.

The best bit (joke) was getting it to be tested completely, which involved a lot of circular referencing in order to ensure that the mock data is always used. However, we got through that, nd need to definitely remember that for next time.

The next trick was merging branch to trunk in svn. 3 conflicts! However, a quick look on svnbook.red-bean.com (http://svnbook.red-bean.com/en/1.1/ch03s05.html) quickly dispelled my panic and all was happy. A quick make test in the trunk before committing ensured that all was still happy, and we deployed to the intweb development area. This all looks hunky dory, so tomorrow we go live AAHHHH!

Shame about FooMongers though today. Hope I didn't miss anything to groundbreaking.

Wednesday 5 December 2007

Lunchtime - FooMongers

We have a small group on site who meet Tuesday and Wednesday lunchtimes to discuss programming stuff.
(This includes people from the EBI and the Wellcome Trust Sanger Institute).

Today, we spent a bit of time debating the pros and cons of Bluetooth technology, and then Rob and I answered a few peoples queries on Ruby on Rails. I also pushed ClearPress (see cpan.org).

Rob mentioned MERB? as something I might lik to look into at the end of the session.

A good lunchtime discussion was had by all. More of the same next week I hope.

First Post

So, here I am. A fully fledged Software Developer. Allegedly.

I have just deployed my first Rails App this morning. An application to store electronic notes about a project in DNA sequencing, and also find out about the life of that project.

It's been a very interesting ride. First learning OO programming in PERL, then learning Ruby and RoRails,
then breaking out of RoR's generic framework to connect to legacy databases (one Oracle, one a collection of many MySQL databases), then adding LDAP authentication and a mailer.

The most interesting part (apart from learning RoR in the first place) though was probably the smallest. I had to connect a small Tk Application written in PERL to this to send information from a form to store in it and send email and a response of success/failure. This essentially brought me back full circle to learning to use PERL with LWP and XML.

A thoroughly enjoyable ride on the Train. And this, all whilst I was a DNA sequencer.

So as I said at the start, now I am a full fledged software developer. It says so on my shiny new contract.
I'm back doing OO Perl (this time without Class::Std) using ClearPress (see cpan.org) which my new boss developed. The nice thing about his framework is that he has made it very Rubyesque in the views (even though he claims no knowledge of Ruby).

And what am I doing it for - New Sequencing Technology - Illumina machines. This is the forefront of modern DNA sequencing, and here I am at the start of it. AAAHHHHH!

As Bob the Builder says': Let's go team!