Perl 6 Hackathon in Oslo: Report From The Second Day

Second day of the Perl 6 Patterns Hackathon. My plans to get the rest of placeholders and prepared statements working in the Postgresql backend for MiniDBI succeed about 10 minutes after midnight. I just wanted to give them a very quick try before going to bed, and was successful. Then I went to sleep.

It was night, and it was morning. Second day.

Next I wrote an SQLite backend for MiniDBI. It blocked on missing features in our native call infrastructure, on which arnsholt worked in parallel. So I haven't had a chance to try the SQLite backend yet. It probably requires some substantial amount of work before it will run, but at least it compiles.

I also investigated prepared statements and placeholders for the mysql backend. This is much less straight forward, because it requires filling in members of structs, not just function calls. This by itself wouldn't be much a problem, our native call infrastructure supports that. The problem is that it's a struct of mixed "private" and "public" members, so modelling the structure in Perl 6 requires modeling private data of the mysql client library. While possible, I don't find it desirable, because it is rather fragile.

Another notable event was the hacking dojo, where about 8 of us collaborated to write a roman numeral conversion, using pair programming, and fixed cycles of first writing a failing test, then getting it to run in the simplest possible way, and finally refactoring it. It was quite an interesting and fun experience.

I spent much of the rest of the hackathon discussing things. For example Patrick Michaud gave a quick walk through of how lists and related types are implemented and iterated in Rakudo.

In the evening we had very tasty Vietnamese food, and generally a good time.

Again it was a very productive and enjoyable day, and I'm very grateful for being invited to the Hackathon.