2007Aug15SPDojo

**Participants:** [DaniloSato](/people/DaniloSato) ,

FabricioSousa , HugoCorbucci , JacquelineMarchetti , JulianMonteiro , MarcioSantos , MarianaBravo , PauloCheque and RafaelBarroso

Scribe: DaniloSato

Kata: Poker Hands in Python (an slightly easier version of KataTexasHoldEm )

Schedule

: 20:00 ~ 20:20 Discussion of the Kata

The problem was already chosen during the week, so the group discussed a little about the Poker rules and briefly discussed the approach to solve it.

20:20 ~ 21:50 Coding Session

We decided to tackle this problem in a Randori style, coding in Python. We also decided to avoid an object-oriented approach, at least in the beginning. As a learning exercise, we wanted to see it asking for the objects instead of designing a model before starting to code. As the last time we tried to solve the KataBowling , we started defining an acceptance level test, and that hasn’t helped us evolve the code in small steps. Once we decided to write smaller tests (for parsing a card, comparing a card, etc.) the session started to improve. Inspite of that, we didn’t have time to finish the problem. The partial solution is available at groups.google.com/group/dojo_sp/files.

21:50 ~ 22:15 Retrospective

We discussed some issues with the programming environment. We had some problems with the tools and the keyboard shortcuts (which significantly slowed us down). Also, we decided that the code ended up mixing a procedural and an object-oriented approach. If we had more time, we would probably come up with a new object (a poker Hand) and this problem would not be so evident. We found that starting with an acceptance level test makes it more difficult to test-drive our code in baby steps. Some of the actions for the next meeting were: trying to solve the same problem before the meeting and starting it over again on the next meeting; setup a different environment (Linux, Eclipse, Pydev and a standard keyboard); use a TODO list to track the next steps (as proposed by Kent Beck on his TDD book).