GSoC Week 3: Formalizing design, polishing code

This week was mostly spent in polishing the design, writing mock sessions and adding tests to the code I wrote last week. Thanks to Jason’s constant encouragement towards objective- and test- driven code development, I now have a good idea of what I need to do in the coming week. If the meeting with Jason goes well and we are on the same page, I will soon complete the current WIP PR that is at this link, and then move on to implementing CoordSysRect class (just written the API for now).

What I learnt this week-

Before coding,

1. First write out as many example use cases of the software, as possible. In this case, the quality is more important than the quantity. The use cases should sufficiently touch upon all aspects of working required from the software that you are about to code. The interactions, on a conceptual and UI-level, should be as crytal-clear as needed.

2. Draw out the high-level API -> user-visible attributes(or better, ‘property’s in Python) and methods/functions. These will help you in defining all the ‘internal’ matter and algorithms succinctly.

3. While you are finalizing the API design, start drawing up a set of incremental objectives you would need to complete for building the overall software. Make sure that the objectives make sense chronologically, and reflect the priority and complexity level of the objectives correctly. This is basically to acknowledge that it is not possible to build and entire software together, but rather its better to code it module-by-module – with sufficuent testing at regular intervals, ofcourse.

The follow the timeline you have decided, and start coding.

Respecting this, I will start working on the code full-time next week onwards, so hope to give more detailed reports then. Have a great week :-).

Leave a comment