Posts GSoC 2019: Week 2
Post
Cancel

GSoC 2019: Week 2

The second week of official coding period has concluded. I will give you a brief summary of the week’s work.

I started this week with working on PR #16956, the PR is now passing tests. I will try to get it merged soon. Later this week, I tried to include complex -> finite in the assumptions system. I started PR #16978 over this. This PR changes the assumptions system to work over the complex -> finite clause. All the instances of use of is_complex call has to be modified. As suggested by Aaron, the is_complex call was changed with a disjunction of is_complex and is_infinite calls. Luckily for me, the usage of complex assumption was not much in the codebase 😌. Though it meant less test failures to handle, the core assumptions system (in the old system) is quite complex to understand and debug, and hence handling all the failing tests proved quite difficult. Also, I tried to work on it without including any new assumptions to the system. One important result of this is, zoo.is_complex returns False now.

Apart from that I have worked on to sync these changes in the new assumptions system as well and I will be sending PR over it soon.

For the coming week,

  • I will start to work on SAT solver and to speed it up. As suggested by Aaron, I will first look into the other unfinished PRs over this.
  • I will sync the new assumptions with the recent developments in the old system.

With enough said, these PRs made me realize how much the codebase can break with small changes in the assumptions. Also, I know this post is a little bit late(because of bad health). I have some pending work over from the last week and I will try to catch speed as soon as possible.

Trending Tags