Home

Building SciPy & ImpliedDoLoops

Date: July 23, 2023.

Hii there! Welcome to blog of week 10. This week I continued implementation of ImpliedDoLoops made a decent progress over it and also somehow manage to get SciPy Specfun build and run tests with LFortran.

Let's pick up ImpliedDoLoops from where we left, we had initial implementation of implied do loops ready, but it lacked support for nested implied do loops, accessing array elements, variables casting, etc. So, I started workin on it and provided support for these. Later, I realised that exisiting implementation of implied do loops was working fine for few cases that latest implementation was not able to handle. So, I decided to keep both implementations and first apply exisiting approach and then the latest one. This worked fine and I was able to pass all tests. You can have a look at it PR: 1993. But, I am not satisfied with this approach as it is not a clean implementation and also it is not very efficient. So, I am still working on it and will try to come up with a better approach.

Secondly, with help from Ondřej Čertík, I was able to get SciPy Specfun build and run tests with LFortran. We had to comment and remove few tests as they were not compatible with LFortran. You can track the progress at Issue: 1979. We planned to setup CI for Specfun in this week, but I was unable to do that.

This week I was unable to give much time to GSoC due to some personal reasons, worked only for 12.5 hours. For next week I will try to setup CI for Specfun and also work on ImpliedDoLoops.

I would like to thank Ondřej Čertík for taking out time and helping me when I was stuck. Looking forward to many more commits.