Hi Nikhil, > Hello Ondrej, > > Thank you for showing interest in my project. It will be an honor to > work with you on this project. > > I would love to include parsing Fortran in the proposal. I would like > to help as much as I can. Sympy has made doing mathematics so much > convenient, whatever I can do to make it more convenient for others, I > will try my best. > > There was a dicussion i saw in a thread about being able to understand > simple instructions like 'integrate sin x over x' and convert them to > sympy syntax. That will be a long and complex task. But this project > can be extended further after fortran and C to understand natural > language. > But, for now first i would like to focus on Fortran and C first. I'll > read the clang and Lfortran documentation asap and start working on a > plan. > > I would also like to thank you for creating sympy. it made doing > complex mathematics so much better. Instead of just looking at some mix > of keywords and operators which is very hard to understand, we can > actually see and understand what we're doing. > > I'll get to work now
Perfect. I am glad you found SymPy useful. For the GSoC, also don't forget the patch requirement and send one or more patches to SymPy. To clarify my motivations regarding LFortran: one or two students working on it during GSoC will allow me to pull it off. And I talked with Aaron and as long as the projects help both LFortran and SymPy, it's ok to to do it as part of SymPy's GSoC. What I am trying to achieve with LFortran is to have an interactive environment for numerical scientific computing like Python, Julia or MATLAB. Of these projects, it's closest to Julia, also in terms of the motivation (fixing the so-called two language problem of coding in Python for interactivity, and then rewriting to C++: instead just do everything in Julia... or Fortran) as well as the technology of interactive compiling. But instead of inventing a new language, I am simply using Fortran, just making it interactive. Fortran as a language is actually really nice for array oriented scientific computing --- if it had nice modern tooling supporting it, which is what I am trying to do. While doing it, I am also designing the Fortran compiler to be nicely designed as a library with a nice Python API. SymPy does this to symbolic mathematics --- instead of using some symbolic manipulation software as a black box, it nicely gives you access to the symbolic tree as a Python library and you can do things with it. With LFortran I am trying to do the same with Fortran, instead of using the compiler as a black box, it nicely exposes Fortran as an AST and ASR trees/representations, allowing people to write tools on top. I open sourced it just a few days ago, and I am getting very positive feedback on it, e.g.: https://twitter.com/DegenerateConic/status/1109930553069973516 http://degenerateconic.com/fortran-and-llvm/ and the GitLab repo has 14 stars, the GitHub repo has 26 stars, that's the largest rate of increase of any of my open source projects that I open sourced. So I have no doubts a lot of people want this and will use it, we just need to develop it enough. Ondrej -- You received this message because you are subscribed to the Google Groups "sympy" group. To unsubscribe from this group and stop receiving emails from it, send an email to sympy+unsubscr...@googlegroups.com. To post to this group, send email to sympy@googlegroups.com. Visit this group at https://groups.google.com/group/sympy. To view this discussion on the web visit https://groups.google.com/d/msgid/sympy/9bc4a7dc-2590-4b3b-bd14-d765189767a6%40www.fastmail.com. For more options, visit https://groups.google.com/d/optout.
