Thanks a lot. Le ven. 13 févr. 2026 à 21:15, Richard O'Keefe <[email protected]> a écrit :
> If you want to learn cdunctional programming, you’ve been given some > wxcellent advice. R is not the best place for that. If you want to become > excellent at R programming, functional programming is one tool in the box. > > FP has a few key ideas. Working with immutable data is one. Functions as > values is one. And composition is a very important one. Designing > computations so that they are built out of reusable pieces that can be > combined using a handful of combinators. > > A lot of R code, some of it extremely useful, has been written by people > who ae expert statisticians but poor programmers. A lot of useful R > packages have been designed by people who weren’t thinking in terms of > making reusable pieces that could be easily combined. Functions with > oodles of optional keyword parameters. Interfaces with obvious gaps. > > So don’t just think about rewriting code. Think about redesigning whole > interfaces. > > On Sat, 14 Feb 2026 at 8:13 AM, Ivan Krylov via R-help < > [email protected]> wrote: > >> Dear Saad Lamjadli, >> >> В Thu, 12 Feb 2026 13:01:26 +0100 >> SAAD LAMJADLI <[email protected]> пишет: >> >> > I have analyzed and rewritten dozens of base R functions, and this >> > process has been very instructive. However, I still find it difficult >> > to write more complex and robust functions on my own. >> > >> > I would be very grateful for any advice you might have — whether in >> > the form of recommended books, learning strategies, advanced >> > resources, or suggestions for progressing toward more advanced >> > functional programming skills in R. >> >> If you're solely focused on functional programming, try Structure and >> Implementation of Computer Programs [1] or How to Design Programs [2], >> but be ready to learn some theory and an unfamiliar language (Scheme). >> For purely functional programming, you can try learning OCaml or >> Haskell, but it's not a task for the faint of heart. Once you're done, >> you'll be saying things like "a monad is just a monoid in the category >> of endofunctors" and not understanding why people are giving you >> strange looks in return. >> >> If you'd like to make your programs more complex and robust, you need >> to study software architecture. The Architecture of Open Source >> Applications [3] is a free resource on the topic. Clean Architecture by >> Robert C. Martin is a more fundamental book, very much worth obtaining. >> >> As for learning software development in R... R is an environment where >> you can interact with data; it only moonlights as a programming >> language when you develop packages in it. To this end you can read >> Extending R by John M. Chambers, Writing R Extensions >> (RShowDoc("R-exts")) and many other free resources on the Web. >> >> Finally, Functional Programming by Thomas Mailund predates the native R >> pipe (|>) and the currently experimental tail call support (Tailcall(), >> Exec()), so it's probably worth spending some time to learn the new >> tools as well. >> >> -- >> Best regards, >> Ivan >> >> [1] >> >> https://mitp-content-server.mit.edu/books/content/sectbyfn/books_pres_0/6515/sicp.zip/index.html >> >> [2] >> https://htdp.org/2025-12-27/Book/index.html >> >> [3] >> https://aosabook.org/en/ >> >> ______________________________________________ >> [email protected] mailing list -- To UNSUBSCRIBE and more, see >> https://stat.ethz.ch/mailman/listinfo/r-help >> PLEASE do read the posting guide >> https://www.R-project.org/posting-guide.html >> and provide commented, minimal, self-contained, reproducible code. >> > [[alternative HTML version deleted]] ______________________________________________ [email protected] mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide https://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
