Data analysis in Python using the concatenative paradigm

#1

Post by punkbrwstr » Thu, Oct 24, 2019, 7:04 PM UTC

Data analysis in Python using the concatenative paradigm
github.com

Re: Data analysis in Python using the concatenative paradigm

#2

Post by eindiran » Thu, Oct 24, 2019, 7:41 PM UTC

Very cool. OP, what is your previous exposure to concatenative programming? Have you written or used a Forth before?

One thing I noticed was that you opted for passing parameters to words directly, rather than placing them on the stack; it does seem to improve readability quite significantly. (eg `c(10.0) | some_word`)

The anonymous quotations for combinators are a nice addition as well.

Re: Data analysis in Python using the concatenative paradigm

#3

Post by punkbrwstr » Thu, Oct 24, 2019, 9:13 PM UTC

eindiran wrote: post #2

Very cool. OP, what is your previous exposure to concatenative programming? Have you written or used a Forth before? One thing I noticed was that you opted for passing parameters to words directly, rather than placing them on the stack; it does seem to improve readability quite significantly. (eg `c(10.0) | some_word`) The anonymous quotations for combinators are a nice addition as well.

Thanks! I tried keeping it all stack-based, but I really wanted default arguments. In the end the parameters felt more usable even if they are "impure". I made a bunch of iterations of a postfix calculator for time series before realizing there is a whole world of concatenative langauges. The Factor paper really blew my mind.