On 23/01/14 04:51, Mark H Weaver wrote: > Hello all, > > For a short time I liked 'cut' from SRFI-26, but I soon became > frustrated by its limitations, most notably not being able to reference > the arguments out of order or within nested expressions. I don't like > the inconsistent style that results when I use 'cut' wherever possible > and 'lambda' everywhere else. So I just stopped using it altogether. > > I prefer what Shiro Kawai did in Gauche: ^ is short for lambda, (^x ...) > is short for (lambda (x) ...), and similarly for all the letters. ^_ is > short for lambda with one ignored argument. > > http://blog.practical-scheme.net/gauche/20100428-shorter-names > > Here's a module that implements that idea, but takes it slightly > further. It exports the unary lambda shorthands described above, and > also a few non-unary ones: ^xy, ^xyz, ^ab, ^abc, ^uv. It also exports > variants that use λ instead of ^, for a nicer look if you're willing to > venture outside of ASCII. > > Finally, it exports the transformer procedure that's bound to all of the > keywords above. It's called 'short-lambda'. Bind it to any keyword > like this: > > (define-syntax ^kw short-lambda) > > and now (^kw ...) will expand to (lambda (k w) ...). > > 'short-lambda' splits the keyword into individual characters. The first > character (usually ^ or λ) is ignored. The other characters become the > variable names. Underscores are treated specially: they become gensyms. > > Thoughts? > > Mark > >
Hi, Any chance for the curried version of these? -- Mateusz K.
