00:22 <shawnw> Am I weird for wanting to reimplement the csvkit bundle of utilities in racket from the original python?

00:25 <shawnw> (curry comes from Haskell Curry)

00:26 <shawnw> (but Moses Schönfinkel was first)

03:09 <norm_tsp> i'm pretty clueless about racket, could someone please tell me if my interpretation of foldl functioning is wrong

03:09 <norm_tsp> http://pasterack.org/pastes/10933

04:59 <norm_tsp> i suspect that i'm wrong, but i can't pin point the order of operations and evaluations here

09:51 <norm_tsp> not a lot of activity around here

09:59 <bremner> norm_tsp: not too much. Alas most racket people seem to like shiny web things.

10:00 <bremner> norm_tsp: did you have a racket question?

10:20 <norm_tsp> yes, i will repost it

10:20 <norm_tsp> <norm_tsp> i'm pretty clueless about racket, could someone please tell me if my interpretation of foldl functioning is wrong > http://pasterack.org/pastes/10933

10:21 <norm_tsp> it probably is, but this construct is pretty tricky for people unfamiliar with functional style

10:27 <bremner> norm_tsp: is it intended that your initial value is a function?

10:27 <norm_tsp> yes

10:28 <norm_tsp> i'm not sure why

10:28 <norm_tsp> with i understood how this works

10:28 <norm_tsp> wish*

10:28 <bremner> right, it is pretty cryptic.

10:33 <norm_tsp> looks like a good interview question

10:37 <bremner> norm_tsp: are you by chance studying continuation passing style? because that is what it looks like to me

10:37 <norm_tsp> it's just one of the examples of racket magic i came across, but now i want to understand it

10:37 <norm_tsp> it's above my current understanding for sure

10:38 <norm_tsp> i will get back to it once i reach continuations theory

10:39 <bremner> norm_tsp: you can trace the execution of brainteaser in the debugger, maybe that helps

10:39 <norm_tsp> you mean the "compiled" version?

10:39 <bremner> the foldl happens at "compile" time, and generates a complicated function

10:39 <bremner> yeah

10:39 <norm_tsp> i can't read that either

10:39 <bremner> just break at the call to brainteaser

10:39 <bremner> (pause, in the jargon of drracket)

10:41 <bremner> replacing + with cons makes the order of operations clear

10:42 <norm_tsp> i guess something like that might work, maybe on a smaller list too

10:43 <norm_tsp> application of g function in particular is puzzling to me, it certainly does something...

10:44 <bremner> g is like a continuation in that it receives the result and does the remaining calculation

10:44 <norm_tsp> that sounds plausible

10:44 <bremner> if you replace + with cons, you get '((((5 . 4) . 3) . 2) . 1)

10:46 <bremner> so things kindof happen backwards from how you might expect with foldl

10:48 <norm_tsp> like i suspected, this functional style is beyond me at the moment

10:49 <norm_tsp> gotta start small

10:53 <norm_tsp> at least my expanded code seems to be correct

14:28 <samth> cpli: quasisyntax/loc is like quasiquote, but with a few differences. first, it constructs a syntax object, not just a list. second, the unquote is spelled #, and third syntax objects have source location information and the first form tells what location to use