00:12 <ski> schlick : it's an upper case gamma, it's prolly just looking somewhat different in your font

00:13 <ski> the context usually contains types for all free variables that could possibly be used in the expression (in some type systems it can contain some other things, too, not sure if that's the case here)

00:15 <ski> x : int,y : int,z : int |- (if x > y then x else y) : int

00:15 <ski> 'x : int,y : int,z : int' is your context / Gamma, there

00:16 <ski> so Gamma is usually a (comma-separated) list of typings of variables

00:16 <Smerdyakov> P.S., all humans are invited to join #coq.

01:09 <JosephRivers> Suppose that you wanted to define two functions f and g where g was used in the definition of f and f was used in the definition of g. Would there be a way to do this in ocaml?

01:13 <ski> that is called mutually recursive functions

01:13 <ski> let rec f ... = ...

01:13 <ski> and g ... = ...

01:14 <JosephRivers> Okay, I'll give that a try. Thanks.

20:09 <ulfdoz> Ehm, how to do export some additional types and functions, made in the header of a .mly-file? ocamlyacc generates an .mli, but I find no way to get automatically some additional definitions from the header included.

20:11 <ulfdoz> btw. am I the only one, who dislikes header-files and variants of it?

20:14 <zmdkrbou> here, you're probably the only one yes

20:14 <zmdkrbou> but code in a one huge file if you like to be dirty :)

20:16 <zmdkrbou> and header and trailer sections in mly files are directly included in the resulting ml and mli files

20:18 <ulfdoz> But for some reason I cannot refer to the type of my parse tree. I coded some variant types, to constrain the parse tree to hopefully only valid ones.

20:20 <ulfdoz> damn, tricked myself. a typo. :(

20:25 <ulfdoz> boah, and I'm out of practice.