14:54 <snusnu> yo mbj

14:54 <snusnu> sorry for not getting back earlier, i was too busy

15:03 <mbj> snusnu: no problem.

16:36 <dkubb> good morning

16:49 <jgaskins> Good mornin

16:56 <mbj> dkubb: hi

16:56 <mbj> jgaskins: hi

16:57 <jgaskins> mbj: howdy

17:04 <dkubb> mbj: there's one thing I dislike about racc, how it names the variables inside the actions

17:05 <dkubb> for example, in https://github.com/dkubb/sql/pull/12/files#diff-8cc3aad00565b46210b90b3c5983ee27R107 I have to refer to the second position element using val[1]. why can't I just use a local variable named "correlation_name"

17:05 <dkubb> I think that would clean up so much of the grammar

17:05 <dkubb> at least it might make it easier to read in some cases

17:36 <mbj> dkubb: racc is IMHO pure ruby.

17:36 <mbj> dkubb: We should be able to change it?

17:37 <dkubb> mbj: I suppose I could, but I don't know if I want to go down a rabbit hole. I suppose I could suggest it as a feature request though

17:37 <mbj> dkubb: Yeah

17:38 <dkubb> sometimes using the array is still useful, like where you want to "slurp" in all the arguments at once, but most of the time you want to refer to specific positions, and you have the names right there in the grammar. it makes sense to me when the ruby code is generated you could add some local variables that reference the array positions directly

17:38 <mbj> dkubb: Actually the ruby behind racc looks "simple"

17:39 <mbj> dkubb: Its really nice their grammar parser is implemented via racc himself :D

17:39 <dkubb> I bet it would be a relatively simple change, but I don't want to get into it until I'm nearly done with the parser and have every statement I can generate round-tripping

17:39 <mbj> dkubb: Sure.