tgbugs
2021-2-19 09:15:02

reporting that this seems to be working as expected with only one issue remaining, which is that the generation of /usr/bin/gracketcs somehow seems to ignore the setting from --collectsdir (/usr/bin/racketcs seems to be ok?) otherwise the only note is that install-both doesn’t install drracketbc and doesn’t install gracketbc so I switched to running install-cs, install-bc, and install-cgc separately (and in that order)


capfredf
2021-2-19 19:22:43

I ran into a kind of weird problem. #lang racket (define-generics some-hi (gen-hi some-hi b)) (struct foo () #:methods gen:some-hi [(define (gen-hi f b) 10)]) Apparently define-generics is unbound without require racket/generic . If I run this program in DrRacket, I will get an expected error. However If I run this program at the command line, i.e. “racket -t this-file.rkt”, the error message will be something like : > struct: the first argument to the #:methods specification is not a name for a generic interface If I deleted the definition of struct foo, then racket -t this-file.rkt would also report define-generics was unbound


sorawee
2021-2-19 19:31:14

For the original program posted above, I consistently get the error:

struct: the first argument to the #:methods specification is not a name for a generic interface in: gen:some-hi


capfredf
2021-2-19 19:31:27

In DrRacket?


sorawee
2021-2-19 19:31:39

yes


shu--hung
2021-2-19 19:31:46

errortrace?


sorawee
2021-2-19 19:32:31

I tried both (with errortrace on and off). Doesn’t make a difference


sorawee
2021-2-19 19:33:06

Are you using Racket CS or BC?


shu--hung
2021-2-19 19:33:16

I’m consistently getting struct: the first ... message both with and without errortrace

Welcome to DrRacket, version 8.0.0.6 [cs]. Language: racket [custom]; memory limit: 512 MB.


capfredf
2021-2-19 19:33:23

Let me try again.


capfredf
2021-2-19 19:40:52

Thank you guys! I must have had an illusion


sorawee
2021-2-19 19:45:17

:mage::magic_wand: