jerome.martin.dev
2018-6-7 07:24:34

@greg Thank you :heart:


jerome.martin.dev
2018-6-7 07:26:25

also, I didn’t thought about all-from-out, it’s definitely better than doing struct-out mutiple times (which I think is part of the issue here).


jerome.martin.dev
2018-6-7 07:28:15

and I didn’t knew about #lang reprovide, nice to see someone had the idea to simplify this process :slightly_smiling_face:


pocmatos
2018-6-7 10:47:20

I have just reported https://github.com/racket/racket/issues/2122 however I am now unsure if this is a bug at all or just a hole on my knowledge of errortrace.


pocmatos
2018-6-7 10:48:25

If you look at the bug, the whole issue boils down to the fact that racket -i -l errortrace is not the same as setting current-compile to errortrace-compile-handler and dynamically requiring the target file. Does anybody know what’s going on here?


pocmatos
2018-6-7 10:57:00

Opps, I also need to set the error-display-handler. :slightly_smiling_face:


leif
2018-6-7 16:42:36

@jeapostrophe How hard would it be to add a flag to raco pkg install to only install the dependencies for a package, and not the package itself.


leif
2018-6-7 16:43:13

I imagine this being useful if I have an application/paper I want to use the info.rkt file to manage dependencies for me, but I don’t want to actually install it as a collection.


ben
2018-6-7 18:35:34

@pnwamk @samth do you know of examples using TR intersection types?


ben
2018-6-7 18:35:58

Fabian M. just asked me — I can forward the email or/and cc you on the reply


pnwamk
2018-6-7 19:34:47

@ben case-> is an intersection type that we reason incompletely about and is used all over the place. The more general intersections (i.e. ) were originally added to simplify Typed Racket’s implementation… off the top of my head I’m not aware of any “real world” programs using them (TBH I haven’t looked), but some searching on Github may turn some up.


ben
2018-6-7 19:37:22

ok, yeah I just meant