Can I know if a module has been imported or not?

module-declared? could do the job, I think.

https://docs.racket-lang.org/reference/Module_Names_and_Loading.html?q=module-declared#%28def._%28%28quote._~23~25kernel%29._module-declared~3f%29%29

E.g.,

#lang racket (module-declared? 'xml) ;=> #f but

#lang racket (require xml) (module-declared? 'xml) ;=> #t

Thanks!

@dan.ml.901 What problem are you trying to solve? Usually define-match-expander is the right way to define new match forms. The prop:match-expander property is for the rare case when you want to define a name simultaneously as a match expander and as some other sort of static information carrier. That is, when the compile-time value associated with the name needs to be a single (struct) value that implements multiple interfaces.

Is a good idea to use _ptr to receive pointer function arguments? (In the Racket FFI)

I was trying (_ptr i functionType) and it is giving me errors

FWIW there is an example with _ptr here: http://prl.ccs.neu.edu/blog/2016/06/29/tutorial-racket-ffi-part-2/

Thanks, but there is not what i’m looking for. I’ll keep searching

All comes from this issue: https://github.com/hsalokor/rkt-glfw/issues/4

The glfw library uses _ptr for all function types

Instead of only _fun

I looked up: GLFWkeyfun https://www.glfw.org/docs/3.3/group__input.html#ga0192a232a41e4e82948217c8ba94fdfd

> typedef void(* GLFWkeyfun) (https://www.glfw.org/docs/3.3/group__window.html#ga3c96d80d363e67d13a41b5d1821f3242\|GLFWwindow, int, int, int, int) > void function_name(https://www.glfw.org/docs/3.3/group__window.html#ga3c96d80d363e67d13a41b5d1821f3242\|GLFWwindow window, int key, int scancode, int action, int mods)

Maybe * GLFWkeyfun was interpreted as a “pointer to a function” and then (_ptr ...) was added?

But if it works without, then I’d say just remove it.

Yes, I thought that

Thanks!

Then again, I see https://github.com/BourgondAries/rkt-glfw/blob/master/main.rkt use a similar method.

Maybe the difference is between us passing a function to the C side - and calling a C function in order to get a callback back? In the latter situation, the C side probably needs an “pointer to a cell in which store the function” aka (ptr o …).

It seems that both files are the same

Also, the definitions are not updated

GLFW have more functions

I think Im going to update it. Ill do a pull request when it is done

Is this correct? (define-cstruct _GLFWgammaramp ([red (_ptr o _ushort)] [green (_ptr o _ushort)] [blue (_ptr o _ushort)] [size _int])) red, green and blue are pointers to unsigned short

Or it should be (define-cstruct _GLFWgammaramp ([red _pointer] [green _pointer] [blue _pointer] [size _int]))

Depends on how you intend to call them, I think. The first will allocate the ushorts automatically.

That works?

Nice then

can someone open Racket (8.0, specifically) through the command line and press Ctrl-?

I don’t know what this keybinding is supposed to do, but on my machine, it core dumps

nevermind this occurs for every program

that’s new

Which os?

NixOS, bash

could just be a shell thing

or a readline bug…

JShell and Scala’s REPL behave really weird when I do this too Python, Racket, and R just abort GHCi doesn’t fail

I think this is a readline issue

Apparently Ctrl- sends SIGQUIT to the foreground process: https://stackoverflow.com/questions/19248556/why-does-pressing-ctrl-backslash-result-in-core-dump

A feature!

a question about PLTCOMPILEDROOTS: when I run PLTCOMPILEDROOTS="$(racket -e '(display (path->string (cadr (current-compiled-file-roots))))')" racket I get the following error: Welcome to Racket v8.0.0.9 [cs]. path-list-string->path-list: contract violation expected: (listof path?) given: '(same #<path:/usr/lib64/racket/compiled>) Is this expected, or should same have been discarded?

It seems that this is caused by an interaction between PLTCOMPILEDROOTS and the changes in https://github.com/racket/racket/commit/a110c58e521e6aa291895e177bd358c0cd7c623b which result in (compiled-file-roots . (same "/usr/lib64/racket/compiled")) being written to /etc/racket/config.rktd I will submit a github issue.

https://github.com/racket/racket/issues/3704

wild

I use this a lot; it’s more effective with processes that don’t respond to ctrl-c

I’m on windows 10 home, I noticed with this new version 8, when I launch DrRcket for the first time (its ok) I close it and it won’t launch again until I restart my computer. Anyone facing the same problem?

running DrRacket as Admin works

And in previous versions it works correctly, right?

yes, I have the old version installed on a different laptop (windows 10 pro). maybe I’ll reinstall it tomorrow on this computer