This comes from _fun, which define-gl expands to,

https://docs.racket-lang.org/foreign/foreign_procedures.html#%28form._%28%28lib._ffi%2Funsafe..rkt%29.__fun%29%29

The first -> is a type spec for the C function's return type and the
second -> is the output expression that is what the Racket wrapper
will return.

Jay

--
Jay McCarthy
Associate Professor @ CS @ UMass Lowell
http://jeapostrophe.github.io
Vincit qui se vincit.

On Fri, Apr 10, 2020 at 10:02 AM Hendrik Boom <[email protected]> wrote:
>
> I'm building the new opengl binding for Racket, and keep running
> into surprises.
>
> Most of them are straightforward, but tedious, but I run into
> something strange.  It's a piece of syntax in the original
> (presumably) corrent, but obsolete, binding -- specifically,
>  a strange type.
>
> I thought I should ask before blundering through with it.
>
> Here's the definition for glAreProgramsResidentNV as present in
> the existing binding:
>
>
> (define-gl glAreProgramsResidentNV 2 ((n : _int32) (programs : (_u32vector 
> i)) (residences : (_vector o _bool n)) -> (result : _bool) -> (values result 
> residences)) (->> exact-integer? u32vector? (values boolean? (vectorof 
> boolean?))) check-gl-error)
>
>
> Now normally the define-gl macro seems to take
>     a number,
>     a type of the function (containing names for the parameters
>         in the style of typed Racket's lambda arguments), and
>     another type to put in the documentation.
>
> Now the type of the function here is
> ((n : _int32) (programs : (_u32vector i)) (residences : (_vector o _bool n)) 
> -> (result : _bool) -> (values result residences))
>
> An ordinary enough looking type except that -> occurs twice.
>
> **: Is this a known Racket construct?
>
> Or is it something peculiar to define-gl, which I'll just have
> to figure out as I build the new opengl binding.  It's presumably
> related to the fact that the 'residences' parameter is for output.
>
> -- hendrik
>
>
> In case anyone is interested, here are the relevant
> machine-readable bits from the nes and the old Khronos API specifications
>
>
> The xml specfile entry for this is:
>
>         <command>
>             <proto group="Boolean"><ptype>GLboolean</ptype> 
> <name>glAreProgramsResidentNV</name></proto>
>             <param><ptype>GLsizei</ptype> <name>n</name></param>
>             <param len="n">const <ptype>GLuint</ptype> 
> *<name>programs</name></param>
>             <param group="Boolean" len="n"><ptype>GLboolean</ptype> 
> *<name>residences</name></param>
>             <glx type="vendor" opcode="1293"/>
>         </command>
>
>
> And here's the old specfile entry from the old Khronos opengl standard:
>
> AreProgramsResidentNV(n, programs, residences)
>         return          Boolean
>         param           n               SizeI in value
>         param           programs        UInt32 in array [n]
>         param           residences      Boolean out array [n]
>         category        NV_vertex_program
>         dlflags         notlistable
>         version         1.2
>         extension       soft WINSOFT NV10
>         glxflags        ignore
>         glxvendorpriv   1293
>         offset          578
>
>
> --
> You received this message because you are subscribed to the Google Groups 
> "Racket Users" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to [email protected].
> To view this discussion on the web visit 
> https://groups.google.com/d/msgid/racket-users/20200410140217.b4e5xizymwunvyu6%40topoi.pooq.com.

-- 
You received this message because you are subscribed to the Google Groups 
"Racket Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/racket-users/CAJYbDakT1AXhd7zBK%2BHQ%2B5-z6RnkoHj__xxyGAD6vDXt21mWrA%40mail.gmail.com.

Reply via email to