On 09/13, Yann Orlarey wrote:
>
> > does the trick, thanks! But,
> >
> > -ID [a-zA-Z][_a-zA-Z0-9]*
> > +ID _*[a-zA-Z][_a-zA-Z0-9]*
> >
> > Why not
> >
> > ID [_a-zA-Z][_a-zA-Z0-9]*
> >
> > ?
> >
>
> That's to avoid underscore+ as valid IDs.
Got it.
Well,
Le mar. 13 sept. 2022, 15:10, Oleg Nesterov a écrit :
> On 09/13, Yann Orlarey wrote:
> >
> > That was fast ;-) The intention was to allow C++ namespace identifiers in
> > foreign functions. But with this minimal extension, C++ namespaces
> > identifiers are also allowed for ordinary identifiers.
On 09/13, Yann Orlarey wrote:
>
> That was fast ;-) The intention was to allow C++ namespace identifiers in
> foreign functions. But with this minimal extension, C++ namespaces
> identifiers are also allowed for ordinary identifiers. For a while, I
> wondered if it should be banned and I admit I do
Hi Oleg,
That was fast ;-) The intention was to allow C++ namespace identifiers in
foreign functions. But with this minimal extension, C++ namespaces
identifiers are also allowed for ordinary identifiers. For a while, I
wondered if it should be banned and I admit I don't have a firm opinion on
the
Hello Yann and Stephane,
I have some questions about the recent commit 1cf7b6d1b57
("Faust identifiers extended to accept c++ namespaces").
So, iiuc, with this patch I can use, say, ffunction(float, std::sin(float)),
and this is nice.
But this also allows "::" in the pure .dsp code, say
