I would never have thought of the second example, a clever rewriting rule. Thank you very much for the advice.
Le jeu. 13 janv. 2022 à 19:43, Dario Sanfilippo <sanfilippo.da...@gmail.com> a écrit : > Hi, Jean-Louis. > > From a first glance, the ba.selectn could be faster. > > Similarly, another easy-to-read solution that is very similar is using > nested ifs: > > lut(n) = ba.if( n == 2, > 3, > ba.if( n == 3, > 4, > ba.if( n == 4, > 2, > n))); > > A while back Oleg Nesterov helped me write a function with pattern > matching to implement nested ifs with arbitrary > if0-then0-if1-then1...ifN-1-thenN-1-else: > > ifN((c, t, e)) = ba.if(c, t, ifN(e)); > ifN(e) = e; > > you could then write your lut function as follows: > > lut(n) = ifN((n == 2, 3, n == 3, 4, n == 4, 2, n)); > > Ciao, > Dr Dario Sanfilippo > http://dariosanfilippo.com > > > On Thu, 13 Jan 2022 at 19:08, Jean-Louis Paquelin <jean-lo...@paquelin.net> > wrote: > >> @James and @bart, thank you for your quick reply. >> >> So (if I understand you well) pattern matching should be thought of as a >> static rewriting tool. >> And, btw, you're right >> >> process = 2 : lut : hbargraph("B", 1, 5); >> >> sets the bargraph to 2, while >> >> process = lut(2) : hbargraph("B", 1, 5); >> >> sets it to 3 (as expected). >> >> I came with 2 ways to write the lut() function I need: >> >> lut1(n) = 1,3,4,2,5 : ba.selectn(5, n - 1); >> lut2(n) = ((n == 1) | (n == 5)) * n + (n == 2) * 3 + (n == 3) * 4 + (n >> == 4) * 2; >> process = hslider("A", 1, 1, 5, 1) <: (lut1 : hbargraph("B", 1, 5)), ( >> lut2 : hbargraph("C", 1, 5)); >> >> Both functions seem to provide the service I'm asking for. But the first >> one is more readable than the second. >> Any idea which one is the lightest for the CPU? >> >> Thanks again, >> >> jlp >> >> Le jeu. 13 janv. 2022 à 16:41, James Mckernon <jmcker...@gmail.com> a >> écrit : >> >>> On 1/13/22, Jean-Louis Paquelin <jean-lo...@paquelin.net> wrote: >>> > Hello there, >>> > >>> > I can't figure out how the pattern matching works in Faust. >>> > >>> > I've tried the following code: >>> > >>> > lut(2) = 3; >>> > lut(3) = 4; >>> > lut(4) = 2; >>> > lut(n) = n; >>> > >>> > process = hslider("A", 1, 1, 5, 1) : int : lut : hbargraph("B", 1, 5); >>> > >>> > Trying this in the IDE, the B bargraph simply replicates the A slider >>> > value. I was expecting a different behavior thru the lut function: the >>> > output of lut should be the same as its input except when the input is >>> 2, 3 >>> > or 4, in those cases the output of lut should be 3, 4, 2 respectively. >>> > >>> > I tried the alternative syntax without any success: >>> > >>> > lut = case { >>> > (2) => 3; >>> > (3) => 4; >>> > (4) => 2; >>> > (n) => n; >>> > }; >>> > >>> > It's as if none of the cases match except the most general one. >>> > >>> > Did I miss something? >>> > >>> > I wish you all the best for 2022, health, joy and great projects! >>> > >>> > Best regards, >>> > >>> > Jean-Louis >>> >>> Pattern matching is done at compile-time, so to match the case of >>> lut(2) (for example), the argument needs to be fixed and known at >>> compile-time to be 2. This is true of lut(2), even of expressions such >>> as lut(1 + 1), but not taking lut as a function of some potentially >>> time-varying expression. >>> >>> Also, I'm not 100% sure of this, but I think pattern-matching may only >>> take effect when a function is applied with explicit, parenthesis >>> syntax, not when it is chained with : (i.e. lut (x) rather than x : >>> lut). If I'm wrong about this then perhaps someone can correct me; in >>> any case I don't think this was the problem with your example. >>> >>> The way to do what you are attempting to do at run-time would probably >>> be by chaining a number of select2 functions together. I'm not sure >>> whether there are any helper functions in the standard library to help >>> with this; it seems like the sort of thing that ought to be present, >>> but I can't think of any off the top of my head. >>> >>> Hope that helps. >>> >> _______________________________________________ >> Faudiostream-users mailing list >> Faudiostream-users@lists.sourceforge.net >> https://lists.sourceforge.net/lists/listinfo/faudiostream-users >> >
_______________________________________________ Faudiostream-users mailing list Faudiostream-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/faudiostream-users
