AFAICS the compiler semantic was made more strict at some point, I hope Yann (in vacation for now...) can comment on that later on.
Stéphane Le 26 août 2014 à 08:26, Dolittle <[email protected]> a écrit : > Stéphane, > > Thanks for the contemporary reference. Do you have any idea why the old > compiler does not throw an error as does the newest one? > > Are you able to compare Yann's original diagram with my previously generated > svg -- linked on dropbox? Basically, my old code is is unusable in the > current faust. I remain puzzled. > > Bob Hamilton > > > On Tue, Aug 26, 2014 at 1:04 AM, Stéphane LETZ <[email protected]> wrote: > From "Faust quick reference" p19 > > Sequential Composition The sequential composition A:B (figure 3.3) expects: > outputs(A) = inputs(B) (3.1) > > It connects each output of A to the corresponding input of B : > A[i ] → [i ]B > > Stéphane > > > Le 26 août 2014 à 04:48, Dolittle <[email protected]> a écrit : > > > Folks, > > It seems that the serial composition operator, : , is either broken or has > > changed since I last used it in my Faust code during 2009. Code that was > > once used to generate svg blocks, c++ , pd, and a vst now throws an error > > in both FaustLive and in my current, offline Faust compiler, Ver-0.9.67. I > > narrowed it down to a problem with the serial operator. > > Instead of referring to my own complex code, I've created a small example > > that illustrates the problem. The following dsp will generate an svg from > > the old Faust : > > a = _, _; > > b = _; > > test = a : b; > > process = test; > > Both the FaustLive and the offline Faust compiler, Ver-0.9.67, throw the > > following error: > > “Error in sequential composition (A:B) > > The number of outputs (2) of A = _,_ must be equal to the number of inputs > > (1) of B : _” > > In contrast, here's a quote from Yann Orlarey's “A Faust Tutorial,” Chapt. > > 3, Sect 3.1, Serial: > > “The remaining inputs or outputs are directly used as inputs or outputs of > > the resulting block-diagram.” > > Now, however, it seems that instead of rerouting unused outputs, the > > operator demands an exact match between inputs and outputs. > > Here's a link to a copy of the svg generated from my example by the older > > version of Faust,V-0.9.9.4d: > > https://www.dropbox.com/s/6sgnprgozbxz4ob/process.svg > > Note the resemblance to the diagram on pg. 22 of the above-referenced > > tutorial. > > Any ideas? > > > > Robert Hamilton > > seedpress > > ------------------------------------------------------------------------------ > > Slashdot TV. > > Video for Nerds. Stuff that matters. > > http://tv.slashdot.org/_______________________________________________ > > Faudiostream-users mailing list > > [email protected] > > https://lists.sourceforge.net/lists/listinfo/faudiostream-users > > ------------------------------------------------------------------------------ Slashdot TV. Video for Nerds. Stuff that matters. http://tv.slashdot.org/ _______________________________________________ Faudiostream-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/faudiostream-users
