Hi, Yann. I get your point.
So would it be enough to put some version of the feature that I described earlier within a "debug" flag so that we enforce that it doesn't belong to the Faust paradigm? If this debug mode doesn't have any flaws regarding performance, chances are that users will default to the debug mode to be able to use the feature. Dr Dario Sanfilippo http://dariosanfilippo.com On Tue, 23 Jan 2024 at 10:50, yann orlarey <orla...@gmail.com> wrote: > Hi Dario, hi everybody, > > I understand the motivation, but unfortunately, it's a proposal at odds > with the fundamental principles of Faust. To take an analogy, it's like > reintroducing "goto" into a structured programming language ( > https://homepages.cwi.nl/~storm/teaching/reader/Dijkstra68.pdf), or > reintroducing imperative features into a functional programming language. > If you gain a little in "flexibility", you gain nothing in expressiveness, > and you lose a lot in terms of modularity and the ability to reason about > programs. Faust's algebraic and functional approach imposes a discipline > that may seem restrictive at first. Still, it enables us to reason about > programs and compose them algorithmically, in a safe and efficient way. On > the other hand, as I said on Discord, we agree on the need for debugging > functionalities... > > Yann > > Le lun. 22 janv. 2024 à 18:26, Dario Sanfilippo < > sanfilippo.da...@gmail.com> a écrit : > >> >>> process = catch("fb") : throw("fb", +(1)); >>> >>> >>> >> I guess that this above should have been >> >> process = catch("fb") : throw("fb", +(1)) , catch("fb"); >> >> >> Dario >> >> _______________________________________________ >> Faudiostream-devel mailing list >> Faudiostream-devel@lists.sourceforge.net >> https://lists.sourceforge.net/lists/listinfo/faudiostream-devel >> >
_______________________________________________ Faudiostream-devel mailing list Faudiostream-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/faudiostream-devel
