On Tue, 2008-08-12 at 17:42 -0500, Mike McGonagle wrote: > Hi Roman, > > Not sure if this is the problem, but I had similar troubles like this, > and to solve the issue, I saved the patch that contains the [declare] > in it, and then reopened it. I guess the issue here is that declare > really only operates when a patch is first opened from a file. (I > could be wrong on this...) > no, you are fully right and i also was aware of this. when testing with binary files (as opposed to abstractions) it gets even trickier: it seems that pd holds a list for the namespace of binary classes, whereas it seems, that abstractions are always searched in the filesystem. which means, that when you once have successfully instantiated an object from a binary class (for example [mrpeach/tcpserver]), then [tcpserver] will instantiate as well (during the whole pd session). this means, that in certain cases it is not even sufficient to reload the patch in order to realiably test [declare], but you would have to restart pd to get rid of the registered classes.
personally i dislike the fact, that binary single object classes behave so differently from abstractions. this means, that patches using [abs~] from extra/zexy might have worked well with the binary 'extra/zexy/abs~.pd_linux', but when it is replaced by 'extra/zexy/abs~.pd' the same patch probably doesn't work as it did before. roman ___________________________________________________________ Telefonate ohne weitere Kosten vom PC zum PC: http://messenger.yahoo.de _______________________________________________ Pd-dev mailing list Pd-dev@iem.at http://lists.puredata.info/listinfo/pd-dev