Wow, this one comes up all the time. $0 doesn't exist in "message"-space. $n in a message means something completely different than $n as a creation argument - this is covered in the first chapter of the help docs. (Now might be a good time to review them.) The standard way to get $0 into a message is to pass it as a message argument, for instance:
[bang( | [$0] | [set $1-table(
or what have you - this will get you what you want.
HOWEVER, I would for one be an advocate of changing Pd to make $0 an exception to this rule, so that $0 means the same thing in messages and in creation arguments. Since it doesn't mean anything in message-space anyway, and is simply used to create a sort of namespace isolating a patch from its environment, it's also strictly speaking not a creation arg in the sense of $1 and so on (even though it is initialized at creation time). This would save LOTS of patching. What does everyone think? I mean, is there a really good reason to keep [$0( meaningless?
-david
On 4/5/06, Peter Worth peterworth@gmail.com wrote:
i'm sending soundfiler a message which says:
read -resize $1 $0-sample
$0-sample is a table (the $0 is required because it is an abstraction so if there is more than one instance of it, there can't be tables with the same name).
but this gives the error: 0-sample: no such table i.e. the $ is being removed.
anyone else had this problem?
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list