This is only the case on 0.40 and newer. [import] on 0.39 loads into
the global namespace. This stuff is currently alpha and not
complete, so it is not very clear. I think for the manual, the best
bet for Pd-extended is to use the [library/objectclass] format for
objectclasses that aren't loaded by default. That has been supported
by every version of Pd for a long time now (as long as you install
the libraries that way).
.hc
On Apr 8, 2008, at 9:22 AM, marius schebella wrote:
as far as I understood [import] is the same as [declare -lib] and that only adds a library to the local namespace of a patch. see the help
file for declare that comes with 0.41. you can declare your library relative to pd [declare -stdlib] or relative to the patch [declare -lib], but - as mentioned in the
helpfile
- the name stdpath is confusing!].
it is also not 100% clear, how this works in abstractions and if the behaviour will be consistent with future pd versions. there might be a chance that [import] really adds to the global namespace, but I don't think so. (otoh I don't know how to add
something to the global namespace.) the idea was that you can use a certain objectclass in one patch and another one with the same name (but from another lib) in another
patch. please correct me, if I'm wrong. marius.Derek Holzer wrote:
Hi all,
it seems that the ways of dealing with externals and paths is
always in flux! I would like to confirm a suspicion that, for the time
being, the following works the way I think it does, assuming PD 0.39:[library/object] imports that specific object into global
namespace, and can accommodate different objects with the same name from different libs. This method does not allow access to help patches or
abstractions in the library path.[import library] imports the whole library into global namespace, including help patches and other abs (usually, although I have often found this broken in Extended). It cannot accommodate different
objects with the same name from different libs, as the last library imported will have priority.Is this correct so far? Has anyone documented this any more substantially anywhere? How does this change for 0.40?
best! d.
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/ listinfo/pd-list
Computer science is no more related to the computer than astronomy is
related to the telescope. -Edsger Dykstra