[PD-dev] bug: loadbanged $1 not passed to outlet

flo at mur.at flo at mur.at
Fri Apr 8 02:37:27 CEST 2011


Hi,

I stumbled across a bug today in an abstraction where a loadbanged $1  
is passed on to an outlet. The message can be picked up at the outlet  
as expected the first time the patch is loaded, but not when it is  
reloaded (for example because its creation arguments are changed in  
the parent patch calling it).

Attached is a minimal working example (abstraction plus calling  
patch). Open mmw_call.pd and follow the instructions there to get the  
idea.

Best,
flo.H

-----------------------------------------------------------------------
..........Sent by mur.at Webmail Testinterface.........................
....this annoying message will disapear in the production version......
.................................................really................
.............promise...................................................

-------------- next part --------------
#N canvas 277 160 438 154 10;
#X obj 29 31 loadbang;
#X obj 30 66 symbol \$1;
#X obj 28 100 outlet;
#X obj 88 98 print inside_abs;
#X text 99 8 This is the abstraction used by mmw_call.pd;
#X text 165 126 Florian Hollerweger \, 2011 \, flo at mur.at;
#X connect 0 0 1 0;
#X connect 1 0 2 0;
#X connect 1 0 3 0;
-------------- next part --------------
#N canvas 277 160 464 313 10;
#X text 149 280 Florian Hollerweger \, 2011 \, flo at mur.at;
#X obj 20 274 print outside_abs;
#X text 13 81 1) observe how the creation argument "foo" of [mmw_abs]
was printed twice on load \, once inside the abstraction itself \,
and once in this patch.;
#X text 14 135 2) reload the [mmw_abs] abstraction \, for example by
renaming its creation argument from "foo" to "bla";
#X text 16 179 3) Note how this time \, [mmw_abs] prints only *inside*
the abstraction itself \, but does not pass the message on to its outlet
as expected!;
#X obj 20 248 mmw_abs foo;
#X text 12 23 To recreate the bug (tested in Pd 0.42.6 under Debian
GNU/Linux):;
#X connect 5 0 1 0;


More information about the Pd-dev mailing list