I think that the Pd API previously required externals to define class_sethelpsymbol(), but now, it mostly causes problems because it prevents Pd from looking for the help file with all possible name and path derivations. Also, when an external has class_sethelpsymbol() set, then its help patch won't be found when an object is instantiated with a geiger namespace prefix (i.e. [mylibrary/myobject]).
Currenty, the only situation where its needed is where many objects share a help patch, but I argue that each object should have its own patch with a distinct example, which can then link to an overarching all_about_ patch. This is the standard that the PDDP has worked out.
I have a sed script which removes class_sethelpsymbol() easily. Does anyone mind if I remove it from their objects? I will leave it in place in the above condition, where its needed. Here are some libraries affected: ann, chaos, beatpipe, cxc, ekext, ext13, jackx, k_cext, k_jack~, OSC, pidip, signal, sprinkler, svf~ unauthorized, zexy, zhzxh~.
Here are some that are affected, but maintained outside of the pure-data CVS (therefore I can't change): gridflow, iemmatrix, iemlib.
Here's the sed script for those who want to do it themselves:
sed -i.bak 's|class_sethelpsymbol(.*class, *gensym(.*));||' *.c
.hc
________________________________________________________________________ ____
"[W]e have invented the technology to eliminate scarcity, but we are deliberately throwing it away to benefit those who profit from scarcity." -John Gilmore