Bob Tanner wrote: > > First, I'd like to make a suggestion. If a function has the side effect of > loading/altering/etc a static data struct it should be documented in that > function somewhere. As I think about this quickly, its a reasonable idea, but lots of functions modify/change something that was not passed to them. But generally speaking, the comments with the source are not all that great. > > I believe a case in point is arch.c:first_arch_pass(). The comments says: > > /* > * Reads/parses the archetype-file, and copies into a linked list > * of archetype-structures. > */ > > Yet the prototype is void, so it looks like it must load the archtype linked > list into a static variable somewhere. > > Once first_arch_pass() is called, what is the name of the linked list it > creates? first_archetype