<oriansj>fosslinux: did that solve your segfault problem? <oriansj>fosslinux: well you can always throw code my way if you need help reasoning about how M2-Planet interacts with it <fosslinux>i'll throw something in the next half hour or so <fosslinux>if you have a random string that is undefined, ie an undefined constant or variable here, that seems to be passed directly to the build_member function which dies <oriansj>fosslinux: that would be struct type* build_member(struct type* last, int offset) in cc_types.c <oriansj>and it does have require_match("Struct only supports [num] form\n", "]"); <oriansj>because it attempts to numerate something that isn't a number and it returns zero as the size <oriansj>fosslinux: well there is a way to catch that <oriansj>simply check if i->size == 0 after the assignment and throw an error <oriansj>easy to fix, I'll have a patch up in one minute <oriansj>thank you fosslinux for helping me improve M2-Planet ^_^ <oriansj>fosslinux: also do you like the new REPL behavior in mes-m2? <oriansj>fosslinux: I am glad you like it ^_^ ***ng0_ is now known as ng0
<real-grue>Do you collect such articles about simpler software/KISS/etc? <`Lion>not really, but occasionally i search around hoping to find something at least mildly interesting and unusual <`Lion>that i believe rain1 maintains <real-grue>Also, from TRIZ: "it is said that we have obtained an ideal system which then has no more existence as a system, but its useful function remains assured. This result is referred to by G. Altshuller: Ideal Final Result".