Nicolas Weeger wrote: >> seems to be a bit of redundant code to check to see if the direction >> is buildable - perhaps reduce that redundancy into one function like >> 'check_buildability(....)? That would seem to be useful if additional >> buildable options is added. > > > Probably. But building functions don't always check the same thing, or > keep some variables for later use (item below floor, for instance, when > building a floor). Ok. I didn't look really close at the exact details, but it seemed to be some amount of duplication of 'is this within map bounds, is there a flag buildable, etc'. Maybe the amount to be reduced isn't so great. OTOH, perhaps a flag can be passed into the check function which also has some indication what to find and what to return. > Those messages are supposed to never happen. If this message ever comes, > it means there may be some issue with the map (or i forgot something in > the tests, quite possible). For instance if you build a floor in a > direction, there should _always_ be something there (wall, floor, > anything). Because if nothing, it's an undefined square, and the player > is standing next to it... If that's the case, then it should also generate a LOG() message. > Ok, even though i do this to avoid mistaking for the evil 'if (op->type > = WALL)'. gcc at least will catch those and print a warning. But it's just easier to me (and really in the style of the rest of the code) that the constant comes second. _______________________________________________ crossfire-devel mailing list crossfire-devel at lists.real-time.com https://mailman.real-time.com/mailman/listinfo/crossfire-devel