Elijah Newren <newren@xxxxxxxxx> writes: > ... would you two prefer > that I add a new affirm() function that is ALSO compiled out in > production? The reason I do not think affirm would fly is because it shares the most problematic trait with assert(): it can be comipled out. There was at least one bug we found and fixed in a piece of code that came later in a codepath that had an assert() in it. The buggy code (IIRC, it was added way after the assert() was written in an unrelated patch) was unknowingly depending on a side-effect of an innocuous function call (i.e. assert(func_tion(args) == expected)) made in the assert condition, and the bug did not reproduce in the developer's environment.
