Re: to be effcient or not to be efficient
Tuukka Toivonen <[email protected]>
| Newsgroups | gmane.comp.multimedia.media-api |
|---|---|
| Message-ID | <Pine.GSO.4.58.0401141909260.19799@stekt37> |
On Wed, 14 Jan 2004, Enrico Weigelt wrote: >* Tuukka Toivonen <[email protected]> [2004-01-08 21:05:36 +0200]: >> - You might not care about possible errors, and in that case no >> checks are necessary at all if non-NULL pointers are guaranteed >> (it's also safer: some buggy code might not check for NULL >Well, in a debuggin build, we could also replace all NULL pointers >by dummy calls No! It's much easier to find bugs when the program really segfaults; the developer might not notice a message on console especially for GUI programs and a segfault forces him to fix the bug instead of postponing and forgetting whole thing. And I was not talking about developers but end users. End users don't want debugging compiled in, but they do want a program that doesn't segfault.