On Wednesday 19 December 2007, Henning Westerholt wrote: > > Well, even though you're busy I still think that the change helps > > nothing and makes it harder for someone to make changes if they need > > to. > > I think its help because one of the main problems for a medium sized > code base like this is the clutter that gets accumulated over the time. > This make maintainance and understanding of the codebase harder then > necessary.
I still do not understand how a named constant improves things over a #define. Can you elaborate on this in the context on this change? > If you had take a look into the actual change, you've noticed that the > TIMEOUT define was actually located in line 429. This is not "on top of > the file" for me. Which means it was in a non-optimal position, but that can be improved. I cannot say the same about a named constant inside a function. > > If you maintain such a patch, and i have made your life a little bit > harder now, then i'll revert this change. If this is not so, and you > still can not accept my change, then we need to start a voting about > this. I don't maintain such a patch, and I'm not affected in any way by this change. Why I raised this issue is because you made a change that was adding little value (if any), while it potentially complicates things. If I do not voice my concerns now, tomorrow you may decide to change something else in a similar manner which would affect me. I think it would only be fair if you would consider such observations because others may have different views on how things should be done, and in the end I think it comes to how the person maintaining that code prefers to do things. > > But this perhaps is a little to much overhead for this type of change. As I recall, the consensus was that changes to code for which you are not maintainer should get the approval of the maintainer unless obvious bug fixes. I do not recall there was any discussion why it would be better to replace #defines with named or even unnamed constants, so that people can state their views on the issue. I do not think this is a minor issue that only one person should start to make changes for without first hearing what others have to say about it. -- Dan _______________________________________________ Devel mailing list [email protected] http://lists.openser.org/cgi-bin/mailman/listinfo/devel
