On Tue, 22 Sep 1998, Jason Wilkins wrote:
>In the file with main(), comm.c I believe, there is a utility function
>that says it was rewritten to stop a warning (passing a structure back on
>the stack, or temporary area I believe), but is says that they sacrificed
>thread safeness to do it. Its not thread safe because it keeps its return
>value in a static structure. Why not just add a parameter to the function
>that points to where to put the answer?
At the time I did that to cause the least changes in bpl14. It has since
been changed to do the same as strcpy() and the like.
>Its just a nit-pick. I was just wondering if there was a reason that it
>was done this way? The way I propose is also the most efficient. So that
>doesn't seem to be an issue. It seems to just be laziness, I understand
>completely. Make the change that will cause you to change the least code.
More of an aversion to a huge change. Once we take the little step and
nothing breaks, we go with the whole thing. Sort of like the IS_NPC()
changes. Some of those are in right now, but the critical one [the actual
macro itself] has not been changed pending further analysis I don't want to
make at the moment.
--
George Greer, greerga@circlemud.org | Genius may have its limitations, but
http://mouse.van.ml.org/ (mostly) | stupidity is not thus handicapped.
http://www.van.ml.org/CircleMUD/ | -- Elbert Hubbard
+------------------------------------------------------------+
| Ensure that you have read the CircleMUD Mailing List FAQ: |
| http://democracy.queensu.ca/~fletcher/Circle/list-faq.html |
+------------------------------------------------------------+
This archive was generated by hypermail 2b30 : 12/15/00 PST