On Mon, 8 Jan 2001, Peter Ajamian wrote: > Actually that's not true, an implementaion must convert 0 to a NULL > pointer when an implicit caste is made however, it does not have to > store NULL as 0 in memory. Actually, it is true. It's simply a matter of pedanticism in both cases, since the net result is unchanged, however, *NULL* is zero. The actual null pointer is machine-dependent, but the definition of NULL is not. I did, however, make a technical mistake (just not the one you thought I did - although you made me realize it): calloc() does not guarantee nulling pointers, since it does a bit-by-bit zero. This is still a matter of concern only to language lawyers: CircleMUD does not run on any platform that does not use address 0 as NULL, and there are no extant platforms that use anything else. IOW, my suggestion remains valid, if not portable to long dead platforms. Thanks for keeping me on my toes. :) -dak -- +---------------------------------------------------------------+ | FAQ: http://qsilver.queensu.ca/~fletchra/Circle/list-faq.html | | Archives: http://post.queensu.ca/listserv/wwwarch/circle.html | +---------------------------------------------------------------+
This archive was generated by hypermail 2b30 : 12/03/01 PST