🚀 go-pugleaf

In article <1992Nov15.230107.29374@gandalf.UMCS.Maine.EDU> jurlwin@gandalf.UMCS.Maine.EDU (Jeff Urlwin) writes:
:In article <1e44n6INN49u@iskut.ucs.ubc.ca> ochealth@unixg.ubc.ca (ochealth) writes:
:>In article <1e3ejjINNdf4@pollux.usc.edu> ckhung@pollux.usc.edu (Chao-Kuei Hung) writes:
:>:
:>:int cmp(const int *i, const int *j)
:>:{
:>:    return(data[*i].x < data[*j].x ? -1 : 1);
:>:}
:>
:>
:>BZZZZZT! How will cmp() ever show that data[*i].x == data[*j].x??
:>Your cmp() can never return 0 to show that the two are equal. Perhaps
:>this worked on other platforms, because that particular sort didn't
:>matter if  a==b. I think the sort is getting into an infinite loop
:>because of this.
:>
:>This works in EMX/GCC:
:>
:>int
:>cmp (const int * i, const int * j)
:>{
:>	if (data[*i].x == data[*j].x)
:>	  return 0;
:>	if (data[*i]].x < data[*j].x)
:>	  return -1;
:>	return 1;
:>}
:
:Better yet, 
:  return (data[*j].x - data[*i].x);


I know, but my ANSI reference says compare() should only return -1, 0, or 1.
for < = >, but maybe that's wrong.

:
:Jeff
:
:-- 
:--------------------------------------
:jurlwin@gandalf.umcs.maine.edu
:


-- 
______________________________________________________________________________
jpm: ochealth@unixg.ubc.ca         
                            Happily using OS/2 2.0 because MS Windows isNT ___
                                                      Insert VapourFeature ^^^