One approach to having a true-negative indexed array is by defining an object-oriented approach. A set of tied-array-based methods might provide the most natural approach. That should allow a transparent mapping of negative array indices onto a perl zero based array implementation while avoiding a string based (hash) array that might result in reqiring more memory to store the same amount of information and more CPU resources to do array element store and fetch operations. Such a set of methods might utililize a definition mechanism that specifies the lowest and highest array indices for an array, e,g, @array=new negarray(-30,-10); The base property for the array would be 30, the value that would have to be added to each index value to translate it back to a zero-based perl array implementation. Such an array would also be created with a index-limit property of 20. Such an approach would allow each array to have it's own base index value rather than using the $[ mechanism which defines an array base value that is used for all arrays. Note, this mechanism could still allow perl-style negative indexing for calculating actual index values relative to the end of an array, ,i.e. $array[-9] could still be used to fetch the last element of the negarray @array. This of course would not be limited to array with just negative indices. There would be no reason which such an array could not be defined as having elements having lower/upper limits defined as -10:+20 or +30:+80 so long as the lower limit was less than or equal to the upper limit.
**** [EMAIL PROTECTED] <Carl Jolley> **** All opinions are my own and not necessarily those of my employer **** _______________________________________________ Perl-Win32-Users mailing list [EMAIL PROTECTED] To unsubscribe: http://listserv.ActiveState.com/mailman/mysubs
