Split and Combine

[Raymond E. Griffith]

on 4/30/04 9:43 PM, J. Landman Gay at jacque at hyperactivesw.com wrote:

> On 4/30/04 6:49 PM, Raymond E. Griffith wrote:
> 
>> I am using split and combine with tab-delimited data. The data is returned
>> in a different order with combine than it was originally configured. I can,
>> of course, regather the data with a repeat for each line in keys(foo) type
>> of coding, but that is less efficient. It also appears to affect lines with
>> larger than nine entries. Any ideas?
> 
> Arrays are recombined according to an internal hash and they don't
> preserve the original order. The only solution is to either sort the
> returned data or reconstruct the order the way you mention.

I was afraid of that. The internal hash almost seems alphabetical.

I have been working on a determinant routine and trying to improve the time
(taking the determinant of a 30x30 matrix can be time consuming). This
rearrangement hinders a line of attack. Oh well, back to the drawing board.

Thanks,

Raymond