Re: Anyone with reverse polish notation experience?
>From: Derek Parnell <ddparnell at bigpond.com>
>Subject: Re: Anyone with reverse polish notation experience?
>
>
>On Tue, 11 Mar 2003 23:31:18 +0000, Pete Lomax <petelomax at blueyonder.co.uk>
>wrote:
>
>>
>>I'm parsing (prefix) expressions and getting reverse polish notation
>>out. The question I have is at the point of pushing a postfix operator
>>on a stack, clearly *all* the necessary pre-calculations will have
>>been done. Should be a no-brainer, but I'm doing it register-based
>>rather than stack-based.
>>
>>I'm adamant about prefix; eg add(3,4) instead of 3+4
>>
>>Maybe later I'll get my head round infix.... (Guess I could stomach a
>>link which makes me confident enough to do both in v0.1, rather than
>>as I am just trying to get it working...)
>>
>>My question at this point is: Has anyone a difficult prefix-only
>>expression hard to parse; rpn which worked unexpectedly, and/or any
>>pointers which may save me some time?
>>
>>Eg: add(add(3,4),add(5,6)); not: (3+4)+(5+6)
>>
>>Equally, does anyone have an expression they feel would be hard to
>>express in prefix rather than infix notation?
>>
>
>I guess this really an oxymoron as by its very nature, prefix notation is
>simple to parse. And providing you have enough registers (or stack space)
>any complex prefix expression is easy using a recursive algorithm.
>
>But in case you up to a neat expression (and much time to evaluate it), I
>present ...
>
>
>power(add(div(power(div(add(27,power(69,div(1,3))),2),div(1,3)),3)
>,power(div(2,add(27,power(69,div(1,3)))),div(1,3))),30000)
>
>This should work out to a very large number that is almost an integer.
>
Assuming add(x, y) equals x + y; div(x, y) equals x / y; and power(x, y)
equals x ^ y; that equals...
1.456616911969342313612567116226969533712140705732487890018328015e2725
which is, ahem, only slightly over the size of an atom...
>
>cheers,
>Derek Parnell
>
|
Not Categorized, Please Help
|
|
