1. Re: Matheval (was: Dos32Lib, Win32Lib)
On 2000-06-06 EUPHORIA at LISTSERV.MUOHIO.EDU said:
EU>[snip]
EU>> EU>> Are right brackets automatically appended now , or are
EU>>these EU>>seen as being unnecessary clutter.
EU>> EU>I'm not sure what you mean here.
EU>> when I entered an expression into eval.e , it returned
EU>> something like
EU>> this for the expression sin(x*(a*x+ , without the closing
EU>>right bracketts.
EU>Hmm...I think you might be confusing matheval with David Cuny's
EU>eval.e. They're two separate packages. Matheval should return all
EU>parenthesis correctly (in fact, the pretty print function will
EU>probably add some parenthesis).
I believe I am , this is what I downloaded , quite awhile ago ; with
one of your attempts at using Davids' parser.
EU>> I'm having some difficulty locating matheval
EU>[snip]
EU>> So where can I find the most recent version , I was unable
EU>> to locate
EU>> this on d.cunys website , certainly not within the more
EU>> recent versions of
EU>> dos32lib.* or win32lib.* .
EU>Yup. Completely different. :)
EU>It's on the recent contributions page. In fact, I just sent an
EU>update to
EU>http://members.xoom.com/matthewlewis/matheval.zip
EU>It's about 32K. I added a graphing feature, and polished up the
EU>parser to allow more flexible variable naming.
After sending this message I shall attempt to download matheval.zip
from that site. My browser doesn't support frames so there's ,
potentially , some difficulty.
Is this in the form of *.ex and *.e files or a *.exe ?
EU>> As you should be aware now , I uploaded this ; via email
EU>> to you ; a few
EU>> days ago. I understand , you might have numerous other
EU>> concerns at this
EU>> time , therefore you haven't evaluated this.
EU>I did take a look at it yesterday (I suspect more reason for the
EU>lag is our different locations on the globe--I'm in California).
EU>It took a little bit to get it to work. Mainly the math.e part,
EU>since some of the functions have been defined since EU 1.5 (arcsin,
EU>etc). Very interesting, though. I like the way you evaluated the
EU>function over the range, although that approach is incompatible to
EU>matheval's way of doing things.
Were you able to , easily , comprehend what I'd done though ?
What's the weather like there ; have you had any unexpected storms ?
Are you close to agricultural land ?
I used EU 1.5 , because this is public domain , there might be a more
recent version. Most functions can be described in terms of the basic
ones , once you know how they're defined and used ; power() etc.
I use -1<= x <=+1 , you can then use a multiple of x to extend the domain
or add a value to x to shift the domain.
for instance f(5*x) is from -5 to +5 , f(x+1) is from 0 to 2 ;
f((x+1)/2) is from 0 to 1 and f(5*(x+1)/2) is from 0 to 5.
Pre-assigning the domain to a sequence can , obviously , be extended
to a number of dimensions.
If you're attempting to actively select a region of the domain then
your approach might well differ.
EU>Take a look at the new graphing routines. They seem to run fairly
EU>fast (at least on my PII 300), as long as you don't have too
EU>complicated a function, and not too many iterations to calculate
EU>(10000 goes pretty quickly).
EU>Matt Lewis
How fast did you find my software to be on your computer , you
can put all of the files on ramdrive to gain a further increase in
speed of evaluation.
I'm hoping that the ability to evaluate complicated functions can be
retained , while improving upon the speed of evaluation.
To do so I need to expand my understanding of this whole area ,
we appear to have taken quite differing approaches ; this seems
encouraging to me.
I'd also like to put all of the software into one package that's
platform independent.
In the meantime I require some assistance on redirection .
Net-Tamer V 1.11 - Test Drive
