original: http://www.pulsar.org/archive/rob/JExpressionEvaluatorObj/
archived: http://web.archive.org/web/*/http://www.pulsar.org/archive/tav/JMSAgentObj/

JExpressionEvaluatorObj

Rob Salgado (salgado@physics.syr.edu)
Tav Hawkins (taviare@physics.syr.edu)
v.990729 (BUGS FIXED! FOR REAL, THIS TIME! [We hope...])

(The computational engine for this external module was based on Laptev Eugene's (ztv@imp.kiev.ua) Fast Expression Evaluator , who has graciously allowed us to use his source code.)

By default,
the module has

two inputs, named "x" and "y". (On the NeatTools desktop, mouseOver the inputs to see this.)
the expression "x^2+y^2".
the label "f(x,y)".

By the changing the module's properties, we can change the module so that it has

three inputs, named "x", "y" and "z".
the expression "sqrt(x^2+y^2+z^2)".
the label mirroring the expression

[Warning: before you connect the RealObj to the output, set that RealObj to have more than 3 digits and more than 1 floating-point place to see more decimal places.]

By supplying a string (say, using a LabelObj) to our module,
we can change the expression. Here, we purposely typed a syntactically-incorrect expression.

Upon a syntax error,
the module turns RED and trips the ERROR_FLAG, as indicated by an LEDObj connected to it. This indicates that the output is not reliable.

Upon an illegal value error,
the module turns YELLOW and trips the ERROR_FLAG, as indicated by an LEDObj connected to it. This indicates that the output is not reliable.

WARNING WARNING WARNING
use PARENTHESES to avoid confusion!

An expression like "-z^2" is evaluated as "(-z)^2" ... not "-(z^2)".

Laptev Eugene's (ztv@imp.kiev.ua) Fast Expression Evaluator uses the following order-of-operations.

From his documentation:

Operators are handled in the following priority: 1) Unary (+) and (-) signs. 2) "To the power of" symbol: (^). 3) Times (*), divide (/) and modulus (%). 4) Plus (+) and minus (-).

Unfortunately, this is different from the standard convention used in C and C++.

If I have time, I'll try to fix this. For now, use PARENTHESES to avoid confusion!

To use our binary:

Download the External Module (.dll) file
JExpressionEvaluatorObj.dll
[right-click, save, but don't use Netscape, you might get some junk too!]

Consult Laptev Eugene's documentation for help with the syntax.

To build your own binary:

Download the latest versions of NeatTools and the developer's kit:
Win95/NT version (oopw.exe, oopw.cfg)
Win95/NT External Module Developer Kit
Download our source and header files:
Place this is a folder called "JExpressionEvaluatorObj" in your "develop" folder.
JExpressionEvaluatorObj.cpp (COMMENTED CODE!)
JExpressionEvaluatorObj.h
EE-src.cpp (can you see how to change the order of operations?)
EE.h
Here's the Visual Studio project.
JExpressionEvaluatorObj.dsp
After copying all five files to your "JExpressionEvaluatorObj" folder (as directed above!), click on this .dsp-file and say okay to the "warning". You're ready to compile!

JExpressionEvaluatorObj.zip - all of the files

Have fun!

Back to the NeatTools page