World: r3wp

regarding the supercalculator - it has wrong associativity as well

like parse-equation.r did yesterday btw :)

Ladislav... sorry, what should be the code returned?

in which case?

3-3+3

add subtract 3 3 3

Do you have an example where this would make a difference?

>> subtract 3.0 add 3.0 3.0
== -3.0

>> add subtract 3.0 3.0 3.0
== 3.0

I think, that counts as a difference

oups :)

Do you think that parsing reversal would fix the problem

not completely

you need to realize, that the ** operator is right-associative

I should review some basic math stuff :))

Ladislav... according to wiki, addition, subtraction, multiplication 
and division are left-assiciative... http://en.wikipedia.org/wiki/Operator_associativity

yes

Now that I take a paper and a pen... left-associativity is indeed 
natural :)

So natural that i forgot about the associativity when thinking on 
how to split an expression i successive blocks! So stupid!

:-D

Thanks for pointing this. Hopefully, I have been clever not to mark 
this version as 1.0.0 :)

I put my %evaluate.r script processing operators with different priorities 
and associativities to the

http://www.rebol.org/view-script.r?script=evaluate.r


as well, since I think there it will be more accessible to the public

Your script is so much more complicated :)... maybe why it works... 
Good idea to put on rebol.org!

A quick doc to show how to use it?

Much more complicated

 - do not forget it processes three different rule sets, each of which 
 uses different priorities and associativities.

How to use it: check the examples at the bottom

yes... the concept of "rule sets" is not straightforward :) what 
do you mean by this?

OK... very nice indeed! But can I use "unknown variable"

the rule sets are described in the comments to the SRA, STD and RLE 
functions

>> mold sra [3 - 3 + 3] translate
== "[subtract 3 add 3 3]"

>> mold std [3 - 3 + 3] translate
== "[add subtract 3 3 3]"

unknown variable
? you need to use the %use-rule.r script:

http://www.rebol.org/view-script.r?script=use-rule.r

I guess, that you actually want to use just the STD function

I will adopt your script! Impressive. But i am still wondering if 
with the strategy in my script it would be possible to implement 
associativity properly?

well, if you modify it appropriately, why not?

except, that it requires some care, since all the operators you use 
are left-associative, except for the exponentiation operator, which 
is right-associative

>> std [3 ** 3 ** 3] translate
== [power 3 power 3 3]

>> std [3 * 3 * 3] translate
== [multiply multiply 3 3 3]

THat's why parsing reversal and taking care of ** specifically might 
be a good start...

Parsing reversal does not help IMO, because you have to process both 
left- and right-associativity at the same time

Notice, that my script parses left to right

I think I found a way... see you later today (or tomorraow). Again, 
many thanks for your feedback and congrat for your impressive script

It was actually so easy to fix: 5 'append to change into 'insert... 
I still have to check this does not break anything before posting 
a new version on rebol.org.

Ladislav... thank you again for pointing this bug!

Voilà... I renamed my script as parse-expression.r and it is on rebol.org 
... http://www.rebol.org/script-information.r?script-name=parse-expression.r

I just had to repace 5 'append into 'insert, so it fixing this big 
did not brak the "philosophy" of the implementation. Ouf :)

Example:
>> do parse-expression "-(-3+2)+3-2+3-sqrt((1+2)**2)-2**3+34"
== 28.0

@Endo: I tried exponent with ^, but as it is the escape character 
for rebol, i consistently end up with an error. If enyone can help 
on this! 

For the english comment, I promise I will take the time to translate 
the french comments soon.

Nice work Francois. Good conversation here too, with important details 
brought out by Ladislav.

Yes... Ladislav reminds me some basic math! God, I felt so stupid 
about this associativity bug! 

The reason why I developped parse-expression.r is because I need 
it to build an companion app for one of the best math book: Calculus 
3d edition from Smith & Minton! For now, I have developped a rebol 
library to transform any vid face into a function plotter, and parse-expression.r 
allows me to use human readable expression in the gui instead of 
guru rebol code :)

Ladislav:

How would you interpret x**-1/2 :
(1) [divide power x -1 2]
or 
(2) [power x divide -1 2]


A previous version of parse-expression.r returned (2)... but i considered 
this as a bug and changed it for (1) (see history 0.9.2 in the header 
of the script). But now, with our discussion on associativity, i 
am not sure anymore... 

Thanks for your help!

Precedence of ** is higher than /
The right form should be (1)

Thanks, this confirms my feeling

the associativity concern operators with same precedence.
** is right associative, the other ones are left associative

So, 2 ** 3 ** 4
is evaluated from rigth to left

like
2 ** (3 ** 4)