Variable does not right!

jstrout

When an error is reported, it will usually include the line number (and if it's not in the main source file, it will include the source file name too). On occasion it can't figure this out, but such occasions should be rare.

There is no comprehensive list of error messages.

There are many MiniScript variants, as it is (like Lua) a language designed to be embedded into other apps. However around here we mostly discuss two: command-line MiniScript, and the MiniScript that is built into Mini Micro. The language of these is identical, but Mini Micro has additional functions and classes for doing all its cool stuff.

EDIT: Then there's also the Try-It! page, which is really command-line MiniScript but without any access to such things as files.

neku

"Index Error (list index 8 out of range)"
this rare "opportunity" has just succeeded. i am a lucky man : ;-)

i thought, i first want to master the ms-language to some extent before i get involved in something bigger. that's why i'm still waiting with micro.

i obviously have trouble with the formating of values with the concatenator: "+"
probably sums are formed in the worst case.

jstrout

So how do you do that? What must I type to see that index error?

neku

well... i got a little further... (but not much).

it looks like the x/y-values are overflowing somehow. the summation is also a problem for me. but in this case, it might be something else. the loop? i don't know... but i'm afraid it might be (again) something like my first message in this forum - that i missed some lines (i have a little problem with my eyes). if it is so, i apologize in advance.

the program is including (marked) debug-prints:

// mnight's tour 	( kt-022a-py.txt )
// quell: python with warnsdorf 
// extended with coin toss
// datum: 210517 - 22:30

n = 8    
cbx = n-1					// width and 
cby = n-1    				// height of the chessboard

cb = []					// make chessboard

for j in range(0, n-1)		// cb fill with zeros
    cb.push([0, 0, 0, 0, 0, 0, 0, 0])
end for

// show chesboard (empty)
prnt = function(cb)
	for x in range(0, n-1)
		print cb[x]
	end for
	end function
prnt(cb)					// show empty chesboard

//between inclusive
between = function( x, mi, mx )		// mi: min  mx: max
	between = (x >= mi) and (x <= mx);
	return between
end function
	

// directions the Knight can move on the chessboard
dx = [-2, -1, 1, 2, -2, -1, 1, 2]
dy = [1, 2, 2, 1, -1, -2, -2, -1]

// start the Knight from field
kx = 7
ky = 7

print "Startfeld: "+ kx +"/"+ ky
k = 0

for k in range(0, 63)      		// check all the fields
    cb[ky][kx] = k + 1	 		
    pq = [] 				// priority queue of available neighbors
	
    for i in range(0, 7)       	// Check thatone for usability (n=8)
    
		nx = kx + dx[i]
        ny = ky + dy[i]
	// *********************** debug	
	print "nx/ny: "+ nx
	print "nx/ny: "+ ny
	// **********************
        // between-1
		//between( x, min, max )		// check all edges
		if between( nx, 0, cbx ) and between( ny, 0, cby ) then
            if cb[ny][nx] == 0 then
                // count the available neighbors of the neighbor
                ctr = 0

                for j in range(0, 7)  	// Check thisone for usability
                     ex = nx + dx[j]
		     ey = ny + dy[j]
	// *********************** debug
	print "ex/ny: "+ ex
	print "ex/ny: "+ ey					
	// *********************** 
                    // between-2
		if between( ex, 0, cbx ) and between( ey, 0, cby ) then
			// increment the counter (pq)
                        if cb[ey][ex] == 0 then ctr = ctr + 1 //; end if
                end if //between-2
                end for //j

                //pq.push(ctr+"/"+i)		// Beware; is a string !! 
		pq.push([ctr, i])                        //  .. as number
                if len(pq) == 0 then
                    print "Field error: "+ k
                end if //len-pq
				
			end if //cb[ny][nx] == 0 
        end if // between-1
	end for // i

    // move to the neighbor that has min number of available neighbors	
    if pq.len > 0 then
        minVal = 7
        minD   = -1						// necessary?			
		
        for dd in range(pq.len-1)
			//pq.sort(0)				// necessary?			
			p = pq[0][0]
			m = pq[0][1]
			
            rdm = floor(rnd * 9)		// coin toss; random
            if p == minVal and rdm < 5 then		// take it.. or not ;-)
                minVal = p
                minD   = m
	    end if
            if p < minVal then
                minVal = p
                minD = m
            end if 
        end for //dd
		
        m = minD
        kx = k + dx[m]		
        ky = k + dy[m]
	// 	error?	>>>------->  kx & ky becomes > as 8
		
	// *********************** debug
	print " --- total: 64 current: "+ k
	print " - ky: "+ ky 
	print " - kx: "+ kx	
	// *********************** 
	
	end if // pq-len
	//x = key.available
end for //k

prnt(cb)		// show result
// ------------------------------------------------------------- pgm end

jstrout

OK, when I run that, I get

Runtime Error: Index Error (list index 9 out of range) [line 43]

Line 43 is the middle line here:

for k in range(0, 63)      		// check all the fields
	cb[ky][kx] = k + 1	 		
	pq = [] 				// priority queue of available neighbors

I see your "error?" note around line 115, and I think it's correct. You're adding dx and dy here, without checking that these values stay within range. Right before the error message, I saw a debug output of "ky: 9" which is of course out of range for a chess board.

neku

hello joe and thanks for the suggestion.

for me it looks like this... (without lines)
excerpt from cmd-window:
...
nx/ny: 7
nx/ny: 7
nx/ny: 8
nx/ny: 8
--- total: 64 current: 9

ky: 7
kx: 8
Runtime Exception: Index Error (list index 8 out of range)

Process Exit Code: -1
Time Taken: 00:01
...

not like yours:

Runtime Error: Index Error (list index 9 out of range) [line 43]

the "knight's tour" is a test-script which i have already done (in this way) in several other languages. in none of them i would have to check the kx/ky values explicitly. that's why i find it difficult to explain why this would be necessary in ms.

is it because of the loops? would it be necessary to abort them (in ms) in time (forcibly)?

ok... we are in any case one step further... thank you. there are obviously still other cracknut waiting... (as you surely have already seen).

startfield: kx: 1 ky: 1
[0, 0, 0, 0, 0, 0, 0, 0]
[0, 1, 0, 0, 0, 0, 0, 0]
[0, 0, 0, 0, 0, 0, 0, 2]
[4, 0, 0, 0, 0, 0, 0, 0]
[0, 0, 0, 0, 0, 0, 0, 0]
[0, 0, 0, 0, 5, 0, 0, 0]
[0, 0, 0, 7, 0, 0, 0, 0]
[0, 0, 3, 0, 8, 0, 10, 64]

:-O

but that's good - it's better than sudoku in any case! :-D

ps
yes.. the creation of a topic: 2darray would be good. how to dislocate the 2d specific contributions into the new category?

pps
a knight's tour category would also make sense... wouldn't it?

by which I mean to say: I would rather leave the forum structure to you ;-)

jstrout

I'm not sure what you mean by "in none of them I would have to check the kx/ky values explicitly." What other languages are these, and what do they do when you index outside the bounds of an array? I have used many languages, and the only one I can think of that wouldn't throw an error in this case is C, but in that case it is just as likely to crash your program or cause unpredictable behavior. Getting errors when you exceed the bounds of an array is a good thing, and all modern languages do it.

by which I mean to say: I would rather leave the forum structure to you ;-)

Just use the "Start a Discussion" button whenever you're changing topics. I can't fix it after the fact.

neku

which languages were they?
some... i like to compare the usability of the languages. it's a free time activity for me.

i won't list all of them (there are really many) only the ones from the last time: easyLang and ring. maybe also basic256 which i always use when i need a step by step execution with debuger.

because of your answer i compared the kx/ky values in the ring and in the easyLang. in easyLang the values overflow similar to ms, but without index errors. in the ring it ran absolutely without overflow. where in the ring also index errors are checked !

i have now also looked at it in the basic256... the kx/ky-values also stay there in the given frame - without overflow.

ps
gladly i send you the ring-print and also the code, if it interests you. have inhibitions to post it here... it is already so pretty overloaded.

neku

neku kx/ky-values also stay there in the given frame - without overflow.

it should mean that the kx/ky values do not exceed the range 0-7.
I apologize for my "google-english" .. I check it there & back .. but there are still misunderstandings. Unfortunately.

jstrout

Interesting. I haven't used any of those. I don't understand what "stay there in the given frame - without overflow" means. If your array only goes from 0 to 7, and you tell it to stuff a value into element 9, what does it do? Just ignore it (fail silently)?

neku

the problem with the kx/ky index makes me desperate. to distract/reward myself a bit, i did something else... got the GODOT endgine. i have had it in mind for a long time, since it takes into account practically all target platforms (ux, apple, windows.. android).

first i constructed a chessboard graphic and then i ventured on the knight jumps. now, after an intensive afternoon, with constant searching in the documentation, it is ready; it already runs very far (up to field 42..). I am quite satisfied.

what i wanted to say; the kx/ky values stay within the set limits of 0-7.
see excerpt:
...
2/4, 4/5, 3/3, 2/5, 4/4, 5/2, 6/4, 5/6, 7/5,5/4, 4/2, 3/4, 5/5,
3/6, 1/5, 2/7, 4/6, 6/5, 5/3, 3/2, 5/1, 6/3, 7/1, 5/0, 6/2, 4/1,
2/2, 1/4, 2/6, 0/5, 1/3, 2/1, 0/2, 1/0, 3/1, 1/2, 0/4, 1/6, 3/7,
...

jstrout

What the kx/ky values do is up to your code. What I'm asking is how they deal with out-of-bounds errors. I virtually guarantee you that if you try to assign to cb[ky][kx] where cb is a 2D array of 8 rows and columns, and kx or ky is something like 9, you will get an error (just as in MiniScript).

Perhaps you are trying to say: your Godot version of the code doesn't contain a bug which causes kx and ky to go out of range, and your MiniScript version of the code does. That may be. I couldn't follow the code well enough to understand exactly what it was trying to do and why the values are going out of bounds. But the fact that they do so is a result of what you have written, and not the fault of MiniScript itself.

neku

hello joe

i don't really know how to answer you. it seems to me that you take my intervention as an attack on you and the MINISCRIPT.

that would be wrong. i like the language. that's why my involvement here.

i have already done the "knight's tour" in various other languages. they were:

rfo-basic
X11 basic
python
lua
rebol
algoid
euphoria
operno
lobster
smallbasic
basic256
ring
easylang online
godot
...
and probably 2-3 more.

the procedure has always been as I have posted here.

if i am correct, easylang is the only one in which the kx/ky values overflow the range 0-7 (but) without effect. the graph appears correct. i cannot judge if it is legal. i discovered the overflow only now in the course of this discussion. i would not have to deal with it in the least.

i have not looked at all of the ones listed here... but i assume that the 0-7 limit is observed. it also seems logical to me.

so i don't understand why i should do it in ms. the instruction is clear: if between( nx, 0, 7 ) and between( ny, 0, 7 ) this works correctly for all languages listed above.

why this should be different with miniscript i don't understand (with my modest knowledge). it seems really mysterious to me why i should care about more than the definition of - BETWEEN.

or did i miss something? is the ms code incorrect?

it would be up to you to explain it and thank you in advance.

jstrout

No, I'm not taking it as an attack, nor am I attacking. I'm just trying to understand. You perceive some important difference that I don't think exists, and I'd like to understand why.

Your between function looks correct. If indeed you're using this around references to cb, then I agree it should work. The formatting of the code above is pretty messed up, so it's hard for me to see if that's the case. And of course the code has probably changed a bit since then anyway.

Can you post the current version of your code? I'd really like to get to the bottom of this.

neku

well... the prerequisite is that there is no error in my code.

the difference i see is that the code works satisfactorily (in the same way) in various other languages - but not in ms.

the code formatting gets a bit messed up when posting here. it would be better to attach it as a zip. is that possible somehow?

the code has not changed in the meantime... and is (for debuging) usable (i checked it).

in my opinion, the error is not in the between and also not in the concatenations.. but (maybe) in the loops (I compare it in python 2.7).

i'm still looking.. it's very grueling... too many numbers. you get tired quickly

jstrout

Well, that's the thing. There is an error in your code. I'm just not sure exactly what it is because I don't understand how it's supposed to work.

It's clear that lines 105-106 (a little past that in your version — I have shortened a bit) are adding dx[m] and dy[m] to k, and this is resulting in values that are out of bounds. There is no between check here. But I don't know what k is, why you are adding dx and dy to it, and what correct could would be doing instead.

I see your for dd loop is indexing over your pq list. Note that a slightly more idiomatic way to do that is for dd in pq.indexes. But you're not actually using this dd index inside your loop; instead you're looking at pq[0] every time. I think this is a bug, but probably not the bug.

I suspect the bug is this: you're taking k, which appears to be a 1-dimensional position number from 0 to 63, and adding dx/dy values to it, which are 2D offsets in the range -2 to 2. And then treating the result as if they were x and y positions (kx and ky).

		kx = k + dx[m]		
		ky = k + dy[m]

That makes no sense. If I change these lines to

		kx = kx + dx[m]		
		ky = ky + dy[m]

Then the program completes without errors, and finds 8 different tours.

jstrout

Incidentally, here is my slightly modified version of your code.

// mnight's tour 	( kt-022a-py.txt )
// quell: python with warnsdorf 
// extended with coin toss
// datum: 210517 - 22:30

n = 8    
cbx = n-1					// width and 
cby = n-1    				// height of the chessboard

cb = []					// make chessboard

for j in range(0, n-1)		// cb fill with zeros
	cb.push([0, 0, 0, 0, 0, 0, 0, 0])
end for

// show chesboard (empty)
prnt = function(cb)
	for x in range(0, n-1)
		print cb[x]
	end for
end function
prnt(cb)					// show empty chesboard

//between inclusive
between = function( x, mi, mx )		// mi: min  mx: max
	between = (x >= mi) and (x <= mx);
	return between
end function


// directions the Knight can move on the chessboard
dx = [-2, -1, 1, 2, -2, -1, 1, 2]
dy = [1, 2, 2, 1, -1, -2, -2, -1]

// start the Knight from field
kx = 7
ky = 7

print "Startfeld: "+ kx +"/"+ ky
k = 0

for k in range(0, 63)      		// check all the fields
	cb[ky][kx] = k + 1	 		
	pq = [] 				// priority queue of available neighbors
	
	for i in range(0, 7)       	// Check thatone for usability (n=8)
		
		nx = kx + dx[i]
		ny = ky + dy[i]
		// *********************** debug	
		print "nx/ny: "+ nx + "/" + ny
		// **********************
		// between-1
		//between( x, min, max )		// check all edges
		if between( nx, 0, cbx ) and between( ny, 0, cby ) then
			if cb[ny][nx] == 0 then
				// count the available neighbors of the neighbor
				ctr = 0
				
				for j in range(0, 7)  	// Check thisone for usability
					ex = nx + dx[j]
					ey = ny + dy[j]
					// *********************** debug
					print "ex/ey: "+ ex + "/" + ey
					// *********************** 
					// between-2
					if between( ex, 0, cbx ) and between( ey, 0, cby ) then
						// increment the counter (pq)
						if cb[ey][ex] == 0 then ctr = ctr + 1
					end if //between-2
				end for //j
				
				//pq.push(ctr+"/"+i)		// Beware; is a string !! 
				pq.push [ctr, i]                        //  .. as number
				print "Pushed onto pq: " + pq[-1]
				
			end if //cb[ny][nx] == 0 
		end if // between-1
	end for // i
	
	// move to the neighbor that has min number of available neighbors	
	if pq then
		minVal = 7
		minD   = -1						// necessary?			
		
		for dd in pq.indexes
			//pq.sort(0)				// necessary?			
			p = pq[dd][0]
			m = pq[dd][1]
			
			rdm = floor(rnd * 9)		// coin toss; random
			if p == minVal and rdm < 5 then		// take it.. or not ;-)
				minVal = p
				minD   = m
			end if
			if p < minVal then
				minVal = p
				minD = m
			end if 
		end for //dd
		
		m = minD
		kx = kx + dx[m]		
		ky = ky + dy[m]
		// 	error?	>>>------->  kx & ky becomes > as 8
		
		// *********************** debug
		print " --- total: 64 current: "+ k + "  kx:" + kx + "  ky:" + ky
		// *********************** 
		
	end if // pq-len
	//x = key.available
end for //k

prnt(cb)		// show result
// ------------------------------------------------------------- pgm end

neku

opps.. that hurts!
:-(
I could have saved myself that!
i formally apologize.
thank you for the clarification.

thanks to the correction in the "pq" treatment;

for dd in pq.indexes		
	p = pq[dd][0]
	m = pq[dd][1]
	...

the "coin toss" now also works!

thanks for the lesson joe.

ps
i will try to explain the process. then i will post it.

« Previous Page