Passing array info into a procedure... again

[Lucas]

What is it you are trying to achieve?
Im trying to pass the information of an array to a procedure. The array is part of a record, and theres a problem when i declare what part of the record im trying to pass to the procedure


What is the problem you are having?
I get -'points' cannot be followed by a '.'- and it highlights the line:
calcthetaR (points.x, points.z)
specifically highliting the '.' after points

Turing:

% variables for points type pointsdata :     record         x, y, z, X, Y, Z, t, r : real         dx, dy : int         dix, diy : real         selected : boolean     end record var points : flexible array 0 .. -1 of pointsdata var numpoints : int := -1     procedure calcthetaR (A, B : array 0 .. * of real)         for a : 0 .. numpoints             if A (a) > 0 and B (a) > 0 then                 points (a).t := arctand (A (a) / B (a))             elsif A (a) > 0 and B (a) < 1 then                 points (a).t := arctand (-B (a) / A (a)) + 90             elsif A (a) < 1 and B (a) < 1 then                 points (a).t := arctand (-A (a) / -B (a)) + 180             else                 points (a).t := arctand (B (a) / -A (a)) + 270             end if             points (a).r := sqrt (A (a) ** 2 + B (a) ** 2)         end for     end calcthetaR     calcthetaR (points.x, points.z)

Please specify what version of Turing you are using
4.1.1

[chrisbrown]

points is an array, not a record. You would have to manually aggregate all the x and y fields into new arrays. However, since points is global, you can remove A and B altogether and simply access the fields directly. A(a) is the same as points(a).x.

edit: And if you're just trying to eliminate duplicate code, just create variables at the beginning of the for loop storing the current x and y.

[Lucas]

K thx, is there anyway to make this work though? cause the reason i used A and B instead of points(a).x and points(a).z is because i do exactly the same thing three diff combos of points (a). x/y/z. possible or not?

[Tony]

Instead of an array of records, you could have a record of arrays. Of course that would make adding/removing individual elements a lot more complicated, but that's the tradeoff.

[chrisbrown]

Well, there's the way you want to hear, and there's the better way.

The short-term fix is to create two new arrays, let's call them A and B. For each element in points, assign the corresponding elements in A and B to the values of x and z respectively, for that particular entry. When finished, call calcthetaR(A, B).

The better way is to modify calctheta (drop the R) so that it returns the arctand of two real numbers. That way, you can do something like:

Turing:

for i ...     points(i).t := calcthetha(points(i).x, points(i).z)     points(i).r := ... end for

[Lucas]

@Tony: K thx, i might do that next time, but if i did that now id have to modify my whole code, so i dont think ill do it this time
@ chrisbrown: I thought about ur second idea, but it results in more lines of code, so i did it the way i did. and wouln't i just run into the same problem as before?
also, its called calcthetaR is cause theres a bit in there where calculates the radius

[chrisbrown]

You say you run this calculation over the combinations of x, y and z? How are you saving the values of t and r each time? If you run it for x, y and then y, z, then t and r have been overwritten.

Assuming you have a way to get around this (or I'm just missing something), here's a more complete sketch of what I'm thinking:

Turing:

function calctheta(a, b : real) : real         % code to calculate points(i).t but use the result keyword instead of := end function calcdistance(a, b : real) : real         % you already have this line somewhere, it just needs some tweaks end for i ...         % Three different combos. I've invented record fields to avoid overwriting data         points(i).t_xy := calctheta(...)        % x, y          points(i).r_xy := calcdistance(...)                     points(i).t_xz := calctheta(...)        % x, z          points(i).r_xz := calcdistance(...)              points(i).t_yz := calctheta(...)        % y, z          points(i).r_yz := calcdistance(...)      end for

It may seem repetitive but it's much easier on the eyes than needless temporary arrays. You can look at that and immediately understand what's happening.

[Lucas]

@chrisbrown:

[Lucas]

sry, accidentally hit the enter button a couple of times :/
thx, ill do it ur way then :p
and the reason i only have one variable for t and r is because i only ever need one r and t value at a time, i nvr actualy need seperate ones. just the way im doing it, i reset the t and r variables evrytime the user inputs the direction they want to spin the object. Im sure theres a better way to do this, i just do it this way because its the way that i figured out myself :p