Need C# help: how do I do this (arrays)?

[coolsti]

I have data in a MySQL database that I need to pull out into C# arrays (or hashes or whatever) and I cannot find the best way to do it, and also have some problems with syntax.

The data represents a 2 dimensional mapping of chips on a relatively round wafer, and each row of the returned data will contain the following:

x coordinate, y coordinate, and the chip type at this x-y position.

When I do this in PHP, I simply create a two dimensional array on the fly:

mydata[x coordinate][y coordinate] = chip type.

Because of the rounded shape of the mapping, the result would not be a completely filled rectangular 2 dimensional array. But this is only a minor point as long as I can fill the "empty" locations with a NULL chiptype.

I do not know before getting the data out of the database what the overall dimensions of the mapping will be (maximum x and maximum y values).

How can I do this in C#? The MySql connector does not give the total number of rows returned from the select query, I would need to increment a counter as I read the data row by row.

I would like the data to be in the form of a two-dimensional array, either rectangular or jagged, or in any other container that gives me the ability to navigate through the mapping by looping over x and y in the fashion of one foreach loop embedded in another.

Thanks for any help!

[cloud255]

I
x coordinate, y coordinate, and the chip type at this x-y position.

this sounds like one object with 3 fields: x, y and a 'chip'

When I do this in PHP, I simply create a two dimensional array on the fly:

mydata[x coordinate][y coordinate] = chip type.

this sounds like a chip object consists of a xy co-ordinate pair.

maybe i'm just not reading it right but could you be more clear on this point?

[coolsti]

I guess you can say it is an object with three fields, an x coordinate, a y coodinate, and a chip type.

To be more clear, let us say I represent the chip type by an integer, then some data may look like this, each line showing x : y : chiptype

1 : 34 : 9
1 : 35 : 9
1 : 36 : 8
2 : 31 : 8
2 : 32 : 9

etc.

The point is, I need the data retrieved and placed in a container that allows me to look up any individual value as you normally would do in a two dimensional array, for example mydata[2][31] will give me the value 8 using the above data.

I also need the data in a fashion that I can traverse the entire mapping row by row, as here if this were legal in this language:

foreach ( "loop over the x coordinates" ) {
   foreach ( "loop over the y coordinates") {
       .... loop code ....
   }
}
 

This is very easy to do in PHP, but that is an interpretive script language.

My question is now to (best) do it in C#?

[cloud255]

dont know if this is the best way,

i would use ArrayLists. these allow you to create dynamic size arrays at runtime. this class accepts argumetns of type OBJECT so can put an integer in without any problems, you'll just have to cast back to an int when reading out of the array list.

this method creates one single dimension arrayList inside another one. there is a problem in that C# doesnt realize that you are making a pseudo 2D array so some additional casting is involved.

you will need to include "using System.Collections;"

assing data:

ArrayList x, y = new ArrayList();
 
y.Add(yourChipValue);
 
x.Add(y);

access data:

int myData = Convert.ToInt32(((ArrayList)x[i])[k]);

i think this will work.

good luck

[coolsti]

Thanks! I will give it a try.

I also have tried something perhaps not very efficient when you think about it:

int idchip;
string chipname;
int xpos, ypos;
int curx = 0;
Hashtable layoutrows = new Hashtable();
Hashtable cols = new Hashtable();
 
int i = 0;
int xmax = 0;
int ymax = 0;
 
// Here, already have performed the query with MySqlDataReader reader. 
// Using order by clauses, the data is ordered by the x and y coordinates 
// in ascending order. Now retrieve the data row by row.
while (reader.Read() != false)
{
                // get data and fill arrays
                idchip = reader.GetInt32(0);
                xpos = reader.GetInt32(2);
                ypos = reader.GetInt32(3);
 
                // find the maximum x and y coordinates in this mapping
                xmax = (xmax < xpos) ? xpos : xmax;
                ymax = (ymax < ypos) ? ypos : ymax;
                if (xpos != curx)
                {
                    // A new x coordinate is found, meaning a new row in the
                    // mapping. Copy the hashtable built up so far into the
                    // current "mapping row" hashtableand start with a new hashtable
                    // for the next mapping row.
                    if (curx > 0)
                    {
                        //Note: first coordinate values in the database are 1, not 0, so
                        // do not do this if the curx counter is equal to 0
                        layoutrows[xpos] = cols.Clone();
                        cols = new Hashtable();
                    }
                    cols[ypos] = idchip;
                    curx = xpos;
                }
                else
                {
                    cols[ypos] = idchip;
                }
                i++;
            }
            reader.Close();
 
            // save the number of chips for later use
            numchips = i;
 
            // Now get all the data out of the hashtables and put into a
           //  rectangular array.
            int[,] layoutarray2 = new int[xmax+1,ymax+1];
            foreach (int x in layoutrows.Keys)
            {
                Hashtable hs2 = (Hashtable)layoutrows[x];
                foreach (int y in hs2.Keys)
                {
                    layoutarray2[x,y] = (int)hs2[y];
                }
            }
 

In the above, I decided it may be far easier to work with the data later, if it was in the form of a rectangular array. If a coordinate x,y pair is not in the database, then the chiptype gets set to zero.

The inefficiency here, is that I first pull out all the data as hashtables, and then take the data from the hashtables to place it in the rectangular array.

The reason I cannot put the data into a rectangular array initially is that I do not know before I pull out all the data what the values for xmax and ymax will be. I could get this with an additional database query, but I would rather not.

Would maybe getting the data initially into an ArrayList be better than the hashtables like I am doing here?