Multiple forms and datagridviews

[bwood2020]

Hi Henry, I have made the recommended changes. Also, I don't think using DataSet for the GetDocs property was correct so I changed it to DataTable. Here is what I have: Form1

string Query = ("SELECT * FROM [" + dataGridView1 + "]");
DataTable Sds = new DataTable(Query);
DataRow myNewRow;
myNewRow = Sds.NewRow();
Sds.Rows.Add(myNewRow);

private DataTable sds;
public DataTable Sds
{
get
{
return sds;
}
set
{
sds = value;
}
}

Form2

GetForm1 = new Form1();
object GD = GetForm1.Sds;
this.dataGridView1.DataSource = GD;

modified on Monday, May 18, 2009 1:02 PM

[Henry Minute]

Hi, have a good weekend? The problem is, that there are still two members with the same name, which could get confusing. However if the compiler doesn't complain, why should I. :) The next problem to tackle is the line string Query = ("SELECT * FROM [" + dataGridView1 + "]"); I don't know if you know how to use breakpoints, the debugger and single stepping, but if you do you will be able to determine that it evaluates to

SELECT * FROM [System.Windows.Forms.DataGridView]

at runtime. Whereas I suspect that you want it to get the data from the table whose name is selected in dataGridView1. Is that correct?

Henry Minute Do not read medical books! You could die of a misprint. - Mark Twain Girl: (staring) "Why do you need an icy cucumber?" “I want to report a fraud. The government is lying to us all.”

[bwood2020]

I had a very productive weekend cutting down dead trees in my backyard. Not something I like spending time on but at least it saves me a few bucks. How about you? I didn't use breakpoints to see if the query was working properly but I did throw in a MessageBox.Show(Query) to evaluate the string and it does say "SELECT * FROM [System.Windows.Forms.DataGridView] so I assumed it was correct. What I would like to do is run a query on the dgv in From 1 to populate the dgv on form 2. From there it would populate a new data table and would be used as the data source for the dgv on Form 2. Right now I have it selecting all just to get the dgv on Form 2 to populate. Later the query will include a Select Distinct. Do you think this is a good way to do this or is there a more efficient method to code this? Thank you, Brenton

[Henry Minute]

bwood2020 wrote:

Right now I have it selecting all just to get the dgv on Form 2 to populate. Later the query will include a Select Distinct. Do you think this is a good way to do this or is there a more efficient method to code this?

This is a valid approach as 'proof of concept', and for now I would leave it like that, then, as you say, refine it later. The problem with SELECT * FROM [System.Windows.Forms.DataGridView] is that System.Windows.Forms.DataGridView is not a table in a database and therefore there is nothing there to SELECT. However even if this worked, the data retrieved is not being assigned to private DataTable sds;, which is where Form2 is looking for its data. This problem can be solved with one simple change Your code

string Query = ("SELECT * FROM [" + dataGridView1 + "]");
DataTable Sds = new DataTable(Query);
DataRow myNewRow;
myNewRow = Sds.NewRow();
Sds.Rows.Add(myNewRow);

Changed code

string Query = ("SELECT * FROM [" + dataGridView1 + "]");
sds = new DataTable(Query); <============= Here is the change, get rid of 'DataTable Sds' and replace it with just 'sds'
DataRow myNewRow;
myNewRow = Sds.NewRow();
Sds.Rows.Add(myNewRow);

That way you would be assigning the results of the query to the field that Form2 is using. However, I do not believe that change will cause Form2s dataGridView1 to populate as there is no data being returned from your query. All that

sds = new DataTable(Query);

does is create a new DataTable with its Name set to "SELECT * FROM [System.Windows.Forms.DataGridView]". That DataTable with the very long name is not attached to anything, it is not part of your database, it is an orphan. Overall the problem is that Query is not being executed. So we need to get a correct string into Query and then get your query to execute and return a result. Can I suggest that you look at look at Executing a Query that returns Rows[^], scroll down till you get to the section Executing SQL Statements that Return Rows Using a Command Object, which is what I think will be the best way for you to do it. For now, where it shows 'SELECT * FROM customer' replace customer with the name of one of your tables (I beli

[bwood2020]

I am caught up on the connection string and query for this. I should have made this clear in the beginning so I don't waste your time. The data source/s for this application comes in a file (.xls, .txt, and .csv). There is no SQL Server connection to import data (All the data will be pushed to a staging environment once the data has gone through the app). What I would like to do is connect to the original dataTable that is the source for my dgv on Form 1 and use it as the source for the dgv on Form 2. Is the connection possible? If I connect to the file then the query will fail once I add Distinct because not all headers will be the same and could come in all different ways. I take care of this by allowing the user to select a new header. The addition of the dgv on Form 2 is kicked off once a button has been clicked and that is where all of this comes into play. I went to &lt;a href="http://www.connectionstrings.com/"&gt;&lt;/a&gt; and I could not find anything that connects to a DataTable. It might be because I'm not looking in the correct place or may be overlooking something. Also, I used product as an example and now I see I should not have done that because it does look like I'm trying to query a table in the database. I hope I have not confused you. If I have I apologize in advance. If you need more information about what I'm doing then let me know.

[Henry Minute]

OK. Theoretically you can access the the DataSource from dgv1 on Form1 simply by using dataGridView1.DataSource, in which case putting this into your original Query, instead of just dataGridView1, might just work. Suck it and see. If not, then if you could post the code that loads the data into dgv1 on Form1, so that I can advise you further. BTW If you are dealing with .csv and .txt files, you might find FileHelpers v2.0 - Delimited (CSV) or Fixed Data Import/Export Framework[^] from here on CP useful. It won't help with the .xls files though.

Henry Minute Do not read medical books! You could die of a misprint. - Mark Twain Girl: (staring) "Why do you need an icy cucumber?" “I want to report a fraud. The government is lying to us all.”

[bwood2020]

Good afternoon Henry, I have tried to run the query using datagridview1.DataSource but doesn't seem to be working. Here is the code I run for csv files. Note that this doesn't show the entire method. In this method there are connections for the other file types. <pre> else if (comboBox1.Text == ".csv")                   {                               //Get file path from textbox                               String FilePath;                               FilePath = textBox1.Text;                               //Connect to csv file                               String comString = @"Provider=Microsoft.Jet.OLEDB.4.0;Data Source=" + Path.GetDirectoryName(FilePath) + @";Extended Properties=""Text;HDR=YES;FMT=Delimited;IMEX=1\""";                               String strSql = "SELECT * FROM [" + Path.GetFileName(FilePath) + "]";                               OleDbConnection conCSV = new OleDbConnection(comString);                               conCSV.Open();                               OleDbCommand dbCommand = new OleDbCommand(strSql, conCSV);                               OleDbDataAdapter dAdapter = new OleDbDataAdapter(dbCommand);                               DataTable dFill = new DataTable();                               dAdapter.Fill(dFill); &nbs

[Henry Minute]

Thanks for posting some of your code. From a quick read through, it seems as if there is some redundant code in there. To help in your future development adventures please try this for me. Remove or comment out the following lines (commenting is probably better till you confirm it is OK):

//create new dataTable and add new top row for new headers
DataTable newDataTable = new DataTable();

                          foreach (DataColumn col in dFill.Columns)
                          {
                                try
                                {
                                      newDataTable.Columns.Add(new DataColumn(col.ColumnName));

                                }
                                catch (Exception)
                                {

                                }
                          }
                          foreach (DataRow row in dFill.Rows)
                          {
                                string\[\] rowarr = new string\[dFill.Columns.Count\];

                                for (int k = 0; k < dFill.Columns.Count; k++)
                                {
                                      rowarr\[k\] = row\[k\].ToString();
                                }
                                newDataTable.Rows.Add(rowarr);

                          }


                          DataRow NewRow;
                          NewRow = newDataTable.NewRow();
                          newDataTable.Rows.InsertAt(NewRow, 0);

As I am not sure of the purpose of the last two three lines of code, you might need to keep them, just change newDataTable to dFill then change this row as indicated:

             // bindingSource1.DataSource = newDataTable;
             bindingSource1.DataSource = dFill;

This should give the same results as before. Have a look and let me know, since only you can tell if it is different. What I then need to know is what you need to happen when when the user double-clicks on dataGridView1 (you are still doing it that way aren't you? Tell me if that has changed.). For example, something like "dgv1 has columns for itemName, itemCost and discount (no need to list them all, just enough for me to get an idea what it's about). When dgv1 is double-clicked I want form2 to open and display information abou

[bwood2020]

You suggestion worked. All seems to be working the same after testing. Thank you for cleaning up the code. Once the data has been loaded into the dgv on Form 1, the user clicks on the dgv and changes the field headers. This helps standardize the file so the ETL package doesn't break. Once the new field headers have been selected, the user will then click a button. Form 2 is then activated and data has been loaded into different dgv's. dgv1 will be populated with Documentation Type data. Dgv2 will be populated with Purpose Type data and so on for dgv3, dgv4... So what it comes down to is that each dgv on Form 2 will be populated from a single field from the dgv source on Form 1. Each dgv on Form 2 will have three columns (the data from source, a drop down to standardize the data from source, and the count of each distinct value). Once the data has been standardized on Form 2 then the data in the dgv on Form 1 needs to be updated. I am open to any suggestions you may have if there is an easier way to do this. The requirements however still recommend I use a second form to provide translations of data.

[Henry Minute]

The second Form method seems to be as good as any other, from what you have told me. So on that basis. What happens when the user clicks the button, depends on whether there will only ever be one instance of Form2 where the data displayed will change depending on whatever, or whether there will be more than one instance, each one created specifically to display one set of data. If there will only be one Form2, in Form1 add

private Form2 docsForm = null;
private Form2 DocsForm
{
    get
    {
        if (this.docsForm == null)
        {
            this.docsForm = new Form2();
        }

        return docsForm;
    }
}

then whenever Form1 needs to refer to the Form2 instance just use DocsForm (the property name) NOT docsForm (the field name). This property uses what is known as 'Lazy Initialization', you can look it up later. It just means that the thing (Form2 in this case) only gets created when it needs to be. The rest of the time the existing instance is used. I have made it private deliberately. If there will be many instances of Form2 they will be created inside the click handler for the button, so the above will not be needed. The button click handler, just for illustration purposes I have called it btnShowDocs

	private void btnShowDocs\_Click(object sender, EventArgs e)
	{
                    // \*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*
                    // Only if there are to be many Form2s
                    // \*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*
		Form2 DocsForm = new Form2();
                    // \*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*
                   
                    DocsForm.DataKey = this.dataGridView1.CurrentRow.Cells\[0\].Value.ToString();
                    DocsForm.Show();  // or .ShowDialog(), depends on whether need access to Form1 again before Form2 closes 
	}

There has to be some piece of data in Form1.dgv1, that will enable Form2 to decide the correct data to show, purely for illustration I have said that that data is the value of the first column in the selected row of dgv1. But you will have to let me know what that piece of data is. It could be the value of a cell in dgv as above, or the whole row, each cell in the row determining the data for one of the dgvs of Form2. If the trigger data is not in dgv1, then what is it? Then for Form2, something like:

pu

[bwood2020]

Yes, there will only be one instance of Form 2. I have dragged and dropped a TabControl onto Form 2. Each tab contains a dgv which will be populated with a column from the dgv on Form 1. I have created a method in Form 1 that goes through each cell in row 1 and stores the new headers in an array. if it meets the criteria then it should load the Distinct values of that column into Form 2. Here is the code I have written but don't know if you can use it as it is. But maybe you can... <pre>public void button12_Click(object sender, EventArgs e)             {                                                             //MappingsForm is the name of Form 2                         MappingsForm.Show();                                  string[] HeaderArray = new string[dataGridView1.ColumnCount];                                           int n = dataGridView1.ColumnCount;                         int c = 0;                         for (c = 0; c < n; c++)                         {                               HeaderArray[c] = dataGridView1.Rows[0].Cells[c].Value.ToString();                                                             if(HeaderArray[c] == "DocType")                               {                       &nbs

[Henry Minute]

I have modified my suggestions from the previous post in light of this information, using your member names where I noticed them and some of your code, although I have suggested moving it to MappingsForm rather than having it in Form1. The reason is that in OOP programming, as far as is possible, each object (and a Form is an object) should be responsible for handling its own data and controlling access to that data. For Form1:

        private MappingsForm mappingsForm = null;
        private MappingsForm MappingsForm
        {
            get
            {
                if (this.mappingsForm == null)
                {
                    this.mappingsForm = new MappingsForm();
                }

                return mappingsForm;
            }
        }

    private void btnShowDocs\_Click(object sender, EventArgs e)
    {
            MappingsForm.DataKey = this.dataGridView1.Rows\[0\];
            MappingsForm.Show();
    }

    private void Form1\_FormClosing(object sender, FormClosingEventArgs e)
    {
        if (this.mappingsForm != null)
	{
	    this.mappingsForm.Dispose();
	    this.mappingsForm = null;
	}
    }

For MappingsForm:

public partial class MappingsForm : Form
{
	private bool dataDisplayed = false;

	public MappingsForm()
	{
	    InitializeComponent();
	}

	private void MappingsForm\_Load(object sender, EventArgs e)
	{
                if (this.dataKey != null)
                { 
	        this.LoadData();
                }
	}

	private void LoadData()
	{
	    // this is part of your code, slightly modified
                for (int c = 0; c < this.dataKey.Columns.Count; c++)
                {
                    if(this.dataKey.Cells\[c\].Value.ToString() == "DocType")
                    {
                        //pass data to dgv1
                    }
                    else if (this.dataKey.Cells\[c\].Value.ToString() == "PurposeType") // else here to avoid unnecessary processing
                    {
                        //pass data to dgv2
                    }
                    else if (this.dataKey.Cells\[c\].Value.ToString() == "SomeOtherType") // else here to avoid unnecessary processing
                    {
                        // and so on 
                    }

                    // This could also be written like this, so either have t

[bwood2020]

Thanks Henry for the short run down of your concerns. I will look into this... I agree with you about renaming the controls as descriptive as possible. As soon as I get this last part I will go back through and rename them. I implemented the code you wrote and I get the error 'System.Windows.Forms.DataGridViewRow' does not contain a definition for 'Columns'. This occurs at 'this.dataKey.Columns.Count' in the for loop on Form 2. I have tried looking this up and I get similar hits but what I have tried doesn't work. Any ideas?

[Henry Minute]

Oops. :-O I'm having a root round. I'll get back to you.

Henry Minute Do not read medical books! You could die of a misprint. - Mark Twain Girl: (staring) "Why do you need an icy cucumber?" “I want to report a fraud. The government is lying to us all.”

[Henry Minute]

Try changing Columns for Cells. It should work exactly the same. Only The DataGridView has Columns, each of it's rows contains a cell for each of the columns.

Henry Minute Do not read medical books! You could die of a misprint. - Mark Twain Girl: (staring) "Why do you need an icy cucumber?" “I want to report a fraud. The government is lying to us all.”

[bwood2020]

Hi Henry, I have tried to pass dataKey.Cells[c].Value a couple of ways but I am not having any luck. Here are just three ways I have tried that make sense to me but aren't working: dataGridView1.DataSource = dataKey.Cells[c].Value; dataGridView1.Rows.Add(dataKey.Cells[c].Value); dataGridView1.SelectedRows[c].Cells[c].Value = dataKey.Cells[c].Value; I have also tried adding ToString() to the end of Value and that isn't working either. I have spent a good amount of time doing research on how to pass the values of the cells to the dgv and I haven't found anything that would work. Do you have a little more time to help me out on this? Thank you, Brenton

[Henry Minute]

Hi Brenton, None of the options that you have tried will work. I'll try to explain. As we left it, MappingForm has a DataViewRow containing a series of cells, each cell holding something like "thisType" or "thatType". To populate each of your DataGridViews with relevant information you need to get it from a file, in the same way that you did for the dgv on Form1. Only you know which file contains data for "thisType" and which file holds data for "thatType". The if (dataKey.cells[c].Value.ToString == "thisType") lines (or the switch block, if that is what you used), simply enables you differentiate between the various 'Types' so that you know which to use the in the SELECT statement when retrieving the data into a DataTable/DataSet or whatever you decide to use as the datasource for your dgvs. Something like

SELECT * FROM TheThisTypeFile.csv
WHERE DataType == "thisType"

Obviously the WHERE clause is only required if that file holds data for various 'Types'. Hope that helps.

Henry Minute Do not read medical books! You could die of a misprint. - Mark Twain Girl: (staring) "Why do you need an icy cucumber?" “I want to report a fraud. The government is lying to us all.”

[bwood2020]

I am still confused because I would still have to know what the field names are called in the file. For instance, I have 2 files. One file may have 32 fields and the other 10 fields which come in different days (say one today and one tomorrow). In these columns we are looking for Loan Purpose Type and Loan Documentation type. In file one these fields are named Doc Type and Purp Type and in the second file they are named Documentation Type and Purpose Type. Also, in the where clause you would need to specify all the doc types. However, each client has there own doc types. One client may have say Full Doc, Limited Doc, and No Doc. Another client may have 1, 2, and 3. so it seems the query would have to change every time. Is this correct?

[Henry Minute]

Ah-ha! I am beginning to understand a little more about how this is supposed to work. If I am right, probably not, but here goes anyway, the purpose of the dgv in Form1 is to allow for translation/modification of what is in the files from client, to a 'standard' wording. If I have got that right, you have been doing things backwards! :laugh: Sorry, I shouldn't laugh, but it is just the sort of thing I do all the time. I will wait for your response before going further.

Henry Minute Do not read medical books! You could die of a misprint. - Mark Twain Girl: (staring) "Why do you need an icy cucumber?" “I want to report a fraud. The government is lying to us all.”

[bwood2020]

That is correct, with one exception. The only thing I want to accomplish in Form 1 is changing the field headers to a standardized naming convention. In Form 2 I want to standardize the data (but this only needs to be done to certain fields like Doc types). I will explain more to you so we are both on the same page. So we receive files on a daily basis from many different clients. These files contain different information in different columns with different field headers. We use SSIS to get them into our systems however, SSIS breaks every time we run the files through because of the differentiation between field headers and data. So, to get around this I am creating a prototype to standardize all the files which mainly include field headers and data related to some of the fields (i.e. Doc Type, Purpose Type, and so on). So For example, if the field header Doc Type comes in then in form 1 row 1 of the dgv there is a combo dropdown where the user can select the "New" header name. Then the second form needs to include the distinct data for the specific field headers so I can standardize them as well. This second form will activate upon clicking a button. Example: File 1: Loan Documentation Type(field name from file) = DocType(new header that will be selected in row 1 of the dgv on Form1) Low Doc, Full, and No Doc(Doc Types data from file) = L, F, ND(translate doc types to standardized naming convention in dgv on Form 2) File 2: Docs(field name from file) = DocType(new header that will be selected in row 1 of the dgv on Form1) 1, 2, and 3 (data from file) = L, F, ND(translate doc types to standardized naming convention in dgv on Form 2) So the files we get have 100's of loans. For this example, the field Doc Types can have many types (i.e. 50 Full docs, 25 Low Docs, and 25 No docs) which is why I would like to select Distinct values only having to map, in this case, 3 doc types rather then 100. I am sure I have not explained myself very well in previous posts so please let me know if you need additional information or if you are confused on anything I have said to this point.