Value of tag property disappears - c#

I'm busy with a simple application. It reads xml and puts the information in a treeview.
I do this by creating TreeNodes and nest them, and finaly, return the root treenode. Because I want to show some extra information when a treenode is selected, I put the information in the tag property of the TreeNode. In this way, I should be able to retrieve the information when the node is selected.
But when I try to retrieve the information in the Tag property, it says the value = null.
Here is the code where I fill the tag. This is in a function which is recursively used to read the XML dom. treeNode is a paramater given to this function.
if (treeNode.Tag == null)
{
treeNode.Tag = new List<AttributePair>();
}
(treeNode.Tag as List<AttributePair>).Add(new AttributePair(currentNode.Name, currentNode.Value));
This is the event where a treenode is selected
private void tvXML_AfterSelect(object sender, TreeViewEventArgs e)
{
if (tvXML.SelectedNode.Tag != null)
{
}
if (e.Node.Tag != null)
{
}
}
Both values evaluate to null. How can I solve this problem?

The code you posted should work as-is. Something else in your code, code that you didn't post here, is causing this to break. It could be clearing the Tag, it could be a data binding set on the tag, etc.
Without seeing all your code, the best I can do is guess and help you isolate the problem.
Here's what I'd do: setup Visual Studio to allow stepping into the .NET framework source code with the debugger. Then, set a breakpoint on the setter for the TreeNode.Tag property. After you set the tag in your code to your AttributePair List, see when it gets set again. The breakpoint will hit, you'll look at the stack trace and see what exactly is clearing your Tag property.

If using Tag property isn't in principle, I'm recommend inherit TreeItem:
public class MyTreeNode : TreeNode
{
public List<AttributePair> list;
public MyTreeNode (string text,List<AttributePair> list) : base(text)
{
this.list = list;
}
//or
public MyTreeNode (string text) : base(text)
{
this.list = new List<AttributePair>();
}
}
And use it:
private void tvXML_AfterSelect(object sender, TreeViewEventArgs e)
{
if (tvXML.SelectedNode is MyTreeNode)
{
MyTreeNode selectedNode = tvXML.SelectedNode as MyTreeNode;
selectedNode.list.Add(.., ..);
}
if (e.Node is MyTreeNode)
{
MyTreeNode node = e.Node as MyTreeNode;
node.list.Add(.., ..);
}
}

Maybe you are assigning the values after Select event. Otherwise you can maintain a dictionary of TreeNode and tag values as workaround.

Try declaring/initialising your List object somewhere above (outside of the inner scope you are in) and when you assign to the .tag property - don't create a new list but rather assign previously created List object.

private TreeViewItem _subsender;
private object _senderTag;
public TreeViewItem _sender
{
get {
return _subsender;
}
set
{
_senderTag = value.Tag;
_subsender = value;
}
}
Got the same problem this the solution that i found
Just don't use the .tag but _senderTag
(don't change the lines in the set for some reason :D )
(You cant just reset the tag (maybe new TreeViewItem ))

Related

C# ListView from Another Function

Please forgive me for such a stupid question. I am sure many of you will find this easy, where I have sent almost half the day reading trying to figure this out.
Here is the problem:
I have a FORM (Form1.cs) made. In that form I created a listview, and named it "ListView1".
Within the Form1.cs, I call a function called FileManager(this), where I pass in the THIS object.
In FileManager.cs I was able to listviewArray= originalForm.Controls.Find("listView1", true) and find that 'listview'.
When I do a listviewArray[0]<-- I can't seem to add a list to it.
FileManager.cs
FileManager(object sender)
{
if (sender != null)
{
originalForm = (Form)sender;
}
}
public void getFiles()
{
filePaths = Directory.GetFiles(hsocDir);
if(filePaths != null)
{
listviewArray= originalForm.Controls.Find("listView1", true);
if(listviewArray != null)
{
ListViewItem lvi = new ListViewItem("text");
// My Array is listViewArray
// How to add things to Lvi to it.
}
}
== Form1.cs
public Form1()
{
InitializeComponent(`enter code here`);
mysql = new MySQLCheck(this);
fileManager = new FileManager(this);
fileManager.getFiles();
}
You can't access element 0 of the collection because the collection is empty. To add an item, use:
listViewArray.Items.Add(lvi);
You need to modify the Items collection instead of the ListView itself for this to work, as ListView is not a collection (its a control).
listViewArray.Items.Add(lvi);
Also in your listview,setting this properties will help :
// Set the view to show details.
listViewArray.View = View.Details;
// Select the item and subitems when selection is made.
listViewArray.FullRowSelect = true;
// Display grid lines.
listViewArray.GridLines = true;

WPF iterate through datagrid

Using WPF C#.NET4.5 using visual studio 2012 ulti.
Old winforms code:
foreach (DataGridViewRow paretoRow in ParetoGrid.Rows)
{
if ((Convert.ToInt32(paretoRow.Cells["CurrentPareto"].Value) < (Convert.ToInt32(paretoRow.Cells["NewPareto"].Value))))
{
paretoRow.Cells["pNew"].Value = downArrow
}
}
As you can see each row I cycle through I check a specific cell, if true I then populate another cell. This was good old winforms code I used many times before...however.
Switching over to WPF was alot more different than i previously assumed.
DataGrid does not contain the Row property. Instead, I think you need to use:
DataGridRow paretoRow in paretogrid.Items
But im still at a loss on who to now get the cell.
So my question is, is there syntax changes to perform, if so where? Or as I'm beginning to believe datagrids in WPF operate with Objects more so than winforms thus not needing to use a propertie called "row", if this is the case what logic/syntax should i know use in this example?
Thanks for your patience guys, think when I go home for the bank holiday I'll do a bit of WPF digging to see how different it actually is.
People seem to be overcomplicating this, this worked for me:
foreach (System.Data.DataRowView dr in yourDataGrid.ItemsSource)
{
MessageBox.Show(dr[0].ToString());
}
I think first think you want to do is to get all rows of your DataGrid:
public IEnumerable<Microsoft.Windows.Controls.DataGridRow> GetDataGridRows(Microsoft.Windows.Controls.DataGrid grid)
{
var itemsSource = grid.ItemsSource as IEnumerable;
if (null == itemsSource) yield return null;
foreach (var item in itemsSource)
{
var row = grid.ItemContainerGenerator.ContainerFromItem(item) as Microsoft.Windows.Controls.DataGridRow;
if (null != row) yield return row;
}
}
and then iterate through your grid:
var rows = GetDataGridRows(nameofyordatagrid);
foreach (DataGridRow row in rows)
{
DataRowView rowView = (DataRowView)row.Item;
foreach (DataGridColumn column in nameofyordatagrid.Columns)
{
if (column.GetCellContent(row) is TextBlock)
{
TextBlock cellContent = column.GetCellContent(row) as TextBlock;
MessageBox.Show(cellContent.Text);
}
}
Yes, you are right. WPF DataGrid is built around better supporting the use of objects.
You could use a ViewModel similar to the following. Build them all into a collection and then set that collection as your ItemsSource. You would also need to use a ValueConverter if you want to display and image instead of a checkmark for pNew being true/false.
public class FooViewModel : INotifyPropertyChanged
{
private int currentPareto;
public int CurrentPareto
{
get
{
return currentPareto;
}
set
{
if (currentPareto == value)
return;
currentPareto = value;
OnPropertyChanged("CurrentPareto");
OnPropertyChanged("pNew");
}
}
private int newPareto;
public int NewPareto
{
get
{
return newPareto;
}
set
{
if (newPareto == value)
return;
newPareto = value;
OnPropertyChanged("NewPareto");
OnPropertyChanged("pNew");
}
}
public bool pNew
{
get
{
return CurrentPareto < NewPareto;
}
}
public event PropertyChangedEventHandler PropertyChanged;
public void OnPropertyChanged(string propertyName)
{
if (PropertyChanged != null)
{
PropertyChanged(this, new PropertyChangedEventArgs(propertyName));
}
}
}
Edit
To simplify it a little, you could use a base ViewModel class and use PropertyChanged weaving. The code would simplify to this:
public class FooViewModel : ViewModelBase
{
public int CurrentPareto { get; set; }
public int NewPareto { get; set; }
public bool pNew { get { return CurrentPareto < NewPareto; } }
}
I don't even understand why is it just so complicated to get rows and their values in a datagrid. It feels like hell finding how. The api even give funny funny event names which is not so direct to the point also. Why can't just people concentrate on the baseline and give what exactly is needed and not all sorts of different options with no use and sense at all. I mean to eat all you need is a spoon and fork right. Never even changed since 100,000 years ago. This is my code thanks to the guy who mentioned some people just try to over-complicate things and waste your time.
private void dtaResultGrid_DataContextChanged(object sender, DependencyPropertyChangedEventArgs e)
{
ActivateTestDatagridAccess();
}
public async void ActivateTestDatagridAccess()
{
try
{
await Task.Delay(500);
foreach (System.Data.DataRowView dr in dtaResultGrid.ItemsSource)
{
for (int j = 0; j < dtaResultGrid.Columns.Count; j++)
{
Console.WriteLine(dr[j].ToString());
}
Console.Write(Environment.NewLine);
}
}
catch (Exception exrr)
{
Console.WriteLine(exrr.ToString());
}
}
The 'simplest' answer, from Charles, did it for me. But I used Items instead of ItemsSource.
Now, for people getting this error:
System.InvalidCastException
Unable to cast object of type 'MS.Internal.NamedObject' to type 'System.Data.DataRowView'.
What did it for me was disabling the DataGrid's property CanUserAddRows. This removes the placeholder line for a new line, thus removing the placeholder object (which is NOT a DataRowView, but something else). If you already have this disabled, then I don't know.
Since I wanted to loop through each element of each row, I added another foreach:
foreach (System.Data.DataRowView dr in nameofyourgrid.Items)
{
foreach (var item in dr.Row.ItemArray)
{
MessageBox.Show(item.ToString());
}
}
In WPF you go about it a lot more dynamic and ObjectOrientated. You can bind the Column "pNew" on a Property of the element you put in the DataGrid, which returns downarrow.
If the value changes you can raise the Event PropertyChanged (Interface INotifyPropertyChanged) and the bound Property will get reevaluated.
Also interesting for beginning with WPF is DataTemplate, ControlTemplate, Converter.
Converter changes the Property Value to a usable Value for WPF (e.g. BoolToVisibility) when the Property gets called.
DataTemplate and ControlTemplate can be used to alter the appearance of the Control.
There are several good Tutorials for WPF out there. I would also recommend to look into the MVVM-Pattern to use as a between layer of your Businessobject and your WPF-Control, especially to handle things like what you try to do here.
if you fill your datagridview rows using an instance of a class (like struct_class)
this would be the fastest way to have a foreach loop
foreach (struct_class row in dgv.Items)
{
MessageBox.Show(row.name);
}
Why can't you just use this property to get the number of rows and then use a For loop to iterate through?
dataGridView1.Rows.Count

Checking a Checkbox after checking another one (TreeView)

So i have this TreeView with 3 parent nodes, each one with some childs. There's a specific child that depends on others two childs in the same parent node. Here's the code i've done:
private void tvMorgan_AfterCheck(object sender, TreeViewEventArgs e)
{
if ((e.Node.Text.Contains("BRL/EUR")) && (e.Node.Parent.Text.Contains("FWD")))
{
TreeNode tnParent = e.Node.Parent;
tnParent.Nodes["BRL/USD"].Checked = true;
tnParent.Nodes["EUR/USD"].Checked = true;
}
}
When i run it, it can't find those two nodes i want to check.
Thanks in advance
As commentors pointed out, your nodes are most likely missing a name, which is the key used in the string index lookup. Easy to happen since most nodes get created at runtime.
Simple way to add the name:
myParent.Nodes.Add(new TreeNode("BRL/USD") { Name = "BRL/USD" });
myParent.Nodes.Add(new TreeNode("EUR/USD") { Name = "EUR/USD" });
Then your code should work:
if (tnParent.Nodes.ContainsKey("EUR/USD"))
tnParent.Nodes["EUR/USD"].Checked = true;
You should probably use the same logic on your if condition so that you are less reliant on the text property, which really should be only used for display purposes:
if ((e.Node.Name == "BRL/EUR") ...
If not setting the names, then you would have to search the node tree yourself to find the "text":
TreeNode found = tnParent.Nodes.OfType<TreeNode>().Where(x => x.Text.Contains("EUR/USD")).First();
if (found != null)
found.Checked = true;

Correct way to implement web part personalisation for listboxes

Trying to work out this whole web part personalisation, and trying to implement it for a list box.
Well the end result will be two list boxes, with interchangeable values (ie, a value will only exist in one of the listboxes)
But I can't maintain the datasource for it. So maybe I'm going about it wrong?
This is what I have for a test H2 tag on the page
[Personalizable(PersonalizationScope.User)]
public string LabelText {
get { return h2Test.InnerText; }
set { h2Test.InnerText = value; }
}
And it works fine, if I have a textbox and use it to change the value of LabelText, then when I close the browser it automagically persists the change.
So I thought, ok, then maybe the same will work with a list box
[Personalizable(PersonalizationScope.User)]
public DomainList Domains {
get { return (DomainList)lstBxDomains.DataSource; }
set {
lstBxDomains.DataSource = value;
lstBxDomains.DataBind();
}
}
Where DomainList is just a class which extends List, and Domain is just a three field class, int, string, string.
But it doesn't, so is this too complicated for the webpart personalisation automagican, or have i just implement it wrongly (Which is more than likely)
This is my event handler to remove the items from the list:
protected void btnRemDomain_Click(object sender, EventArgs e) {
if (IsPostBack && lstBxDomains.SelectedIndex > -1) {
for (int i = 0; i < lstBxDomains.Items.Count; i++) {
if (lstBxDomains.Items[i].Selected) {
Domains.Remove(Domains.Find(d => d.ID.ToString() == lstBxDomains.Items[i].Value));
}
}
Domains = Domains;
}
}
The Domains=Domains; line is in there to see if explicitly setting the value made a difference (as Removing doesn't acutally reset the value of the field), but it doesn't. I've also tried creating a new local DomainList setting it to the global one, and then doing the remove/find on it, and then setting the local one to the global. But not working either.
I have managed to resolve this by using WebPart.SetPersonalizationDirty(this); in the set accessor of Domains, but would someone mind confirming if this is an appropriate way to do it?

TreeView child node populating problem

I need to construct a huge treeview from a composite database table with Grouping.
Grouping is, what we see in SQL Server Management Studio Express. After a Database node, some fixed folders are shown (like, Database Diagrams, Tables, Views, Synonyms, Programmability and Security) and children are grouped in those folders.
Up to this point I have used AfterSelect event and handler to achieve this.
But the problem with AfterSelect is, before selecting the node, the viewer is not able to know whether there is any child available. This is because, the expandable plus sign is not visible.
I want to use BeforeExpand. But the problem with BeforeExpand is, it works if the children are already populated. In that case, when I click groups, nothing happens.
How to solve this?
So codes/web-link will be appreciated.
What I usually do is to add a "dummy child node" wherever there may be children that should be loaded in a lazy manner. This will make the parent have the plus sign, and then you can add code to the AfterExpand event where you do the following:
Check if there are are exactly one child, and if that child is the dummy node (you can use the Tag property to identify the dummy node)
If the dummy node is found, launch a search to get the children and add them to the parent node, finish it off by removing the dummy node.
I typically give the dummy node a text like "Loading data. Please wait..." or so, so that the user gets some info on what is going on.
Update
I put together a simple example:
public class TreeViewSample : Form
{
private TreeView _treeView;
public TreeViewSample()
{
this._treeView = new System.Windows.Forms.TreeView();
this._treeView.Location = new System.Drawing.Point(12, 12);
this._treeView.Size = new System.Drawing.Size(200, 400);
this._treeView.AfterExpand +=
new TreeViewEventHandler(TreeView_AfterExpand);
this.ClientSize = new System.Drawing.Size(224, 424);
this.Controls.Add(this._treeView);
this.Text = "TreeView Lazy Load Sample";
InitializeTreeView();
}
void TreeView_AfterExpand(object sender, TreeViewEventArgs e)
{
if (e.Node.Nodes.Count == 1 && e.Node.Nodes[0].Tag == "dummy")
{
// this node has not yet been populated, launch a thread
// to get the data
ThreadPool.QueueUserWorkItem(state =>
{
IEnumerable<SomeClass> childItems = GetData();
// load the data into the tree view (on the UI thread)
_treeView.BeginInvoke((Action)delegate
{
PopulateChildren(e.Node, childItems);
});
});
}
}
private void PopulateChildren(TreeNode parent, IEnumerable<SomeClass> childItems)
{
TreeNode child;
TreeNode dummy;
TreeNode originalDummyItem = parent.Nodes[0];
foreach (var item in childItems)
{
child = new TreeNode(item.Text);
dummy = new TreeNode("Loading. Please wait...");
dummy.Tag = "dummy";
child.Nodes.Add(dummy);
parent.Nodes.Add(child);
}
originalDummyItem.Remove();
}
private IEnumerable<SomeClass> GetData()
{
// simulate that this takes some time
Thread.Sleep(500);
return new List<SomeClass>
{
new SomeClass{Text = "One"},
new SomeClass{Text = "Two"},
new SomeClass{Text = "Three"}
};
}
private void InitializeTreeView()
{
TreeNode rootNode = new TreeNode("Root");
TreeNode dummyNode = new TreeNode("Loading. Please wait...");
dummyNode.Tag = "dummy";
rootNode.Nodes.Add(dummyNode);
_treeView.Nodes.Add(rootNode);
}
}
public class SomeClass
{
public string Text { get; set; }
}
It's standard behaviour for a tree to show a "+" in front of every folder/group, and the plus dissapears when clicked on if it's found to have no children, this saves the expensive "do you have children" check.
Alternatively you can provide this information if you have a cheap way of determining if a node has children. This question provides more information.

Categories

Resources