• 0

C# and making a TreeView out of arrays


Question

So I have encountered a probleem while trying to populate a TreeView control with nodes. I hope that there are a some kind people who t?ke the time and help me figuure out a correct solution.

So lets say I have 3 arrays:

 

["Example", "2", "1"]

{"Example", "2", "2"]

["Example", "3", "0"]

 

I know that my arrays have always exactly 3 elements in them, so the treeview will never have more than 3 childnodes

 

But all I am able to get is

 

Example

 2

  1

Example

 2

  2

Example

 3

  0

 

But the goal would be:

 

Example

 2

  1

  2

 3

  0

 

So where would I start chasing down the answer to my probleem?
Big thanks in advance :)

 

 

Recommended Posts

  • 0


void InitTreeView(List<string[]> arrays) {

            foreach (var array in arrays) {

                AddArray(array, 0, m_treeView.Nodes);

            }

        }

        /// <summary>

        /// Recursively adds each element of the array to the tree

        /// </summary>

        void AddArray(string[] array, int index, TreeNodeCollection nodes) {

            // Termination condition: we've past the bounds of the array

            if (index < array.Length) {

                var nextNode = AddValue(array[index], nodes);

                AddArray(array, index + 1, nextNode.Nodes);

            }

        }

        /// <summary>

        /// If the value exists, returns it, otherwise creates a new value

        /// </summary>

        TreeNode AddValue(string value, TreeNodeCollection nodes) {

            var index = nodes.IndexOfKey(value);

            if (index == -1) {

                var newNode = new TreeNode(value) { Name = value };

                nodes.Add(newNode);

                return newNode;

            }

            return nodes[index];

        }

  • 0

        private static TreeView GetTree(TreeView Tree, List<string[]> Items)
        {
            foreach (string[] item in Items)
            {
                TreeNode parent = new TreeNode();
                parent.Text = item[0];
                TreeNode child = new TreeNode();
                child.Text = item[1];
                TreeNode child2 = new TreeNode();
                child2.Text = item[2];
                child.Nodes.Add(child2);
                parent.Nodes.Add(child);
                Tree.Nodes.Add(parent);
            }
            return Tree;
        }

Well, I have a code  that goes through each array and creates parentnode with array first item, childnode with second and childnode of childnode with third. So thats pretty basic approach

 

  • 0

I imagine you want to do something like the following pseudo code.
Idea is that you create the array of children you want before you create the treenode with the label.

tv = new TreeView();
dict = new Dictionary<String, List<String>>();
arrs = array of those arrays;

for arr in arrs:
     dict[arr[1]].Add(arr[2]);

for (key, value) in dict:
    tv.Nodes.Add(new TreeNode(key, value.ToArray()))
  • 0
  On 22/12/2013 at 14:17, Lant said:

 

I imagine you want to do something like the following pseudo code.

Idea is that you create the array of children you want before you create the treenode with the label.

 

 

Got lost trying to implement something like that, although the approach is right, I need to create the last level nodes first

  • 0

What sort of performance are you looking for?  If you're just talking a few hundred items, then try this approach.

 

Each node should be identifiable via a predictable, unique key of your creation (it could be, literally, the value you're trying to insert).  Try to find your parent node by looking for the key it should have.  If you can't find it, create a new node and insert it.  If it returns something instead, then use that as the parent of the key you're trying to insert.

  • 0
  On 22/12/2013 at 16:55, jakem1 said:

Try a recursive function.

This is the right way to go, I know that, but I am totally unsure how the code should look like.

Regarding to performance, it is important in my case since we're talking about thousands of nodes here.

  • 0
  On 22/12/2013 at 17:01, ka-la said:

This is the right way to go, I know that, but I am totally unsure how the code should look like.

Regarding to performance, it is important in my case since we're talking about thousands of nodes here.

 

As a general rule of thumb, if you find yourself pushing program state into some sort of explicit stack when using an iterative solution (pushing state to a queue for example): you are may just be doing an explicit recursion (what I mean by this is that you are doing the same work that a recursion would do for you) and are may be better suited to just use a recursive solution.

  • 0

This should be very straightforward to convert to C#, but here's the general idea for a recursive solution:

InitTreeView:
    var data = new List<string[]> { " Your data here " };
    foreach array in data:
        AddArray(array, 0, root collection)

AddArray(array, index, nodes):
    // Termination condition: index is past the end of the array
    // Otherwise:
    nextNode = AddValue(array[index], nodes)
    AddArray(array, index + 1, nextNode collection)

AddValue(value, nodes):
    // If node exists in the collection, return it
    // Otherwise create new node with the value, add it and return it
  • 0
  On 26/12/2013 at 17:08, Andre S. said:

 

This should be very straightforward to convert to C#, but here's the general idea for a recursive solution:


InitTreeView:
    var data = new List<string[]> { " Your data here " };
    foreach array in data:
        AddArray(array, 0, root collection)

AddArray(array, index, nodes):
    // Termination condition: index is past the end of the array
    // Otherwise:
    nextNode = AddValue(array[index], nodes)
    AddArray(array[index], index + 1, nextNode collection)

AddValue(value, nodes):
    // If node exists in the collection, return it
    // Otherwise create new node with the value, add it and return it

Some parts of the code still remain as a puzzle for me:

1)  What should AddValue return, a collection or a treenode?

2) AddArray( array[index].... seems weird to me, it expects an array, but your code sends it a value from array with that index

3) What should AddArray return? my best guess so far is, it is just a void and does not return anything.

4) nextNode collection should be nextNode.Nodes ? (case the AddValue returns a treenode)

  • 0
  On 27/12/2013 at 15:08, ka-la said:

Some parts of the code still remain as a puzzle for me:

1)  What should AddValue return, a collection or a treenode?

2) AddArray( array[index].... seems weird to me, it expects an array, but your code sends it a value from array with that index

3) What should AddArray return? my best guess so far is, it is just a void and does not return anything.

4) nextNode collection should be nextNode.Nodes ? (case the AddValue returns a treenode)

1) As I wrote it, it returns a TreeNode. You could just return the collection as well as that's the only thing AddArray needs.

2) Oops that was a mistake, fixed. Should pass array.

3) AddArray is void.

4) Yup.

 

Is this homework?

  • 0
        private static TreeView GetTree(TreeView Tree)
        {
            Tree.BeginUpdate();
            List<string[]> l_tree = new List<string[]>();
            string[] str1 = new string[] {"Example", "2", "1"};
            string[] str2 = new string[] { "Example", "2", "2"};
            string[] str3 = new string[] { "Example", "3", "0" };
            l_tree.Add(str1);
            l_tree.Add(str2);
            l_tree.Add(str3);

            foreach (string[] array in l_tree)
            {
                AddArray(array, 0, Tree.Nodes);
            }
            Tree.EndUpdate();
            return Tree;
        }

        private static void AddArray(string[] array, int index, TreeNodeCollection nodes)
        {
            if (index < 3)
            {
                var nextNode = AddValue(array[index], nodes);
                AddArray(array, index + 1, nextNode.Nodes);
            }
        }

        private static TreeNode AddValue(string value, TreeNodeCollection nodes)
        {
            if (nodes.Count == 0)
            {
                TreeNode newnode = new TreeNode();
                newnode.Text = value;
                nodes.Add(newnode);
                return newnode;
            }
            else
            {
                foreach (TreeNode node in nodes)
                {
                    if (node.Text == value)
                    {
                        return node;
                    }
                    else
                    {
                        TreeNode newnode = new TreeNode();
                        newnode.Text = value;
                        node.Nodes.Add(newnode);
                        return node;
                    }
                }
            }
            return null;
        }

If the initial tree is empty, I get nullreferenceexception, since I cant go through the empty collection in addvalue

Example

 2

  1

    2

    0

  3

 

 

  • 0
  On 27/12/2013 at 22:59, ka-la said:

Sir I am truly thankful for your help, not sure how to return the regards :)

Glad to be of help. :) By the way, you can write this:

    List<string[]> l_tree = new List<string[]>();
    string[] str1 = new string[] {"Example", "2", "1"};
    string[] str2 = new string[] { "Example", "2", "2"};
    string[] str3 = new string[] { "Example", "3", "0" };
    l_tree.Add(str1);
    l_tree.Add(str2);
    l_tree.Add(str3);

more simply as:

    var l_tree = new List<string[]> {
        new[] { "Example", "2", "1" },
        new[] { "Example", "2", "2" },
        new[] { "Example", "3", "0" }
    }
  • 0

I haven't tested but are you required to use the new[] syntax for the string arrays(i.e. string[]) in this example or can you get away with removing that part because of the string type?

  • 0
  On 28/12/2013 at 02:56, snaphat (Myles Landwehr) said:

I haven't tested but are you required to use the new[] syntax for the string arrays(i.e. string[]) in this example or can you get away with removing that part because of the string type?

new[] { "a", "b" }

 is shorthand (since C# 3.0) for :

new string[] { "a", "b" }

There's also a special case (which was just to copy C/C++ in C# 1.0) for assignment to a new variable, where you can omit everything:

string[] array = { "a", "b" }

But then you have to explicitely type the variable, you can't use var. IMO this is old fashioned and for consistency everyone should write:

var array = new[] { "a", "b" }

Of course this works for all types and not just string.

  • 0
  On 28/12/2013 at 03:12, Andre S. said:
new[] { "a", "b" }

 is shorthand (since C# 3.0) for :

new string[] { "a", "b" }

There's also a special case (which was just to copy C/C++ in C# 1.0) for assignment to a new variable, where you can omit everything:

string[] array = { "a", "b" }

But then you have to explicitely type the variable, you can't use var. IMO this is old fashioned and for consistency everyone should write:

var array = new[] { "a", "b" }

Of course this works for all types and not just string.

 

 

Right, right, Is the special case for assignment still possible in newer C# revisions or is is not-recommended/depreciated? I initially thought that you were being a bit more verbose with the new[] statement in the compound list assignment because it was required. Sounds like you were doing it more for consistencies sake; which is fair.

  • 0
  On 28/12/2013 at 03:28, snaphat (Myles Landwehr) said:

Right, right, Is the special case for assignment still possible in newer C# revisions or is is not-recommended/depreciated? I initially thought that you were being a bit more verbose with the new[] statement in the compound list assignment because it was required. Sounds like you were doing it more for consistencies sake; which is fair.

I'm doing it in the list assignment because it is and has always been required in that case. You can omit new[] only if you assign the array expression to an explicitely typed variable. This, in addition to the fact that this syntax contradicts the usual C# principle that every expression has a type, makes it inconsistent and confusing in my eyes, but it's still supported. It's only my opinion and not any sort of official deprecation or recommendation.

  • Like 1
  • 0
  On 28/12/2013 at 03:56, Andre S. said:

I'm doing it in the list assignment because it is and has always been required in that case. You can omit new[] only if you assign the array expression to an explicitely typed variable. This, in addition to the fact that this syntax contradicts the usual C# principle that every expression has a type, makes it inconsistent and confusing in my eyes, but it's still supported. It's only my opinion and not any sort of official deprecation or recommendation.

 

Fair enough, I'll say that the bolded text is a good reason to use it in the above code regardless of opinion ;-)

  • 0
  On 28/12/2013 at 04:09, snaphat (Myles Landwehr) said:

Fair enough, I'll say that the bolded text is a good reason to use it in the above code regardless of opinion ;-)

I'm not sure if we're talking past each other... this:

new List<string[]> {
     { "a", "b", "c" },
     { "b", "c", "d" }
}

is not and never was legal C#.

This topic is now closed to further replies.
  • Posts

    • Or, it is Apple simply overinflated the screens.
    • Is it that easy though? WhatsApp is the default way the majority message in a lot of countries these days. I would prefer Signal to be as popular as WhatsApp and probably could get a few people to use it, most people are probably going to stick with WhatsApp sadly. Which leaves SMS or Facebook Messenger as alternatives a lot of people also have. (Here anyway, I know iMessage, LINE and WeChat dominate in certain parts of the world). It annoying Meta purchased WhatsApp all those years ago.
    • Do they tell Google not to scrape their content via something like robots.txt? Do they specify anywhere that certain agents aren't to scrape? If not, tough. Plus there's no obligation on anyone's part to adhere to any directives that might be in this file anyway...
    • DMDE 4.3.5.823 Beta by Razvan Serea DMDE is a software designed to effectively recover lost data. It retrieves files and folders swiftly and stores them in the user-defined location. It is an easy to use yet powerful tool that will assist both novice and experienced users in getting back lost files in just a few simple steps. Free Edition includes all basic features but a single recovery operation recovers up to 4000 files in the current panel only (you should first open a subdirectory in the current panel and then recover files in the panel). In paid licenses there is no this restriction, and recovery of nested directories is allowed. Can paid versions recover more files than the free version of DMDE? If a file cannot be recovered in the DMDE Free Edition (or it is damaged after recovery) the same will occur in the paid versions. DMDE paid versions are capable of recovering the same files. The only difference is that paid versions can recover all found files in one go, as well as restore the directory structure presented in the free version. Professional Edition provides additional features: rights to provide data recovery services portable use on different computers one-time activation on client computers (including remote use) data recovery reports (include logs and file checksums) read support for E01 disk image files using logs when copying a disk (resume copying, multiple passes) customizable I/O handler script recovery of NTFS alternate data streams DMA access in DOS (for ATA interface) DMDE key features: Portable run without installation Support for NTFS, FAT12/16, FAT32, exFAT, ReFS, Ext2/Ext3/Ext4, btrfs, HFS+/HFSX, APFS Thorough FS and Raw scan, FS reconstruction for data recovery in complex cases Simple partition manager for express search, diagnostics, and restoration of partitions Disk cloning and disk image creating, including I/O error handling, reverse copying, and other features RAID constructor for virtual RAID reconstruction supporting levels RAID-0, RAID-1, RAID-4, RAID-5, RAID-6, delayed parity, custom striping, JBOD/spanned disks; automatic calculation of RAID configurations Cluster map to investigate file allocation Disk editor compatible with the most recent Windows versions which allows viewing, editing, and navigating through different disk structures using built-in and custom templates NTFS tools to work bypassing NTFS driver (copy, delete file, create, repair directory) Support for various device I/O interfaces and settings to work with damaged devices, disk images, NTFS compression and encryption, national names, large disks, large files, large sectors, and other features DMDE 4.3.5.823 Beta changelog: Expanded built-in signatures for RAW search functionality Added file list export to HTML format (DMDE Professional Edition only, view sample) Improved handling of I/O errors with selective skipping by error code Enabled preview support for additional image (graphic) file types (Windows only) Improved extfs reconstruction when copies of superblocks with group descriptors are found Fixed potential hang during Btrfs volume reconstruction Resolved issue with cluster list creation when subfolders are present Other improvements and fixes Download: DMDE 64-bit | 2.4 MB (Free, paid upgrade available) Download: DMDE 32-bit | 2.0 MB Link: DMDE Home Page | DMDE Manual | Screenshot Get alerted to all of our Software updates on Twitter at @NeowinSoftware
    • The BBC might have gone about this the wrong way, but if there is a revenue sharing program then they and all other "providers" of data should be included in the plan.
  • Recent Achievements

    • Week One Done
      Crunchy6 earned a badge
      Week One Done
    • One Month Later
      KynanSEIT earned a badge
      One Month Later
    • One Month Later
      gowtham07 earned a badge
      One Month Later
    • Collaborator
      lethalman went up a rank
      Collaborator
    • Week One Done
      Wayne Robinson earned a badge
      Week One Done
  • Popular Contributors

    1. 1
      +primortal
      680
    2. 2
      ATLien_0
      276
    3. 3
      Michael Scrip
      221
    4. 4
      +FloatingFatMan
      170
    5. 5
      Steven P.
      164
  • Tell a friend

    Love Neowin? Tell a friend!