We Keep Coding

Create Intersect of HashSet and remove Items from class structure with extracted ID's

lets assume I have the following classes:
public class ServiceStatistics
{
public string LocalId { get; set; }
public string OrganizationId { get; set; }
public List<StatisticElements> Elements { get; } = new List<StatisticElements>();
}
public class StatisticElements
{
public string StatisticId { get; set; }
public string Type { get; set; }
public string ServiceName { get; set; }
}
I retrieve such ServiceStatistics by a soap service and I use serialization/deserialization.
Each ServiceStatistics contains a set of StatisticElements. I also have a static list of StatisticElements-ID's which are relevant for calculation. All other incoming StatisticElements-ID's can be dropped. I need to do this on my side
because the SOAP Service does not support selecting specific StatisticElements-ID's
So I have generated a static Class with a HashSet:
public static class RelevantDutyPlans
{
private static HashSet<int> relevantDutyPlans;
static RelevantDutyPlans()
{
// only a subset of the original ID's
relevantDutyPlans = new HashSet<int>()
{
530,
1150,
1095,
};
}
public static HashSet<int> GetRelevantDutyPlans()
{
return relevantDutyPlans;
}
public static bool Contains(int planId)
{
return relevantDutyPlans.Contains(planId);
}
// Extracts all DutyPlans which are relevant (HashSet) for validation from
// the incoming data
public static List<int> ExtractRelevantDutyPlans(List<int> planIds)
{
var relevantPlans = new HashSet<int>(planIds);
relevantPlans.IntersectWith(relevantDutyPlans);
return relevantDutyPlans.ToList();
}
}
So my thought was, to create an Intersect like this:
List<ServiceStatistics> statistics = SoapService.GetStatistics(Now);
List<int> incomingIds = new List<int>();
foreach(var item in statistics)
{
foreach(var element in item.Statistic)
{
incomingIds.Add(int.Parse(element.StatisticId));
}
}
List<int> extract = RelevantDutyPlans.ExtractRelevantDutyPlans(incomingIds);
So now I have a List of ID's which are relevant for further processing. What I want to achieve is to remove all class elements "StatisticElements" with "StatisticId" not contained in the the extract list generated above.
Any ideas?
Any help is very appreciated

How about a little bit different approach. Simply remove irrelevant plans right away!
List<ServiceStatistics> statistics = SoapService.GetStatistics(Now);
foreach(var item in statistics)
{
item.Elements.RemoveAll(x => !RelevantDutyPlans.Contains(int.Parse(x.StatisticId)));
}
Now you are only left with the relevant once.

Hope you can use selectMany to flatten the collection and proceed the filter.
var filteredItems = statistics.SelectMany(s => s.Elements)
.Where(s => extract.Contains(Convert.ToInt32(s.StatisticId)))
.ToList();

You could also use LINQ to create a new List<> if you need to keep the original statistcs intact - e.g. if you might run multiple plans against it.
var relevantStatistics = statistics.Select(s => new {
LocalId = s.LocalId,
OrganizationId = s.OrganizationId,
Elements = s.Elements.Where(e => !RelevantDutyPlans.Contains(Convert.ToInt32(e.StatisticId))).ToList()
});
Since ServiceStatistics doesn't provide for construction, I return an anonymous object instead, but you could create an appropriate DTO class.

check for removed slots in list

I would like to generate some Items with IDs. Item is just a class with an ID integer attribute. They should get added to a list after creating them. So I have a class that manages all the items
internal static class ItemPool
{
public static readonly List<Item> items = new List<Item>(); // Store all the items here
public static Item CreateItem()
{
int itemId = items.Count; // Just take the next free slot index
Item itemToCreate = new Item(itemId);
items.Add(itemToCreate);
return itemToCreate;
}
public static void DestroyItem(int itemId)
{
activeItems.RemoveAt(itemId);
}
}
Now I can create some items
Item firstItem = ItemPool.CreateItem(); // generates item with ID 0
Item secondItem = ItemPool.CreateItem(); // generates item with ID 1
ItemPool.DestroyItem(firstItem.id); // Recudes the list count to 1
Item thirdItem = ItemPool.CreateItem(); // generates item with ID 1 - ERROR
The third item is not allowed to have the ID 1 because item 2 already has it.
When changing the code two questions come up:
How to manage the IDs for items in this list so that none of them have the same ID?
public static Item CreateItem()
{
int itemId = temporaryUniqueId; // create a temporary unique ID
// ... other code
}
What is a more optimized way than going for
public static void DestroyItem(int itemId)
{
activeItems = activeItems.Where(item => item.id != itemId).ToList();
}
I know I could do
public static void DestroyItem(Item item)
{
activeItems.Remove(item);
}
but I think removing by ID is more safe. I am asking this because for this case performance is everything.
The IDs don't have to be integer values

Since having an ID as an integer is not a requirement, one way to do this would be to use a GUID as a unique identifier to avoid having to deal with potential conflict.
public class Item
{
public Guid Id { get; }
public Item()
{
Id = Guid.NewGuid();
}
}
Then, your ItemPool class could be changed to use a ConcurrentDictionary (to avoid any race conditions):
internal sealed class ItemPool
{
private readonly ConcurrentDictionary<Guid, Item> items = new ConcurrentDictionary<Guid, Item>(); // Store all the items here
public Item CreateItem()
{
Item itemToCreate = new Item();
items.TryAdd(itemToCreate.Id, itemToCreate);
return itemToCreate;
}
public void DestroyItem(Guid itemId)
{
activeItems.TryRemove(itemId, out Item _);
}
}
I took the liberty of removing the static parameter of the class to make it easier to test. I also made the items field private for better encapsulation. You do not want anyone other class to bypass ItemPool and start manipulating the collection by themselves :)

Serialize list of objects

I need to serialize/deserialize some XML code and part of it looks like next example:
<CoordGeom>
<Curve rot="cw" chord="830.754618036885" crvType="arc" delta="72.796763873948" dirEnd="283.177582669379" dirStart="355.974346543327" external="169.661846548051" length="889.38025007632" midOrd="136.562611151675" radius="699.999999998612" tangent="516.053996536113">
<Start>4897794.2800513292 6491234.9390137056</Start>
<Center>4897096.0071489429 6491185.7968343571</Center>
<End>4897255.5861026254 6491867.3645547926</End>
<PI>4897758.0514541129 6491749.7197593488</PI>
</Curve>
<Spiral length="109.418078418008" radiusEnd="INF" radiusStart="699.999999999025" rot="cw" spiType="clothoid" theta="4.477995782709" totalY="2.849307921907" totalX="109.351261203955" tanLong="72.968738862921" tanShort="36.493923980983">
<Start>4897255.5861026254 6491867.3645547936</Start>
<PI>4897220.0531303799 6491875.6840722272</PI>
<End>4897147.9238984985 6491886.7208634559</End>
</Spiral>
<Spiral length="153.185309785019" radiusEnd="499.99999999993" radiusStart="INF" rot="ccw" spiType="clothoid" theta="8.776871734087" totalY="7.808812331497" totalX="152.826239431476" tanLong="102.249348442205" tanShort="51.176160975293">
<Start>4897147.9238985004 6491886.7208634559</Start>
<PI>4897046.8509311257 6491902.186455016</PI>
<End>4896998.0370401107 6491917.5553683294</End>
</Spiral>
<Curve rot="ccw" chord="936.510896488672" crvType="arc" delta="138.94725576785" dirEnd="66.423714388543" dirStart="287.476458620693" external="925.970149937768" length="1212.543549877849" midOrd="324.680762068264" radius="499.999999999181" tangent="1335.436583485725">
<Start>4896998.0370401107 6491917.5553683294</Start>
<Center>4897148.1939981515 6492394.4755796343</Center>
<End>4896948.2091376046 6492852.7397562303</End>
<PI>4895724.243644949 6492318.6055583945</PI>
</Curve>
</CoordGeom>
I've generated automatically classes using xsd.exe. Part of generated code looks like this:
public partial class CoordGeom
{
private List<object> _items;
private List<Feature> _feature;
private string _desc;
private string _name;
private stateType _state;
private string _oID;
public CoordGeom()
{
_feature = new List<Feature>();
_items = new List<object>();
}
[XmlElementAttribute("Chain", typeof(Chain))]
[XmlElementAttribute("Curve", typeof(Curve))]
[XmlElementAttribute("IrregularLine", typeof(IrregularLine))]
[XmlElementAttribute("Line", typeof(Line))]
[XmlElementAttribute("Spiral", typeof(Spiral))]
public List<object> Items
{
get { return this._items; }
set { this._items = value; }
}
[XmlElement("Feature")]
public List<Feature> Feature { get; set; }
[XmlAttribute()]
public string desc { get; set; }
[XmlAttribute()]
public string name { get; set; }
[XmlAttribute()]
public stateType state { get; set; }
[System.Xml.Serialization.XmlAttributeAttribute()]
public string oID
{
get{ return this._oID; }
set{ this._oID = value; }
}
}
And my code for deserialization look like this:
XmlSerializer mySerializer = new XmlSerializer(typeof(LandXML), new XmlRootAttribute(""));
TextReader myFileStream = new StreamReader("myFile.xml");
LandXML myObject = (LandXML)mySerializer.Deserialize(myFileStream);
var coordGeomItems = myObject.Alignments.Alignment[0].CoordGeom;
My problem is that, when I deserialize file, it is deserialized as list of items of type {LandXML.Curve}, {LandXML.Spiral} etc. and I don't know how to access their properties. It would be great if I can do this directly. Here is a screenshot:
EDIT 1
Here is inital screen
then I have items:
When I unfold this
And this is at the top layer of object - it has some InnerXml, InnerText... If I want to achieve CoordGeom, there is a lot object.Item(i).ChildNodes.Item(j).ChildNodes...
And all of that is because in some lines, lists of objects are made like List as for CoordGeom

Because there are multiple allowed types, the Items collection is typed as object. The simplest approach is to enumerate and cast each item:
foreach(var item in coordGeomItems.Items)
{
var curve = item as Curve;
if (curve != null)
{
// access curve properties here
}
var spiral = item as Spiral
if (spiral != null)
{
// access spiral properties here
}
// ...
}

You could build up a list of Curves and Spirals and access them using properties with custom getters:
class CoordGeom
{
public List<object> Items;
List<Curve> _curves;
public List<Curve> Curves
{
get
{
return _curves ?? (_curves = Items
.Where(item => item is Curve).Select(curve => (Curve)curve).ToList());
}
}
}
The null coalescing operator (??) will cause the Curves property to set and return the value of _curves as a list of curves if _curves is null. This basically causes it to initialize the list on the first get and on all subsequent gets it will return the already initialized list.

As you cannot change the generated class nor the XML.The best possible approach would be to write an extension method.
public static List<Curve> GetCurves(this CoordGeom cg)
{
return cg.Items.OfType<Curve>().ToList();
}
public static List<Spiral> GetSpirals(this CoordGeom cg)
{
return cg.Items.OfType<Spiral>().ToList();
}
Once you do this, you can get items like this
var coordGeomItems = myObject.Alignments.Alignment[0].CoordGeom;
var curves = coordGeomItems.GetCurves();
var spirals = coordGeomItems.GetSpirals();

Why isn't my object being updated

I'm doing something wrong because after the loop executed myData still contains objects with blank ids. Why isn't the myData object being updated in the following foreach loop, and how do I fix it?
I thought it could be that I wasn't passing the object by reference, but added a ref keyword and also moved to the main method and I'm still showing the object not being updated.
Additional Information
The user object in the foreach loop is being updated, but the myData list does not reflect the updates I see being applied to the user object.
** Solution **
I was not creating a List but an Enumerable which was pulling the json each time I went through myData in a foreach list. Adding a ToList() fixed my issue.
public class MyData
{
public string ID { get; set; }
public Dictionary<string, string> Properties { get; set; }
}
int index = 0;
// Does not allow me to up, creates an IEnumerable
//IEnumerable<MyData> myData = JObject.Parse(json)["Users"]
// .Select(x => new MyData()
// {
// ID = x["id"].ToString(),
// Properties = x.OfType<JProperty>()
// .ToDictionary(y => y.Name, y => y.Value.ToString())
// });
//Works allows me to update the resulting list.
IEnumerable<MyData> myData = JObject.Parse(json)["Users"]
.Select(x => new MyData()
{
ID = x["id"].ToString(),
Properties = x.OfType<JProperty>()
.ToDictionary(y => y.Name, y => y.Value.ToString())
}).ToList();
foreach (var user in myData) // Also tried myData.ToList()
{
if (string.IsNullOrEmpty(user.ID))
{
user.ID = index.ToString();
user.Properties["id"] = index.ToString();
}
index++;
}

public class MyData
{
public MyData()
{
this.Properties = new Dictionary<string,string>();
}
public string ID { get; set; }
public Dictionary<string, string> Properties { get; set; }
}
public static void Main(string[] args)
{
IEnumerable<MyData> myDataList = new List<MyData>();
int index = 0; // Assuming your starting point is 0
foreach (var obj in myDataList)
{
if (obj != null && string.IsNullOrEmpty(obj.ID))
{
obj.ID = index.ToString();
// Checks if the Properties dictionary has the key "id"
if (obj.Properties.ContainsKey("id"))
{
// If it does, then update it
obj.Properties["id"] = obj.ID;
}
else
{
// Else add it to the dictionary
obj.Properties.Add("id", obj.ID);
}
}
index++;
}
I believe the reason why your objects are not updating because it's probably still referring to the memory block before your objects were changed. Perhaps. The easiest way (that I can think of, there are thousands of smarter programmers than me) is to create a new list and have it contain all of your updated objects.
Edit
I updated the code above with the code that I have. I created a method to set a small amount of objects to test:
private static IEnumerable<MyData> GetMyData()
{
return new List<MyData>()
{
new MyData(),
new MyData() {ID = "2"},
new MyData() {ID = "3"},
new MyData()
};
}
I was able to view my changes and then go through a foreach loop to view my changes. If the ID of the object is Null or Empty, then it steps into the if check and adds the current index to the ID as you know.
Now for my question: Which "id" is blank? The "id" in the dictionary or is it the ID of the model? Are all of your (Model).ID blank? As the updated code of yours, if your dictionary doesn't have "id" as a key, it's going to throw an exception saying it doesn't exist so you will need to do a check to make sure it does exist or add it if it doesn't.

Implementing a find system for a ListView in C#

I am creating an application in C# using a ListView control that lets you create lists. I am implementing a Find function using the Find() method. Here’s my code:
if (findTextBox.Text != "")
{
ListViewItem[] lviFoundList = listItemsList.Items.Find(findTextBox.Text, true);
amountFound.Text = "Found " + Convert.ToString(lviFoundList.Count());
if (lviFoundList.Count() != 0)
{
int firstItemIndex = lviFoundList[0].Index;
listItemsList.Items[firstItemIndex].Selected = true;
}
}
else
{
amountFound.Text = "Found 0";
}
However, it doesn’t return any matches. What am I doing wrong?

Find method requires your listView item's Name, did you set your list view item's name property? If you want to search for text you can use this:
var lviFoundList = new List<ListViewItem>();
foreach(var item in listItemsList.Items)
{
if(item.Text == findTextBox.Text) lviFoundList.Add(item);
}

The Find() Method looks at the ListViewItem's name, not it's text.
You want this instead:
if (findTextBox.Text != "")
{
List<ListViewItem> items = new List<ListViewItems>();
foreach ListViewItem lvi in listItemsList.Items
{
if (lvi.Text == findTextBox.Text)
items.Add(lvi);
}
amountFound.Text = "Found " + Convert.ToString(lviFoundList.Count());
if(lviFoundList.Count() != 0)
{
int firstItemIndex = lviFoundList[0].Index;
listItemsList.Items[firstItemIndex].Selected = true;
}
}
else
{
amountFound.Text = "Found 0";
}

Honestly, the ListView.Find() method is rather poor and it's much easier to roll your own with LINQ. Think about what Find really is trying to accomplish -- a specific filtering, typically one record.
The first step, if you haven't already, would be to keep a cached collection of your data objects. Let's assume you have a list of Person classes like so:
class Person
{
public string FirstName { get; set; }
public string LastName { get; set; }
public string JobTitle { get; set; }
public DateTime DateOfBirth { get; set; }
}
Then in your MainForm you have a ListView and a member variable people defined as a List<Person>. Your ListView.Items should reflect the contents of this List<Person>.
So now maybe you want to find a person based on their FirstName or LastName, right? You could use LINQ in a function like so:
int FindFirstIndexOfPersonNamed(string firstOrLastName)
{
// WARNING: This is case sensitive!
return people.FindIndex(p => p.FirstName.Contains(firstOrLastName) || p.LastName.Contains(firstOrLastName));
}
Since your ListView.Items should be reflecting your List<Person> the index should be identical:
// Get the found item and do whatever you want with it...
var selectedListViewItem = listView.Items[index];