具有更改变量的C＃列表

Question

I have a List containing an object like this:

List<unit> unitlist = new List<unit>();

With unit beeing initialized like this:

 public class unit
    {
        public string[] records;
    }

I then use a variable to add into the list:

var temp = new unit();
temp.records = csv.GetFieldHeaders; // using lumenworks framework to read a csv
unitlist.Add(temp);

When I now override temp with a new item line from the csv, the entry in the list unitlist is also changed:

while (csv.ReadNextRecord())
{
    for (int i = 0; i < csv.Fieldcount; i++) 
    {
        // Read the csv entry in the temp variable
        temp.records[i] = csv[i];
    }
    // check for specific field, and write to list
    if (temp.records[8] == "Ja")
        unitlist.Add(temp);
}

When I now check unitlist, all the entries are the last read line from the csv, because they all are changed when the temp-variable changes. Why is that the case? How can I separate the List unitlist from the variable temp?

Answer 1

Because you're using the same bucket to store things in.

If you create temp every time, this should fix your problem

var header = new unit();
header.records = csv.GetFieldHeaders; 
unitlist.Add(header);

...

while (csv.ReadNextRecord())
{
    var temp = new unit();
    temp.records = new string[header.records.Length]

    for (int i = 0; i < csv.Fieldcount; i++) 
    {
        // Read the csv entry in the temp variable
        temp.records[i] = csv[i];
    }
    // check for specific field, and write to list
    if (temp.records[8] == "Ja")
        unitlist.Add(temp);
}

Answer 2

You need to create temp object in each iteration like below. Please try this and check.

while (csv.ReadNextRecord())
{
    var temp = new unit();
    temp.records = csv.GetFieldHeaders;

    for (int i = 0; i < csv.Fieldcount; i++) 
    {
        // Read the csv entry in the temp variable
        temp.records[i] = csv[i];
    }
    // check for specific field, and write to list
    if (temp.records[8] == "Ja")
        unitlist.Add(temp);
}

Answer 3

When you create the temp variable, it is referencing a location in memory where your object data is allocated. When you add it to unitlist , a reference is added to the list that points to the same location in memory.

Now, when you change temp.records[i] , it is updated in that same memory location. So, you end up with a list of items, all pointing to the same object in memory, containing the last records in the CSV file.

Simply adding a temp = new unit(); at the beginning of your while loop will cause each iteration to allocation a new object, with a new memory location, and have temp reference it.