We Keep Coding

How do you return an array in a procedure back to the main procedure in C#?

static string readfileName(string[] name)
{
using (StreamReader file = new StreamReader("StudentMarks.txt"))
{
int counter = 0;
string ln;
while ((ln = file.ReadLine()) != null)
{
if (ln.Length > 4)
{
name[counter] = ln;
counter++;
}
}
file.Close();
return name;
}
}
This is the procedure I'm currently trying to return the array name[50] but the compile time error I can't fix states
"Error CS0029 Cannot implicitly convert type 'string[]' to 'string' "

You don't need to. Your main method passed the array to this method, this method filled it. It doesn't need to hand it back because the object pointed to by your 'name` variable is the same object as pointed to by the original variable in the main method; your main method already has all the array data:
static void Main(){
var x = new string[10];
MyMethod(x);
Console.Write(x[0]); //prints "Hello"
}
static void MyMethod(string[] y){
y[0] = "Hello";
}
In this demo code above we start out with an array of size 10 that is referred to by a variable x. In memory it looks like:
x --refers to--> arraydata
When you call MyMethod and pass x in, c# will create another reference y that points to the same data:
x --refers to--> arraydata <--refers to-- y
Now because both references point to the same area of memory anything that you do with y, will also affect what x sees. You put a string (like I did with Hello) in slot 0, both x and y see it. When MyMethod finishes, the reference y is thrown away, but x survives and sees all the changes you made when working with y
The only thing you can't do is point y itself to another different array object somewhere else in memory. That won't change x. You can't do this:
static void MyMethod(string[] y){
y = new string[20];
}
If you do this your useful reference of x and y pointing to the same area of memory:
x ---> array10 <--- y
Will change to:
x ---> array10 y ---> array20
And then the whole array20 and the y reference will be thrown away when MyMethod finishes.
The same rule applies if you call a method that supplies you an array:
static void MyMethod(string[] y){
y = File.ReadAllLines("some path"); //this also points y away to a new array made by ReadAllLines
}
It doesn't matter how or who makes the new array. Just remember that you can fiddle with the contents of an object pointed to by y all you like and the changes will be seen by x, but you can't change out the entire object pointed to by y and hope x will see it
in that case you WOULD have to pass it back when you're done:
static string[] MyMethod(string[] y){
y = new ...
return y;
}
And the main method would have to capture the change:
Main(...){
string[] x = new string[10];
string[] result = MyMethod(x);
}
Now, while I'm giving this mini lesson of "pass by reference" and "pass by value" (which should have been called "pass by original reference" and "pass by copy of reference") it would be useful to note that there is a way to change things so MyMethod can swap y out for a whole new object and x will see the change too.
We don't really use it, ever; there is rarely any need to. Just about the only time it's used is in things like int.Parse. I'm telling you for completeness if education so that if you encounter it you understand it but you should always prefer a "change the contents but not the whole object" or a "if you make a new object pass it back" approach
By marking the y argument with the ref keyword, c# wont make a copy of the reference when calling the method, it will use the original reference and temporarily allow you to call it y:
static void MyMethod(ref string[] y){
y = new array[20];
}
Our diagram:
x ---> array10data
Temporarily becomes:
x a.k.a y ---> array10data
So if you point y to a new array, x experiences the change too, because they're the same reference; y is no longer a different reference to the same data
x a.k.a y ---> array20data
Like I say, don't use it- we always seek to avoid it for various reasons.
Now, I said at the start "you don't need to" - by that, and for the reasons above, I meant you don't need to return anything from this method
Your method receives the array it shall fill (from the file) as a parameter; it doesn't make a new array anywhere so there isn't any need to return the array when done. It will just put any line longer than 4 chars into an array slot. It could then finish without returning anything and the method that called this method will see the changes it made in the array. This is just like my code, where MyMethod changes slot 0 of the array, MyMethod was declared as void so it didn't need to make a return statement , and my Main method god could still see the Hello that I put in the array. In the same vein, your Main method will see all those lines from the file if you make your ReadFileName method (which should perhaps be called FillArray) because it fills the array called name
The most useful thing your method could return is actually an integer saying how many lines were read; the array passed in is of a fixed size. You can't resize it because that entails making a new array which won't work for all those reasons I talked about above. If you were to make a new array and return it there wouldn't be any point in passing an array in.
There are thus several ways we could improve this code but to my mind they come down to two:
don't pass an array in; let this method make a new array and return it. The new array passed back can be exactly sized to fit
keep with the "pass an array in" idea and return an integer of how many lines were actually read instead
For the second idea (which is the simplest to implement) you have to change the return type to int:
static int ReadFileName(string[] name)
And you have to return that variable you use to track which slot to put the next thing in, counter. Counter is always 1 greater than the number of things you've stored so:
return counter - 1;
Your calling method can now look like:
string[] fileData = new string[10000]; //needs to be big enough to hold the whole file!
int numberOfLinesRead = ReadFileName(fileData);
Can you see now why ReadFileName is a bad name for the method? Calling it FillArrayFromFile would be better. This last line of code doesn't read like a book, it doesn't make sense from a natural language perspective. Why would something that looks like it reads a file name (if that even makes sense) take an array and return an int - calling it ReadFileName makes it sound more like it searches an array for a filename and returns the slot number it was found in. Here ends the "name your methods appropriately 101"
So the other idea was to have the Read method make its own array and return it. While we are at it, let's call it ReadFileNamed, and have it take a file path in so it's not hard coded to reading just that one file. And we will have it return an array
static string[] ReadFileNamed(string filepath)
^^^^^^^^ ^^^^^^^^^^^^^^^
the return type the argument passed in
Make it so the first thing it does is declare an array big enough to hold the file (there are still problems with this idea, but this is programming 101; I'll let them go. Can't fix everything using stuff you haven't been taught yet)
Put this somewhere sensible:
string lines = new string[10000];
And change all your occurrences of "name" to be "lines" instead - again we name our variables we'll just like we name our methods sensibly
Change the line that reads the fixed filename to use the variable name we pass in..
using (StreamReader file = new StreamReader(filepath))
At the end of the method, the only thing left to do is size the array accurately before we return it. For a 49 line file, counter will be 50 so let's make an array that is 49 big and then fill it using a loop (I doubt you've been shown Array.Copy)
string[] toReturn = new string[counter-1];
for(int x = 0; x < toReturn.Length; x++)
toReturn[x] = lines[x];
return toReturn;
And now call it like this:
string[] fileLines = ReadFileNamed("student marks.txt");

If you're looking to return name[50] and you know that will be populated, why not go with:
static string readfileName(string[] name)
{
using (StreamReader file = new StreamReader("StudentMarks.txt"))
{
int counter = 0;
string ln;
while ((ln = file.ReadLine()) != null)
{
if (ln.Length > 4)
{
name[counter] = ln;
counter++;
}
}
file.Close();
return name[50];
}
}
You're getting the error because your method signature indicates that you're going to return a string, but you're defining name as a string[] in the argument. If you simply select a single index of your array in the return statement, you'll only return a string.

You have defined your method to return a string, yet the code inside is returning name, which is a string[]. If you want it to return a string[], then change the signature to specify that:
static string[] ReadFileName(string[] name)
However, since your method is only populating the array that was passed in, it's not really necessary to return the array, since the caller already has a reference to the array we're modifying (they passed it to our method in the first place).
There is a potential problem here, though
We're expecting the caller to pass us an array of the appropriate length to hold all the valid lines from the file, yet that number is unknown until we read the file. We could return an array of the size they specified with either empty indexes at the end if it was too big, or incomplete data if it was too small, but instead we should probably just return them a new array, and not require them to pass one to us.
Note that it's easier to use a List<string> instead of a string[], since lists don't require any knowledge of their size at instantiation (they can grow dynamically). Also, we no longer need a counter variable (since we're using the Add method of the list to add new items), and we can remove the file.Close() call since the using block will call that automatically (one of the cool things about them):
static string[] ReadFileName()
{
List<string> validLines = new List<string>();
using (StreamReader file = new StreamReader("StudentMarks.txt"))
{
string ln;
while ((ln = file.ReadLine()) != null)
{
if (ln.Length > 4)
{
validLines.Add(ln);
}
}
}
return validLines.ToArray();
}
And we can simplify the code even more if we use some static methods of the System.IO.File class:
static string[] ReadFileName()
{
return File.ReadLines("StudentMarks.txt").Where(line => line.Length > 4).ToArray();
}
We could also make the method a little more robust by allowing the caller to specify the file name as well as the minimum line length requirement:
static string[] ReadFileName(string fileName, int minLineLength)
{
return File.ReadLines(fileName)
.Where(line => line.Length >= minLineLength).ToArray();
}

Well, you are trying to do several thing in one method:
Read "StudentMarks.txt" file
Put top lines into name existing array (what if you have too few lines in the file?)
return 50th (magic number!) item
If you insist on such implementation:
using System.Linq;
...
static string readfileName(string[] name)
{
var data = File
.ReadLines("StudentMarks.txt")
.Where(line => line.Length > 4)
.Take(name.Length);
int counter = 0;
foreach (item in data)
if (counter < name.Length)
name[counter++] = item;
return name.Length > 50 ? name[50] : "";
}
However, I suggest doing all things separately:
// Reading file lines, materialize them into string[] name
string[] name = File
.ReadLines("StudentMarks.txt")
.Where(line => line.Length > 4)
// .Take(51) // uncomment, if you want at most 51 items
.ToArray();
...
// 50th item of string[] name if any
string item50 = name.Length > 50 ? name[50] : "";
Edit: Splitting single record (name and score) into different collections (name[] and score[]?) often is a bad idea;
the criterium itself (line.Length > 4) is dubious as well (what if we have Lee - 3 letter name - with 187 score?).
Let's implement Finite State Machine with 2 states (when we read name or score) and read (name, score) pairs:
var data = File
.ReadLines("StudentMarks.txt")
.Select(line => line.Trim())
.Where(line => !string.IsNullOrEmpty(line));
List<(string name, int score)> namesAndScores = new List<(string name, int score)>();
string currentName = null;
foreach (string item in data) {
if (null == currentName)
currentName = item;
else {
namesAndScores.Add((currentName, int.Parse(item)));
currentName = null;
}
}
Now it's easy to deal with namesAndScores:
// 25th student and his/her score:
if (namesAndScores.Count > 25)
Console.Write($"{namesAndScores[25].name} achieve {namesAndScores[25].score}");

Parsing string as int with "Model.x" property in for loop remains in string format

I have a simple form that accepts a number from a radio button selection (1-5) of 11 questions and posts the values into a database as varchar(10) data. I intend to send the user to a result page that lists the sum of these scores through a simple for loop, but when I try parsing the data to integer format, it simply results in zero due to error parsing. Here's an example of my code:
// Q1 - Q11 are the questions in my Db, using Model property
int sum = 0;
int temp = 0;
String question;
for (int i = 11; i >= 1; i--)
{
question = "Model.Q" + i.ToString();
temp = int.Parse(question);
sum += temp;
}
return sum;
What's strange is that if I parse them individually, such as writing:
Int32.TryParse(Model.Q5, out temp);
I am able to parse the data just fine. My console shows that the loop keeps the question variable as "Model.Qx" with quotations, ultimately resulting in 0 for the sum. I have also tried using Int32.TryParse(); for that as well and it resulted in no difference, besides handling the error.
Can a string simply not be parsed if it contains punctuation in concatenation with the i variable, or am I missing something else here? I want to avoid parsing each question individually, as it looks rather ugly in code.
Thanks in advance.

You problem is that you're trying to access a variable by using a string with the same name. This won't work, in the same way that the name gitgecko is not you.
If your model has got a number of properties with similar names, you could write a function to switch between them:
object GetQ(int number)
{
switch(number)
{
case 1: return Model.Q1;
case 2: return Model.Q2;
// etc...
}
}
Or you could change your model to store these variables in an array or list, or whatever is appropriate.
For example, if you've currently got:
class Model
{
string Q1;
string Q2:
// repeated 11 times
You could have:
class Model
{
string[] Q = new string[11];
}
which gives you the ability to do Model.Q[x]

Catching value of the concatenated variable

My environment is: W7 Ultimate 64 bits, VS2010 Ultimate, C#, WinForm, target XP and W7.
With the help of #dasblinkenlight, the concatenation on the for loop was very good.
I feel we are making great progress.
As you ca see, we are putting into the array sMntHour[d,h] the string "csv_001_01" if d=1 and h=1 and so on.
This csv_001_01, csv_001_02,.. ; are variables that contains an integer value.
csv_001_01=5111;
csv_001_02=236; // This is a sample, because has 365 days in normal year
// and 366 days in leaf year. "csv_day_hour"
Directly we could do this:
sMntHour[d,h] = csv_001_01.ToString(); // d is day and h is hour
sMntHour[d,h] = csv_001_02.ToString();
As we put the value of this concatenated variable in the array and not the name of the variables?
for(int d=1;d<=365;d++) //I'll put the code to not leap years.
{
for(int h=1; h<=24; h++)
{
sMntHour[d,h] = string.Format("csv_{0:000}_{1:00}", d, h)
}
}

If I understand what you mean, you have all the variable names and now you want to get their values.
You could do this using Reflection and you can create a dictionary where keys are the variable names and the values are the actual values. It is really hard to help without seeing how these variables are declared, are they fields / properties ? are the private, static ? etc... But something like this should work, in theory:
var type = this.GetType();
var values = sMntHour.OfType<string>()
.ToDictionary(
x => x,
x => (int)type.GetField(x).GetValue(this));
Then you can access the values using values["variable_name"]
Or if you don't want this, instead if you want to access them using index like [d,h] as mentioned in comments, do not store the variable names in the first place instead store the values in your array:
var type = this.GetType();
for(int d=1;d<=365;d++)
{
for(int h=1; h<=24; h++)
{
var name = string.Format("csv_{0:000}_{1:00}", d, h);
sMntHour[d,h] = (int)type.GetField(name).GetValue(this);
}
}
Ofcourse you need to change the type of sMntHour, in order to make it work.

An object reference is required for the non-static field error

I have a long method that ends up returning a DataTable. The method opens like this:
public DataTable ImportCsvData(string folderPath, string[] csvFileNames, int compIdValue)
{
int countDupFiles = 0;// Count number of previously imported csv files
int countImportedFiles = 0;// Count imported files
for (int i = 0; i < csvFileNames.Length; i++)
{
frmImportCsvData.importProgressBar(i);
This method is inside a class where I am doing all of my SQL requests but is called from another class that controls my Form. The last line of code above calls a method back in my Form class which should update my progress bar. The code for that method is:
public void importProgressBar(int i)
{
progressTableLayoutPanel.Visible = true;//display progress bar
int percProgress = 100 * (i + 1) / csvFileNames.Length;
while (percProgress < 100)
{
if (percProgress <= 99)// Required to prevent values above 100 that crash the code
progressBar.Value = percProgress + 1;//hack that makes the progress bar update when progress value decreases
progressBar.Value = percProgress;
percProgressLabel.Text = percProgress.ToString();
progressTableLayoutPanel.Update();//Required to display all progress bar table contents
//Thread.Sleep(200);
}
progressTableLayoutPanel.Visible = false;
}
This gives me the error: An object reference is required for the non-static field, method, or property 'BetfairDatabaseMDI.frmImportCsvData.importProgressBar(int)'. If I set the importProgressBar method to static I get a bunch of other similar errors. Could anybody advise me how to fix this please?

frmImportCsvData is a class, so a type, which is an abstract thing, not a real object in memory. The error message says you need an instance of a type, so a concrete, existing thing, an object in memory of that given type.
So, essentially, you need to pass the instance of frmImportCsvData into ImportCsvData. For example, like this:
public DataTable ImportCsvData(frmImportCsvData myForm, …)
{
myForm.importProgressBar(i);
}
Then to call the method from within frmImportCsvData you just pass in this, which denotes the current instance (in run-time sense) of the enclosing class:
dataLayer.ImportCsvData(this, …);
Simplest doesn't mean best or correct in respect to the particular UI framework you are using (being it WinForms, WPF, whatever). BUT: You definitelly have to understand the differences between a type and an instance of a type first, before you run into any more complex constructs. So grab a C# book and get back to the basics, that will help you the most.
Note: It is a common convention to start identifiers of methods, classes and properties with an upper-case letter.

Inline use of function returning more than one value in C#

I am used to using functions that return a single value "in-line" like so:
Label1.Text = firstString + functionReturnSecondString(aGivenParameter);
Can this be done for a function that returns two values?
Hypothetical example:
label1.Text = multipleReturnFunction(parameter).firstValue
I have been looking into returning more than one value and it looks like the best options are using a tuple, struct, or an array list.
I made a working function that retuns a struct. However the way I got it to work I need to first call the function, then I can use the values. It doesn't seem possible to make it happen all on the same line without writing another function.
multipleReturnFunction(parameter);
Label1.Text = firstString + classOfStruct.secondString;
I haven't made a function that returns a tuple or array list yet, so I'm not sure. Is it possible to call those functions and reference the return values "inline"?
I appreciate your feedback.

I have a grotty hack for exactly this type of scenario - when you want to perform multiple operations on the return value without defining an extra variable to store it:
public static TResult Apply<TInput, TResult>(this TInput input, Func<TInput, TResult> transformation)
{
return transformation(input);
}
... and here's the reason it came about in the first place:
var collection = Enumerable.Range(1, 3);
// Average reimplemented with Aggregate.
double average = collection
.Aggregate(
new { Count = 0, Sum = 0 },
(acc, i) => new { Count = acc.Count + 1, Sum = acc.Sum + i })
.Apply(a => (double)a.Sum / (double)a.Count); // Note: we have access to both Sum and Count despite never having stored the result of the call to .Aggregate().
Console.WriteLine("Average: {0}", average);
Needless to say this is better suited for academic exercises than actual production code.

Alternatively, use the ref or they out keyword.
Example:
int a = 0, b = 0;
void DoSomething(ref int a, ref int b) {
a = 1;
b = 2;
}
Console.WriteLine(a); // Prints 1
Console.WriteLine(b); // Prints 2
It's not inline and I personally would consider a class or a struct before using the ref or the out keyword. Let's consider the theory: when you want to return multiple things, you have in fact an object that has multiple properties which you want to make available to the caller of your function.
Therefore it is much more correct to actually create an object (either by using a class or a struct) that represents what you want to make available and returning that.
The only time I use the ref or the out keyword is when using DLL imports because those functions often have pointers as their calling arguments and I personally don't see any benefit in using them in your typical normal application.

To do this inline, I think you would have to have another method that takes your struct and gives you the string you are looking for.
public string NewMethod(object yourStruct)
{
return string.Format("{0} {1}", yourStruct.value1, yourStruct.value2);
}
Then in the page, you do this:
Label1.Text = NewMethod(multipleReturnFunction(parameter));

C# doesn't have Inline functions, but it does support anonymous functions which can be closures.
With these techniques, you can say:
var firstString=default(String);
var secondString=default(String);
((Action<String>)(arg => {
firstString="abc"+arg;
secondString="xyz";
}))("wtf");
label1.Text=firstString+secondString;
Debug.Print("{0}", label1.Text);
((Action<String>)(arg => {
firstString="123"+arg;
secondString="456";
}))("???");
label1.Text=firstString+secondString;
Debug.Print("{0}", label1.Text);
or name the delegate and reuse it:
var firstString=default(String);
var secondString=default(String);
Action<String> m=
arg => {
firstString="abc"+arg;
secondString="xyz";
};
m("wtf");
label1.Text=firstString+secondString;
Debug.Print("{0}", label1.Text);
m("???");
label1.Text=firstString+secondString;
Debug.Print("{0}", label1.Text);
So, do you really need a method returns multiple values?

Each method can return only one value. Thats how methods defined in .NET
Methods are declared in a class or struct by specifying the access
level such as public or private, optional modifiers such as abstract
or sealed, the return value, the name of the method, and any method
parameters
If you need to return more than one value from method, then you have three options:
Return complex type which will hold all values. That cannot help you in this case, because you will need local variable to store value returned by method.
Use out parameters. Also not your case - you will need to declare parameters before method call.
Create another method, which does all work and returns single value.
Third option looks like
Label1.Text = AnotherMethod(parameters);
And implementation
public string AnotherMethod(parameters)
{
// use option 1 or 2 to get both values
// return combined string which uses both values and parameters
}
BTW One more option - do not return values at all - you can use method which sets several class fields.