Evaluating string “3*(4+2)” yield int 18 [duplicate]
https://stackoverflow.com/questions/333737
italiano
english
français
española
中国
日本の
العربية
Deutsch
한국어
Português
RussianQuestion
This question already has an answer here:
Is there a function the .NET framework that can evaluate a numeric expression contained in a string and return the result? F.e.:
string mystring = "3*(2+4)";
int result = EvaluateExpression(mystring);
Console.Writeln(result); // Outputs 18
Is there a standard framework function that you can replace my EvaluateExpression method with?
Solution
Yes, you can let C# compiler evaluate it at runtime.
See: CSharpCorner
OTHER TIPS
If you want to evaluate a string expression use the below code snippet.
using System.Data;
DataTable dt = new DataTable();
var v = dt.Compute("3 * (2+4)","");
Using the compiler to do implies memory leaks as the generated assemblies are loaded and never released. It's also less performant than using a real expression interpreter. For this purpose you can use Ncalc which is an open-source framework with this solely intent. You can also define your own variables and custom functions if the ones already included aren't enough.
Example:
Expression e = new Expression("2 + 3 * 5");
Debug.Assert(17 == e.Evaluate());
Try this:
static double Evaluate(string expression) {
var loDataTable = new DataTable();
var loDataColumn = new DataColumn("Eval", typeof (double), expression);
loDataTable.Columns.Add(loDataColumn);
loDataTable.Rows.Add(0);
return (double) (loDataTable.Rows[0]["Eval"]);
}
You could look at "XpathNavigator.Evaluate" I have used this to process mathematical expressions for my GridView and it works fine for me.
Here is the code I used for my program:
public static double Evaluate(string expression)
{
return (double)new System.Xml.XPath.XPathDocument
(new StringReader("<r/>")).CreateNavigator().Evaluate
(string.Format("number({0})", new
System.Text.RegularExpressions.Regex(@"([\+\-\*])")
.Replace(expression, " ${1} ")
.Replace("/", " div ")
.Replace("%", " mod ")));
}
static double Evaluate(string expression) {
var loDataTable = new DataTable();
var loDataColumn = new DataColumn("Eval", typeof (double), expression);
loDataTable.Columns.Add(loDataColumn);
loDataTable.Rows.Add(0);
return (double) (loDataTable.Rows[0]["Eval"]);
}
Explanation of how it works:
First, we make a table in the part var loDataTable = new DataTable();, just like in a Data Base Engine (MS SQL for example).
Then, a column, with some specific parameters (var loDataColumn = new DataColumn("Eval", typeof (double), expression);).
The "Eval" parameter is the name of the column (ColumnName attribute).
typeof (double) is the type of data to be stored in the column, which is equal to put System.Type.GetType("System.Double"); instead.
expression is the string that the Evaluate method receives, and is stored in the attribute Expression of the column. This attribute is for a really specific purpose (obvious), which is that every row that's put on the column will be fullfilled with the "Expression", and it accepts practically wathever can be put in a SQL Query. Refer to http://msdn.microsoft.com/en-us/library/system.data.datacolumn.expression(v=vs.100).aspx to know what can be put in the Expression attribute, and how it's evaluated.
Then, loDataTable.Columns.Add(loDataColumn); adds the column loDataColumn to the loDataTable table.
Then, a row is added to the table with a personalized column with a Expression attribute, done via loDataTable.Rows.Add(0);. When we add this row, the cell of the column "Eval" of the table loDataTable is fullfilled automatically with its "Expression" attribute, and, if it has operators and SQL Queries, etc, it's evaluated and then stored to the cell, so, here happens the "magic", the string with operators is evaluated and stored to a cell...
Finally, just return the value stored to the cell of the column "Eval" in row 0 (it's an index, starts counting from zero), and making a conversion to a double with return (double) (loDataTable.Rows[0]["Eval"]);.
And that's all... job done!
And here a code eaiser to understand, which does the same... It's not inside a method, and it's explained too.
DataTable MyTable = new DataTable();
DataColumn MyColumn = new DataColumn();
MyColumn.ColumnName = "MyColumn";
MyColumn.Expression = "5+5/5"
MyColumn.DataType = typeof(double);
MyTable.Columns.Add(MyColumn);
DataRow MyRow = MyTable.NewRow();
MyTable.Rows.Add(MyRow);
return (double)(MyTable.Rows[0]["MyColumn"]);
First, create the table with DataTable MyTable = new DataTable();
Then, a column with DataColumn MyColumn = new DataColumn();
Next, we put a name to the column. This so we can search into it's contents when it's stored to the table. Done via MyColumn.ColumnName = "MyColumn";
Then, the Expression, here we can put a variable of type string, in this case there's a predefined string "5+5/5", which result is 6.
The type of data to be stored to the column MyColumn.DataType = typeof(double);
Add the column to the table... MyTable.Columns.Add(MyColumn);
Make a row to be inserted to the table, which copies the table structure DataRow MyRow = MyTable.NewRow();
Add the row to the table with MyTable.Rows.Add(MyRow);
And return the value of the cell in row 0 of the column MyColumn of the table MyTable with return (double)(MyTable.Rows[0]["MyColumn"]);
Lesson done!!!
This is a simple Expression Evaluator using Stacks
public class MathEvaluator
{
public static void Run()
{
Eval("(1+2)");
Eval("5*4/2");
Eval("((3+5)-6)");
}
public static void Eval(string input)
{
var ans = Evaluate(input);
Console.WriteLine(input + " = " + ans);
}
public static double Evaluate(String input)
{
String expr = "(" + input + ")";
Stack<String> ops = new Stack<String>();
Stack<Double> vals = new Stack<Double>();
for (int i = 0; i < expr.Length; i++)
{
String s = expr.Substring(i, 1);
if (s.Equals("(")){}
else if (s.Equals("+")) ops.Push(s);
else if (s.Equals("-")) ops.Push(s);
else if (s.Equals("*")) ops.Push(s);
else if (s.Equals("/")) ops.Push(s);
else if (s.Equals("sqrt")) ops.Push(s);
else if (s.Equals(")"))
{
int count = ops.Count;
while (count > 0)
{
String op = ops.Pop();
double v = vals.Pop();
if (op.Equals("+")) v = vals.Pop() + v;
else if (op.Equals("-")) v = vals.Pop() - v;
else if (op.Equals("*")) v = vals.Pop()*v;
else if (op.Equals("/")) v = vals.Pop()/v;
else if (op.Equals("sqrt")) v = Math.Sqrt(v);
vals.Push(v);
count--;
}
}
else vals.Push(Double.Parse(s));
}
return vals.Pop();
}
}
This is right to left execution, so need to use proper parathesis to execute expression
// 2+(100/5)+10 = 32
//((2.5+10)/5)+2.5 = 5
// (2.5+10)/5+2.5 = 1.6666
public static double Evaluate(String expr)
{
Stack<String> stack = new Stack<String>();
string value = "";
for (int i = 0; i < expr.Length; i++)
{
String s = expr.Substring(i, 1);
char chr = s.ToCharArray()[0];
if (!char.IsDigit(chr) && chr != '.' && value != "")
{
stack.Push(value);
value = "";
}
if (s.Equals("(")) {
string innerExp = "";
i++; //Fetch Next Character
int bracketCount=0;
for (; i < expr.Length; i++)
{
s = expr.Substring(i, 1);
if (s.Equals("("))
bracketCount++;
if (s.Equals(")"))
if (bracketCount == 0)
break;
else
bracketCount--;
innerExp += s;
}
stack.Push(Evaluate(innerExp).ToString());
}
else if (s.Equals("+")) stack.Push(s);
else if (s.Equals("-")) stack.Push(s);
else if (s.Equals("*")) stack.Push(s);
else if (s.Equals("/")) stack.Push(s);
else if (s.Equals("sqrt")) stack.Push(s);
else if (s.Equals(")"))
{
}
else if (char.IsDigit(chr) || chr == '.')
{
value += s;
if (value.Split('.').Length > 2)
throw new Exception("Invalid decimal.");
if (i == (expr.Length - 1))
stack.Push(value);
}
else
throw new Exception("Invalid character.");
}
double result = 0;
while (stack.Count >= 3)
{
double right = Convert.ToDouble(stack.Pop());
string op = stack.Pop();
double left = Convert.ToDouble(stack.Pop());
if (op == "+") result = left + right;
else if (op == "+") result = left + right;
else if (op == "-") result = left - right;
else if (op == "*") result = left * right;
else if (op == "/") result = left / right;
stack.Push(result.ToString());
}
return Convert.ToDouble(stack.Pop());
}
You could fairly easily run this through the CSharpCodeProvider with suitable fluff wrapping it (a type and a method, basically). Likewise you could go through VB etc - or JavaScript, as another answer has suggested. I don't know of anything else built into the framework at this point.
I'd expect that .NET 4.0 with its support for dynamic languages may well have better capabilities on this front.
I recently needed to do this for a project and I ended up using IronPython to do it. You can declare an instance of the engine, and then pass any valid python expression and get the result. If you're just doing simple math expressions, then it would suffice. My code ended up looking similar to:
IronPython.Hosting.PythonEngine pythonEngine = new IronPython.Hosting.PythonEngine();
string expression = "3*(2+4)";
double result = pythonEngine.EvaluateAs<double>(expression);
You'd probably not want to create the engine for each expression. You also need a reference to IronPython.dll
EDIT: Realised i should really bring the addition and subtraction out seperately aswell to make it a little bit more BODMAS compliant.
Big thanks to Rajesh Jinaga for his Stack based approach. I found it really useful for my needs. The following code is a slight modification of Rajesh's method, which processes divisions first, then multiplications, then finishes up with addition and subtraction. It will also allow the use of booleans in the expressions, where true is treated as 1 and false 0. allowing the use of boolean logic in expressions.
public static double Evaluate(string expr)
{
expr = expr.ToLower();
expr = expr.Replace(" ", "");
expr = expr.Replace("true", "1");
expr = expr.Replace("false", "0");
Stack<String> stack = new Stack<String>();
string value = "";
for (int i = 0; i < expr.Length; i++)
{
String s = expr.Substring(i, 1);
// pick up any doublelogical operators first.
if (i < expr.Length - 1)
{
String op = expr.Substring(i, 2);
if (op == "<=" || op == ">=" || op == "==")
{
stack.Push(value);
value = "";
stack.Push(op);
i++;
continue;
}
}
char chr = s.ToCharArray()[0];
if (!char.IsDigit(chr) && chr != '.' && value != "")
{
stack.Push(value);
value = "";
}
if (s.Equals("("))
{
string innerExp = "";
i++; //Fetch Next Character
int bracketCount = 0;
for (; i < expr.Length; i++)
{
s = expr.Substring(i, 1);
if (s.Equals("(")) bracketCount++;
if (s.Equals(")"))
{
if (bracketCount == 0) break;
bracketCount--;
}
innerExp += s;
}
stack.Push(Evaluate(innerExp).ToString());
}
else if (s.Equals("+") ||
s.Equals("-") ||
s.Equals("*") ||
s.Equals("/") ||
s.Equals("<") ||
s.Equals(">"))
{
stack.Push(s);
}
else if (char.IsDigit(chr) || chr == '.')
{
value += s;
if (value.Split('.').Length > 2)
throw new Exception("Invalid decimal.");
if (i == (expr.Length - 1))
stack.Push(value);
}
else
{
throw new Exception("Invalid character.");
}
}
double result = 0;
List<String> list = stack.ToList<String>();
for (int i = list.Count - 2; i >= 0; i--)
{
if (list[i] == "/")
{
list[i] = (Convert.ToDouble(list[i - 1]) / Convert.ToDouble(list[i + 1])).ToString();
list.RemoveAt(i + 1);
list.RemoveAt(i - 1);
i -= 2;
}
}
for (int i = list.Count - 2; i >= 0; i--)
{
if (list[i] == "*")
{
list[i] = (Convert.ToDouble(list[i - 1]) * Convert.ToDouble(list[i + 1])).ToString();
list.RemoveAt(i + 1);
list.RemoveAt(i - 1);
i -= 2;
}
}
for (int i = list.Count - 2; i >= 0; i--)
{
if (list[i] == "+")
{
list[i] = (Convert.ToDouble(list[i - 1]) + Convert.ToDouble(list[i + 1])).ToString();
list.RemoveAt(i + 1);
list.RemoveAt(i - 1);
i -= 2;
}
}
for (int i = list.Count - 2; i >= 0; i--)
{
if (list[i] == "-")
{
list[i] = (Convert.ToDouble(list[i - 1]) - Convert.ToDouble(list[i + 1])).ToString();
list.RemoveAt(i + 1);
list.RemoveAt(i - 1);
i -= 2;
}
}
stack.Clear();
for (int i = 0; i < list.Count; i++)
{
stack.Push(list[i]);
}
while (stack.Count >= 3)
{
double right = Convert.ToDouble(stack.Pop());
string op = stack.Pop();
double left = Convert.ToDouble(stack.Pop());
if (op == "<") result = (left < right) ? 1 : 0;
else if (op == ">") result = (left > right) ? 1 : 0;
else if (op == "<=") result = (left <= right) ? 1 : 0;
else if (op == ">=") result = (left >= right) ? 1 : 0;
else if (op == "==") result = (left == right) ? 1 : 0;
stack.Push(result.ToString());
}
return Convert.ToDouble(stack.Pop());
}
I know there is likely to be a cleaner way of doing it, thought id just share the first look at it in case anyone finds it usefull.
Many thanks to Ramesh. I used a version of his simple code to pull a string out a database and use it to do boolean operations in my code.
x is a number like 1500 or 2100 or whatever.
function would be a stored evaluation like x > 1400 and x < 1600
function = relation[0].Replace("and","&&").Replace("x",x);
DataTable f_dt = new DataTable();
var f_var = f_dt.Compute(function,"");
if (bool.Parse(f_var.ToString()) { do stuff }
There is not. You will need to use some external library, or write your own parser. If you have the time to do so, I suggest to write your own parser as it is a quite interesting project. Otherwise you will need to use something like bcParser.
Short answer: I don't think so. C# .Net is compiled (to bytecode) and can't evaluate strings at runtime, as far as I know. JScript .Net can, however; but I would still advise you to code a parser and stack-based evaluator yourself.
