使用DataTable进行统计与分析(.net)

1. Expression属性
DataColumn有一个Expression属性，可以接收一个以列名为基础的算术表达式。定义了Expression属性的列称为计算列。

以下代码创建表结构：
private void CreateDBStru()
        {
            DataColumn dc = new DataColumn("ItemName",typeof(string));
            dc.ReadOnly = true;
            dt.Columns.Add(dc);

            dc = new DataColumn("UnitPrice",typeof(float));
            dt.Columns.Add(dc);

            dc = new DataColumn("Amount",typeof(int));
            dt.Columns.Add(dc);

            dc = new DataColumn("ItemTotal",typeof(float));
            dc.Expression = "UnitPrice*Amount";
            dt.Columns.Add(dc);

            dt.DefaultView.AllowNew = false;

            this.dataGrid1.DataSource = dt;
        }随机添加10条记录
private void AddRecord()
        {
            int count=0;
            DataRow dr;
            Random ran = new Random();
            for(int i=0; i<10; i++)
            {
                count++;
                dr = dt.NewRow();
                dr["ItemName"] = "商品" + count.ToString();
                dr["UnitPrice"] = ran.NextDouble() * 10;
                dr["Amount"] = ran.Next(1,1000);
                dt.Rows.Add(dr);
            }
        }代码中没有一句是直接给ItemTotal赋值，但其值可由计算机自动生成。

如果通过DataRelation定义了数据间的关系，则可以通过计算列对子行进行计算，只需在Expression属性中使用Child关键字，比如假设上面的例子中出现的是另一个DataTable的子表，则在父表中可以建立一个计算列，设置其Expression属性为：
SUM(Child.ItemTotal)
每当子表中ItemTotal列被更新时，父表中对应列自动更新。
2. Compute方法
Datatable的Compute方法可以针对一批数据求和、平均值和方差等。
格式如下：
Object sumObj = dt.Compute("SUM(列名)","筛选条件");//筛选条件为空表示全体
代码示例如下:
private void Analyze()
        {
            DataTable dtAnalyze = dt.Clone();

            DataRow dr = dtAnalyze.NewRow();
            dr["ItemName"] = "平均：";
            dr["UnitPrice"] = dt.Compute("AVG(UnitPrice)","");
            dr["Amount"] = dt.Compute("AVG(Amount)","");
            dr["ItemTotal"] = dt.Compute("AVG(ItemTotal)","");
            dtAnalyze.Rows.Add(dr);

            //dt.Compute("Sum(Age)", "ID like ’%%’").

            dr = dtAnalyze.NewRow();
            dr["ItemName"] = "最大值：";
            dr["UnitPrice"] = dt.Compute("MAX(UnitPrice)","");
            dr["Amount"] = dt.Compute("MAX(Amount)","");
            dr["ItemTotal"] = dt.Compute("MAX(ItemTotal)","");
            dtAnalyze.Rows.Add(dr);

            dr = dtAnalyze.NewRow();
            dr["ItemName"] = "最小值：";
            dr["UnitPrice"] = dt.Compute("MIN(UnitPrice)","");
            dr["Amount"] = dt.Compute("MIN(Amount)","");
            dr["ItemTotal"] = dt.Compute("MIN(ItemTotal)","");
            dtAnalyze.Rows.Add(dr);

            dr = dtAnalyze.NewRow();
            dr["ItemName"] = "方差：";
            dr["UnitPrice"] = dt.Compute("VAR(UnitPrice)","");
            dr["Amount"] = dt.Compute("VAR(Amount)","");
            dr["ItemTotal"] = dt.Compute("VAR(ItemTotal)","");
            dtAnalyze.Rows.Add(dr);

            dr = dtAnalyze.NewRow();
            dr["ItemName"] = "求和：";
            dr["UnitPrice"] = dt.Compute("SUM(UnitPrice)","");
            dr["Amount"] = dt.Compute("SUM(Amount)","");
            dr["ItemTotal"] = dt.Compute("SUM(ItemTotal)","");
            dtAnalyze.Rows.Add(dr);

            dtAnalyze.DefaultView.AllowNew = false;

            this.dataGrid2.DataSource = dtAnalyze;
        }