Bdtjtk

I want to group by datatable using followng method. However i can not select a column which includes string.

dtTAConvert example;

SAMPLE_TIME  WAIT_CLASS 

23:11:12     CPU

23:11:12     IO

23:11:12     IO

23:11:11     CPU

23:11:11     CPU

What i want is this;

SAMPLE_TIME  WAIT_CLASS COUNT

23:11:12     CPU        1

23:11:12     IO         2

23:11:11     CPU        2

So far, i wrote below code;

var dtTAConvertList =

    (from dr1 in dtTAConvert.AsEnumerable()

     group dr1 by dr1.Field<DateTime>("SAMPLE_TIME") into g

     select new

     {

         SAMPLE_TIME = g.Key,

         WAIT_CLASS = g.Field<string>("WAIT_CLASS"),          // ==> I get error in this line

         COUNT = Math.Round(g.Sum(h => h.Field<double>("COUNT")) / 15.0, 2),

     });

It throws the exception

'IGrouping' does not contain a definition for 'Field' and the best extension method overload 'DataRowExtensions.Field(DataRow, string)' requires a receiver of type 'DataRow'

If i change above code to like below, it works;

WAIT_CLASS = g.First().Field<string>("WAIT_CLASS"),

However this returns only the first element, which is not work for me.

How can i select string column like as i explained?

You're receiving the said error because as shown IGrouping i.e. a collection of objects that have a common key does not contain a definition for 'Field'. you probably expected this to be a DataRow.

In addition, currently you're grouping only by SAMPLE_TIME which is not what you need, rather you need to group by both SAMPLE_TIME and WAIT_CLASS to get the aforementioned result.

Here's how to group by both of those fields via query syntax:

var result = from dr in dtTAConvert.AsEnumerable()

             group dr by new 

             {

                  sampleTime = dr.Field<DateTime>("SAMPLE_TIME"),

                  waitClass = dr.Field<string>("WAIT_CLASS")

             } into g

             select new

             {

                  SampleTime = g.Key.sampleTime,

                  WaitClass = g.Key.waitClass,

                  Count = g.Count()

             });

or fluent syntax:

var result = dtTAConvert.AsEnumerable()

                .GroupBy(dr => new

                {

                    sampleTime = dr.Field<DateTime>("SAMPLE_TIME"),

                    waitClass = dr.Field<string>("WAIT_CLASS")

                }).Select(g => new

                {

                    SampleTime = g.Key.sampleTime,

                    WaitClass = g.Key.waitClass,

                    Count = g.Count()

                });