Method 1 : Using setDT() method The setDT() method can be used to coerce the dataframe or the lists into data. table, where the conversion is made to the original dataframe.
Here's a nice 2013 update using FastMember from NuGet:
IEnumerable<SomeType> data = ...
DataTable table = new DataTable();
using(var reader = ObjectReader.Create(data)) {
table.Load(reader);
}
This uses FastMember's meta-programming API for maximum performance. If you want to restrict it to particular members (or enforce the order), then you can do that too:
IEnumerable<SomeType> data = ...
DataTable table = new DataTable();
using(var reader = ObjectReader.Create(data, "Id", "Name", "Description")) {
table.Load(reader);
}
Editor's Dis/claimer: FastMember is a Marc Gravell project. It's gold and full-on flies!
Yes, this is pretty much the exact opposite of this one; reflection would suffice - or if you need quicker, HyperDescriptor
in 2.0, or maybe Expression
in 3.5. Actually, HyperDescriptor
should be more than adequate.
For example:
// remove "this" if not on C# 3.0 / .NET 3.5
public static DataTable ToDataTable<T>(this IList<T> data)
{
PropertyDescriptorCollection props =
TypeDescriptor.GetProperties(typeof(T));
DataTable table = new DataTable();
for(int i = 0 ; i < props.Count ; i++)
{
PropertyDescriptor prop = props[i];
table.Columns.Add(prop.Name, prop.PropertyType);
}
object[] values = new object[props.Count];
foreach (T item in data)
{
for (int i = 0; i < values.Length; i++)
{
values[i] = props[i].GetValue(item);
}
table.Rows.Add(values);
}
return table;
}
Now with one line you can make this many many times faster than reflection (by enabling HyperDescriptor
for the object-type T
).
Edit re performance query; here's a test rig with results:
Vanilla 27179
Hyper 6997
I suspect that the bottleneck has shifted from member-access to DataTable
performance... I doubt you'll improve much on that...
Code:
using System;
using System.Collections.Generic;
using System.ComponentModel;
using System.Data;
using System.Diagnostics;
public class MyData
{
public int A { get; set; }
public string B { get; set; }
public DateTime C { get; set; }
public decimal D { get; set; }
public string E { get; set; }
public int F { get; set; }
}
static class Program
{
static void RunTest(List<MyData> data, string caption)
{
GC.Collect(GC.MaxGeneration, GCCollectionMode.Forced);
GC.WaitForPendingFinalizers();
GC.WaitForFullGCComplete();
Stopwatch watch = Stopwatch.StartNew();
for (int i = 0; i < 500; i++)
{
data.ToDataTable();
}
watch.Stop();
Console.WriteLine(caption + "\t" + watch.ElapsedMilliseconds);
}
static void Main()
{
List<MyData> foos = new List<MyData>();
for (int i = 0 ; i < 5000 ; i++ ){
foos.Add(new MyData
{ // just gibberish...
A = i,
B = i.ToString(),
C = DateTime.Now.AddSeconds(i),
D = i,
E = "hello",
F = i * 2
});
}
RunTest(foos, "Vanilla");
Hyper.ComponentModel.HyperTypeDescriptionProvider.Add(
typeof(MyData));
RunTest(foos, "Hyper");
Console.ReadLine(); // return to exit
}
}
I had to modify Marc Gravell's sample code to handle nullable types and null values. I have included a working version below. Thanks Marc.
public static DataTable ToDataTable<T>(this IList<T> data)
{
PropertyDescriptorCollection properties =
TypeDescriptor.GetProperties(typeof(T));
DataTable table = new DataTable();
foreach (PropertyDescriptor prop in properties)
table.Columns.Add(prop.Name, Nullable.GetUnderlyingType(prop.PropertyType) ?? prop.PropertyType);
foreach (T item in data)
{
DataRow row = table.NewRow();
foreach (PropertyDescriptor prop in properties)
row[prop.Name] = prop.GetValue(item) ?? DBNull.Value;
table.Rows.Add(row);
}
return table;
}
This is a simple mix of the solutions. It work with Nullable types.
public static DataTable ToDataTable<T>(this IList<T> list)
{
PropertyDescriptorCollection props = TypeDescriptor.GetProperties(typeof(T));
DataTable table = new DataTable();
for (int i = 0; i < props.Count; i++)
{
PropertyDescriptor prop = props[i];
table.Columns.Add(prop.Name, Nullable.GetUnderlyingType(prop.PropertyType) ?? prop.PropertyType);
}
object[] values = new object[props.Count];
foreach (T item in list)
{
for (int i = 0; i < values.Length; i++)
values[i] = props[i].GetValue(item) ?? DBNull.Value;
table.Rows.Add(values);
}
return table;
}
A small change to Marc's answer to make it work with value types like List<string>
to data table:
public static DataTable ListToDataTable<T>(IList<T> data)
{
DataTable table = new DataTable();
//special handling for value types and string
if (typeof(T).IsValueType || typeof(T).Equals(typeof(string)))
{
DataColumn dc = new DataColumn("Value", typeof(T));
table.Columns.Add(dc);
foreach (T item in data)
{
DataRow dr = table.NewRow();
dr[0] = item;
table.Rows.Add(dr);
}
}
else
{
PropertyDescriptorCollection properties = TypeDescriptor.GetProperties(typeof(T));
foreach (PropertyDescriptor prop in properties)
{
table.Columns.Add(prop.Name, Nullable.GetUnderlyingType(prop.PropertyType) ?? prop.PropertyType);
}
foreach (T item in data)
{
DataRow row = table.NewRow();
foreach (PropertyDescriptor prop in properties)
{
try
{
row[prop.Name] = prop.GetValue(item) ?? DBNull.Value;
}
catch (Exception ex)
{
row[prop.Name] = DBNull.Value;
}
}
table.Rows.Add(row);
}
}
return table;
}
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With