EnumerableDataReaderAdapter

A lightweight .NET library that converts IEnumerable<T> into an IDataReader, enabling streaming of in-memory collections into APIs that require IDataReader -- most notably SqlBulkCopy for high-performance bulk inserts into SQL Server.

Features

• Streaming -- rows are read lazily from the enumerable; the entire collection is never buffered in memory.
• Automatic property mapping -- public properties are discovered automatically when no explicit mapping is provided.
• Fluent column mapping API -- choose exactly which columns to expose using expression-based or delegate-based mappings.
• Computed columns -- map constant values or derived expressions that don't correspond to a property.
• Multi-target -- supports .NET 8.0, .NET 9.0, and .NET 10.0.

Installation

Add a project reference or include the source in your solution. The library has no external runtime dependencies.

Usage

Basic usage with SqlBulkCopy

var data = Enumerable.Range(1, 10_000)
    .Select(x => new { Id = x, Name = $"name-{x}" });

using var reader = data.ToDataReader(map => map
    .Add(x => x.Id)
    .Add(x => x.Name));

using var bulkCopy = new SqlBulkCopy(connectionString);
bulkCopy.DestinationTableName = "dbo.People";
bulkCopy.EnableStreaming = true;
bulkCopy.ColumnMappings.Add("Id", "Id");
bulkCopy.ColumnMappings.Add("Name", "Name");

await bulkCopy.WriteToServerAsync(reader);

Default mapping (auto-discover all public properties)

If no mapping is configured, every public instance property on T is exposed as a column:

var reader = products.ToDataReader();

Expression-based mapping

Use lambda expressions that point to properties. The column name and type are inferred automatically:

var reader = products.ToDataReader(map => map
    .Add(p => p.Id)
    .Add(p => p.Name)
    .Add(p => p.Price));

Delegate-based mapping with explicit name and type

For full control -- including computed/constant columns -- specify the column name, CLR type, and a value delegate:

var reader = orders.ToDataReader(map => map
    .Add("OrderId", typeof(int), o => o.Id)
    .Add("Total",   typeof(decimal), o => o.Quantity * o.UnitPrice)
    .Add("Source",  typeof(string), _ => "Import"));

Mixing mapping styles

The two Add overloads can be freely combined in a single mapping configuration:

var reader = items.ToDataReader(map => map
    .Add(i => i.Id)
    .Add("DisplayName", typeof(string), i => $"{i.FirstName} {i.LastName}")
    .Add(i => i.CreatedAt));

API Reference

EnumerableExtensions

Method	Description
ToDataReader<T>(this IEnumerable<T>, Action<ColumnMappings<T>>?)	Creates an IDataReader with optional mapping configuration. When no columns are configured, all public properties of T are used.
ToDataReader<T>(this IEnumerable<T>, ColumnMappings<T>)	Creates an IDataReader using a pre-built ColumnMappings<T> instance.

ColumnMappings<T>

Method	Description
Add(Expression<Func<T, object?>>)	Adds a column from a property expression. Column name and type are inferred from the member.
Add(string, Type, Func<T, object?>)	Adds a column with an explicit name, CLR type, and value delegate.

Both Add methods return this, so calls can be chained fluently.

Building

dotnet build

Running tests

dotnet test

Benchmarks

dotnet run --project benchmarks/EnumerableDataReaderAdapter.Benchmarks -c Release

License

MIT -- Copyright (c) 2018 Dimo Terziev