PgBulkOps 0.1.0

																					 PgBulkOps

PgBulkOps is a high-performance .NET library for PostgreSQL bulk operations. It provides ultra-fast bulk insert and bulk update methods using PostgreSQL’s native binary COPY protocol.

Built on top of Npgsql, PgBulkOps can handle millions of rows in just a few seconds.

✨ Features Bulk insert with PostgreSQL binary COPY Bulk update using COPY + temporary table + UPDATE ... FROM join Optional PascalCase → snake_case column name conversion Progress callback with configurable batch size Fully async/await compatible 📦 Installation dotnet add package PgBulkOps

✨ Quick Start

Entity Definition

public class User { public long Id { get; set; } public string Name { get; set; } = default!; public string Email { get; set; } = default!; public int Age { get; set; } public bool IsActive { get; set; } public decimal Balance { get; set; } public DateTime CreatedAt { get; set; } public DateTime? UpdatedAt { get; set; } public string Country { get; set; } = default!; public DateTime? LastLogin { get; set; } }

Bulk Insert Example

using Npgsql; using PgBulkOps;

await using var conn = new NpgsqlConnection("Host=localhost;Database=test;Username=postgres;Password=secret");

var rnd = new Random(); var users = Enumerable.Range(1, 100_000) .Select(i ⇒ new User { Id = i, Name = $"User{i}", Email = $"user{i}@example.com", Age = rnd.Next(18, 70), IsActive = i % 2 == 0, Balance = (decimal)(rnd.NextDouble() * 10000), CreatedAt = DateTime.UtcNow, Country = i % 2 == 0 ? "TR" : "US", LastLogin = DateTime.UtcNow.AddMinutes(-rnd.Next(1, 10000)) });

await conn.BulkInsertAsync(users, "Users", opts ⇒ { opts.BatchSize = 50_000; opts.UseSnakeCase = true; opts.OnProgress = p ⇒ Console.WriteLine($"Inserted {p.Rows} rows..."); });

Bulk Update Example

// Update existing users foreach (var user in users) user.Name = user.Name + "_updated";

await conn.BulkUpdateAsync(users, "Users", "Id", opts ⇒ { opts.BatchSize = 50_000; opts.UseSnakeCase = true; opts.OnProgress = p ⇒ Console.WriteLine($"Updated {p.Rows} rows..."); });

⚡ Performance Typical throughput (depending on hardware, WAL, and indexes):

Bulk Insert → 300k–400k rows/sec

Bulk Update → 30k–80k rows/sec

In benchmarks, inserting 1,000,000 rows often completes in ~3 seconds.

⚠️ Notes PostgreSQL user must have INSERT/UPDATE privileges.

PascalCase properties map automatically to snake_case columns if UseSnakeCase = true.

For massive loads, consider disabling indexes and triggers temporarily.

📄 License MIT License – maintained by Onventus.