Fluent.CSV.Machine 1.0.2

.NET 7.0
dotnet add package Fluent.CSV.Machine --version 1.0.2
NuGet\Install-Package Fluent.CSV.Machine -Version 1.0.2
This command is intended to be used within the Package Manager Console in Visual Studio, as it uses the NuGet module's version of Install-Package.
<PackageReference Include="Fluent.CSV.Machine" Version="1.0.2" />
For projects that support PackageReference, copy this XML node into the project file to reference the package.
paket add Fluent.CSV.Machine --version 1.0.2
#r "nuget: Fluent.CSV.Machine, 1.0.2"
#r directive can be used in F# Interactive and Polyglot Notebooks. Copy this into the interactive tool or source code of the script to reference the package.
// Install Fluent.CSV.Machine as a Cake Addin
#addin nuget:?package=Fluent.CSV.Machine&version=1.0.2

// Install Fluent.CSV.Machine as a Cake Tool
#tool nuget:?package=Fluent.CSV.Machine&version=1.0.2

example workflow Nuget (with prereleases) GitHub .NET Version

Fluent CSV Machine


  • Reads and parses each character only once.
    → results in a blazing fast execution while having a low memory footprint
    <sub>(does not follow the usual pattern: read the line, split it, parse the fields)</sub>
  • Supports all CSV variants, test cases are implemented against those
  • Types are parsed directly. Nullable is also supported on all types except Enums
    • Simple types (int, long, double, decimal, ...)
    • string
    • DateTime → requires an specific InputFormat
    • Enums
  • Fluent interfaces to define the mapping between the Entity class and the CSV file
  • Parsing and entity creation is handled by different threads

Getting started

Please take a look at the documentation as well at those implemented test cases.
The test cases implement a variety of different CSV fixtures (which are mainly forked from from csv-parser)

// CSV file content:
// a,b,c
// 1,2,2012/11/25
// 3,4,2022/12/04

var parser = new CsvParser<EntityClass>();
parser.Property<string?>(c => c.A).ColumnName("a");
parser.Property<int>(c => c.B).ColumnName("b");
parser.Property<DateTime>(c => c.C).ColumnName("c").InputFormat("yyyy/MM/dd");
IReadOnlyList<EntityClass> result = await parser.Parse(path);

// Values are parsed according to their type definition in EntityClass

Have a look at this awesome tool which generates Entity classes. This tool belongs to the popular library CSVHelper.


Parsing 13 columns, the CSV file contains 14.
Data is read from a MemoryStream and returned as List of Entities with the default parser configuration.
Each Entity contains 7 string, 2 int, 2 double, 1 DateTime and 1 Enum Property.
The benchmark ranges from 1 thousand to 1 million CSV lines / entities.

BenchmarkDotNet=v0.13.2, OS=Windows 11 (10.0.22000.1098/21H2)
AMD Ryzen 5 3600, 1 CPU, 12 logical and 6 physical cores
.NET SDK=7.0.100

InvocationCount=1  IterationCount=10  LaunchCount=10
RunStrategy=Monitoring  UnrollFactor=1  WarmupCount=0

|        Lines |         Mean |      Error |      StdDev |       Median |
|------------- |-------------:|-----------:|------------:|-------------:|
|        1,000 |     6.789 ms |  0.1206 ms |   0.3556 ms |     6.667 ms |
|       10,000 |    34.383 ms |  4.3588 ms |  12.8519 ms |    44.196 ms |
|      100,000 |   240.696 ms |  1.4216 ms |   4.1915 ms |   239.989 ms |
|    1,000,000 | 2,742.841 ms | 47.3035 ms | 139.4755 ms | 2,815.532 ms |

Here you can compare those values roughly. Though different libraries have different purposes while all parse CSV.


This started as a CSV library for my personal private projects. My thought back then was the following: Do not test a dozen of libraries, just write one of your one. Since then it has been rewritten a few times. Mostly to show off that I can still write effient code while my occupation doesn't include any programming anymore. Finally I tried to make it as fast as possible while still returning a typed result and not just a set of strings.

tl;dr: Lets see how fast a typed dotnet CSV parser can get

Advanced use cases

Custom Properties

If a simple mapping does not work out for you then you can try to use PropertyCustom

parser.PropertyCustom<string>((x, v) =>
    var split = v.Split(' ').Select(c => c.Trim()).ToArray();
    x.ForeignCurrencyValue = decimal.Parse(split[0]);
    x.Currency = EnumHelper.Parse<Currency>(split[1]);
}).ColumnName("Foreign Transaction");

Beware: You are about to execute this Action on each CSV line. An Action which is a lot slower than the in-built parsers.

Line Actions

Defines one or more Actions which run after all properties (normal as well as custom ones) have been mapped.

var parser = new CsvParser<T>();
parser.LineAction((obj, fields) =>
	if (fields == null || obj is not Entity e)
	// Create an hash value using all parsed columns of a CSV line
	e.HashCode = HashCodeLine(fields);

Have a look at the test case BackTick for another example

CSV files without a header line

This CSV parser only works with a backing class which can be mapped. If you do not have a CSV line which defines the headers and thefore the corresponding properties:
Then you need to use CsvNoHeaderAttribute as an attribute to your properties to define the column order.

internal class BasicString
    [CsvNoHeader(columnIndex: 0)] public string? A { get; set; }
    [CsvNoHeader(columnIndex: 1)] public string? B { get; set; }
    public string? C { get; set; } // This column won't be mapped
    [CsvNoHeader(columnIndex: 2)] public string? D { get; set; }
Product Versions
.NET net7.0 net7.0-android net7.0-ios net7.0-maccatalyst net7.0-macos net7.0-tvos net7.0-windows
Compatible target framework(s)
Additional computed target framework(s)
Learn more about Target Frameworks and .NET Standard.
  • net7.0

    • No dependencies.

NuGet packages

This package is not used by any NuGet packages.

GitHub repositories

This package is not used by any popular GitHub repositories.

Version Downloads Last updated
1.0.2 175 12/8/2022
1.0.1 163 12/6/2022
1.0.0 181 12/3/2022
0.1.0-beta 60 11/19/2022