Flowthru.Extensions.Parquet
0.26.0-preview.112
dotnet add package Flowthru.Extensions.Parquet --version 0.26.0-preview.112
NuGet\Install-Package Flowthru.Extensions.Parquet -Version 0.26.0-preview.112
<PackageReference Include="Flowthru.Extensions.Parquet" Version="0.26.0-preview.112" />
<PackageVersion Include="Flowthru.Extensions.Parquet" Version="0.26.0-preview.112" />
<PackageReference Include="Flowthru.Extensions.Parquet" />
paket add Flowthru.Extensions.Parquet --version 0.26.0-preview.112
#r "nuget: Flowthru.Extensions.Parquet, 0.26.0-preview.112"
#:package Flowthru.Extensions.Parquet@0.26.0-preview.112
#addin nuget:?package=Flowthru.Extensions.Parquet&version=0.26.0-preview.112&prerelease
#tool nuget:?package=Flowthru.Extensions.Parquet&version=0.26.0-preview.112&prerelease
Flowthru.Extensions.Parquet
Read and write Flowthru Catalog Items as Parquet files. Adds the Parquet format to the
Catalog builder, so any Item backed by a row sequence serializes to and from .parquet with a
one-line declaration. Parquet is the format of choice for intermediate and primary data: it's
binary, columnar, and compressed, so it round-trips large typed row sets far faster and smaller
than CSV.
Mental model
Storage in Flowthru is three independent axes: format (how bytes serialize) × medium
(where bytes live) × container (the in-memory shape). This package supplies one format —
Parquet. Bring the columnar mental model: instead of one record per line, values are grouped by
column into compressed row groups, so reads can skip columns and writes pack tightly. You don't
manage any of that — you declare the Item, and Flowthru maps your schema to and from the columnar
layout. It doesn't care where the .parquet lives (that's the medium's job).
Install
dotnet add package Flowthru.Extensions.Parquet
Declare a Parquet-backed Item in your Catalog:
public IItem<IEnumerable<PreprocessedShuttleSchema>> PreprocessedShuttles =>
CreateItem(() => Item.Of<IEnumerable<PreprocessedShuttleSchema>>("PreprocessedShuttles")
.Parquet()
.AtPath($"{_basePath}/Data/_02_Intermediate/preprocessed_shuttles.parquet")
.Build());
Defaults are Snappy compression, 1,000,000-row groups, and dictionary encoding. To tune them,
pass .WithOptions(...).
| Product | Versions Compatible and additional computed target framework versions. |
|---|---|
| .NET | net10.0 is compatible. net10.0-android was computed. net10.0-browser was computed. net10.0-ios was computed. net10.0-maccatalyst was computed. net10.0-macos was computed. net10.0-tvos was computed. net10.0-windows was computed. |
-
net10.0
- Flowthru.Core (>= 0.26.0-preview.112)
- Microsoft.Extensions.Configuration (>= 10.0.4)
- Microsoft.Extensions.Configuration.Abstractions (>= 10.0.4)
- Microsoft.Extensions.Configuration.Binder (>= 10.0.4)
- Microsoft.Extensions.DependencyInjection.Abstractions (>= 10.0.4)
- Microsoft.Extensions.Logging.Abstractions (>= 10.0.4)
- Microsoft.Extensions.Options (>= 10.0.4)
- Parquet.Net (>= 5.6.0)
- Snappier (>= 1.3.1)
NuGet packages (1)
Showing the top 1 NuGet packages that depend on Flowthru.Extensions.Parquet:
| Package | Downloads |
|---|---|
|
Flowthru
A type-safe data engineering framework for .NET. Includes CSV, Parquet, and Excel format support. |
GitHub repositories
This package is not used by any popular GitHub repositories.
| Version | Downloads | Last Updated |
|---|---|---|
| 0.26.0-preview.112 | 35 | 6/5/2026 |
| 0.25.0 | 96 | 6/2/2026 |
| 0.25.0-preview.110 | 42 | 6/2/2026 |
| 0.24.0-preview.108 | 39 | 6/2/2026 |
| 0.21.0 | 104 | 5/24/2026 |
| 0.21.0-preview.101 | 58 | 5/24/2026 |
| 0.20.0 | 104 | 5/23/2026 |
| 0.20.0-preview.100 | 54 | 5/23/2026 |
| 0.19.0-preview.99 | 66 | 5/23/2026 |
| 0.18.5-preview.98 | 69 | 5/22/2026 |
| 0.18.4-preview.97 | 51 | 5/22/2026 |
| 0.18.3 | 105 | 5/20/2026 |
| 0.18.3-preview.95 | 60 | 5/19/2026 |
| 0.18.2 | 101 | 5/18/2026 |
| 0.18.2-preview.93 | 57 | 5/18/2026 |
| 0.18.1 | 110 | 5/15/2026 |
| 0.18.1-preview.92 | 54 | 5/14/2026 |
| 0.18.0-preview.91 | 49 | 5/14/2026 |
| 0.17.5 | 106 | 5/13/2026 |