Skip to content

Latest commit

 

History

History
368 lines (301 loc) · 9.83 KB

README.md

File metadata and controls

368 lines (301 loc) · 9.83 KB

Umbraco.Community.PerformanceTestDataSeeder

A configurable dummy data seeder for Umbraco CMS v13+ designed for performance and load testing.

Features

  • Seeds languages, dictionary items, data types, document types, media, content, and users
  • Nested blocks support - configurable depth of blocks within blocks for realistic load testing
  • Fully configurable via appsettings.json
  • Reproducible test data with configurable Faker seed
  • Enable/disable individual seeders
  • Idempotent - safe to run multiple times
  • Structured logging with progress reporting
  • Auto-registration via Umbraco IComposer

Installation

dotnet add package Umbraco.Community.PerformanceTestDataSeeder

That's it! The package automatically registers with Umbraco and runs on startup.

Local Development

To test the package locally before publishing:

  1. Add a project reference to your Umbraco project's .csproj:

    <ProjectReference Include="..\path\to\Umbraco.Community.PerformanceTestDataSeeder.csproj" />

  2. Add configuration to appsettings.json:

    {
  "PerformanceTestDataSeeder": {
    "Options": {
      "Enabled": true,
      "Preset": "Small",
      "StopOnError": true
    }
  }
}

  3. Run the Umbraco project:

    dotnet run

  4. Check the status endpoint to verify seeding completed:

    GET /umbraco/api/seederstatus/status

Building & Packaging

To create a NuGet package for distribution:

# Build the package
dotnet pack -c Release

# The .nupkg file will be in bin/Release/

To publish to NuGet.org:

# Push to NuGet (requires API key)
dotnet nuget push bin/Release/Umbraco.Community.PerformanceTestDataSeeder.1.0.0.nupkg --api-key YOUR_API_KEY --source https://api.nuget.org/v3/index.json

For local testing with a NuGet package (instead of project reference):

# Create a local NuGet source folder
mkdir ~/local-nugets

# Pack and copy to local source
dotnet pack -c Release -o ~/local-nugets

# Add local source to your test project's NuGet.config or use:
dotnet add package Umbraco.Community.PerformanceTestDataSeeder --source ~/local-nugets

Quick Start with Presets

The easiest way to get started is using a preset. Add this to your appsettings.json:

{
  "PerformanceTestDataSeeder": {
    "Options": {
      "Enabled": true,
      "Preset": "Medium"
    }
  }
}

Available Presets

Preset Languages Content Media Users NestingDepth Total Items
Small 3 50 30 5 2 ~200
Medium 10 500 405 20 4 ~2,000
Large 20 10,000 4,020 50 6 ~25,000
Massive 30 50,000 16,050 100 8 ~100,000
Custom - - - - - Use SeederConfiguration

When using a preset, you don't need to specify the full SeederConfiguration section.

Custom Configuration

For fine-grained control, set Preset to Custom (or omit it) and add configuration to your appsettings.json:

{
  "SeederConfiguration": {
    "Languages": { "Count": 20 },
    "Dictionary": {
      "RootFolders": 10,
      "SectionsPerRoot": 5,
      "ItemsPerSection": 30
    },
    "Users": { "Count": 30 },
    "DataTypes": {
      "ListView": 30,
      "MultiNodeTreePicker": 40,
      "RichTextEditor": 10,
      "MediaPicker": 10,
      "Textarea": 10,
      "Dropdown": 10,
      "Numeric": 10
    },
    "DocumentTypes": {
      "ElementTypes": { "Simple": 65, "Medium": 39, "Complex": 26 },
      "VariantDocTypes": { "Simple": 55, "Medium": 33, "Complex": 22 },
      "InvariantDocTypes": { "Simple": 20, "Medium": 12, "Complex": 8 },
      "BlockList": 40,
      "BlockGrid": 60,
      "NestingDepth": 4
    },
    "Media": {
      "PDF": { "Count": 200, "FolderCount": 10 },
      "PNG": { "Count": 200, "FolderCount": 10 },
      "JPG": { "Count": 200, "FolderCount": 10 },
      "Video": { "Count": 10, "FolderCount": 1 }
    },
    "Content": {
      "TotalTarget": 300,
      "SimplePercent": 50,
      "MediumPercent": 30,
      "ComplexPercent": 20,
      "RootSections": 50,
      "CategoriesPerSection": 10,
      "PagesPerCategory": 100
    }
  },
  "PerformanceTestDataSeeder": {
    "Options": {
      "Preset": "Custom",
      "Enabled": true,
      "StopOnError": false,
      "FakerSeed": 12345,
      "ProgressIntervalPercent": 10,
      "EnabledSeeders": {
        "Languages": true,
        "Dictionary": true,
        "DataTypes": true,
        "DocumentTypes": true,
        "Media": true,
        "Content": true,
        "Users": true
      },
      "Prefixes": {
        "DataType": "Test_",
        "ElementType": "testElement",
        "VariantDocType": "testVariant",
        "InvariantDocType": "testInvariant",
        "Media": "Test_",
        "Content": "Test_",
        "User": "TestUser_",
        "Dictionary": "Dict_"
      }
    }
  }
}

Options

Option Type Default Description
Preset enum Custom Predefined dataset size (Small, Medium, Large, Massive, Custom)
Enabled bool false Master switch - must be explicitly enabled
StopOnError bool false Stop seeding if any seeder fails
FakerSeed int? null Seed for reproducible test data
ProgressIntervalPercent int 10 Log progress every N percent
EnabledSeeders.* bool true Enable/disable individual seeders
Prefixes.* string various Configurable naming prefixes
CustomCultures string[]? null Custom cultures (overrides default 30)

Seeder Execution Order

  1. LanguageSeeder - Creates languages from culture pool
  2. DictionarySeeder - Creates dictionary items with translations
  3. DataTypeSeeder - Creates custom data types
  4. DocumentTypeSeeder - Creates element types, nested container elements, document types, block types, and templates
  5. MediaSeeder - Creates media items (PDFs, images, videos)
  6. ContentSeeder - Creates content hierarchy
  7. UserSeeder - Creates test users

Default Data Quantities

Category Default Count
Languages 20
Dictionary Items 1,500
Data Types 130
Element Types 130
Document Types 110
Templates 110
Block List/Grid Types 100
Media Items 610
Content Nodes 300
Users 30

Nested Blocks

The seeder creates nested block structures to simulate real-world content complexity. The NestingDepth setting controls how deep blocks can be nested.

Example with NestingDepth = 2:

BlockList
  └── NestedBlock_Depth1
        └── nestedBlocks (BlockList)
              └── Simple/Medium/Complex Elements

Example with NestingDepth = 4:

BlockList
  └── NestedBlock_Depth1
        └── nestedBlocks → NestedBlock_Depth2
                            └── nestedBlocks → NestedBlock_Depth3
                                                └── nestedBlocks → Simple/Medium/Complex Elements

This creates realistic load testing scenarios where:

  • JSON serialization/deserialization is stressed
  • Backoffice rendering performance is tested
  • Content API handles deeply nested structures

Reproducible Test Data

Set FakerSeed to a fixed value to generate the same test data across runs:

{
  "PerformanceTestDataSeeder": {
    "Options": {
      "FakerSeed": 12345
    }
  }
}

Disable Specific Seeders

{
  "PerformanceTestDataSeeder": {
    "Options": {
      "EnabledSeeders": {
        "Users": false,
        "Dictionary": false
      }
    }
  }
}

Status Endpoint (for Automation)

Poll the status endpoint to know when seeding is complete:

GET /umbraco/api/seederstatus/status

Response:

{
  "status": "Completed",
  "isComplete": true,
  "currentSeeder": null,
  "executedCount": 7,
  "failedCount": 0,
  "elapsedMs": 45230,
  "errorMessage": null
}

HTTP Status Codes:

  • 200 - Seeding complete (Completed, CompletedWithErrors, or Skipped)
  • 202 - Seeding in progress (Running or NotStarted)
  • 503 - Seeding failed

Automation example (bash):

# Wait for seeding to complete
while true; do
  response=$(curl -s -o /dev/null -w "%{http_code}" http://localhost:5000/umbraco/api/seederstatus/status)
  if [ "$response" = "200" ]; then
    echo "Seeding complete!"
    break
  elif [ "$response" = "503" ]; then
    echo "Seeding failed!"
    exit 1
  fi
  sleep 10
done

Performance Considerations

  • Database: SQL Server is recommended for large datasets. SQLite may timeout with Large/Massive presets.
  • Time estimates:
    • Small: ~30 seconds
    • Medium: ~2-5 minutes
    • Large: ~10-20 minutes
    • Massive: ~30-60+ minutes
  • Memory: Large media generation requires adequate RAM. Consider reducing media counts if memory constrained.
  • Disk space: Media seeder generates actual files. Massive preset creates ~16,000+ media files.

Requirements

  • Umbraco CMS 13+ (13.0.0 and up)
  • .NET 8.0+

Compatibility

This package is designed as a class library that references only Umbraco.Cms.Core and Umbraco.Cms.Infrastructure (not the full Umbraco.Cms metapackage). This means:

  • It does not include any web project dependencies
  • It can be easily added to any existing Umbraco project
  • It will automatically register via the IComposer pattern

Version Support

Package Version Umbraco Version
1.0.x 13.0.0+

Note: While the package allows Umbraco 13+, API compatibility with future major versions (14+, 15+, etc.) depends on Umbraco maintaining backward-compatible APIs. Test against your target version.

Dependencies

  • Bogus (fake data generation)
  • SixLabors.ImageSharp (image generation)
  • Umbraco.Cms.Core (13.x)
  • Umbraco.Cms.Infrastructure (13.x)

License

MIT