DataValidation

DataValidation is a Power Platform ToolBox tool for running simple, repeatable data quality checks against Dataverse tables.

The tool lets you select a table, choose columns, configure tests per column, and run those tests across all records to produce a headline pass rate and per-test results.

Current Capabilities

Table and Column Setup

• Browse Dataverse tables (display name + logical name).
• Load table columns and filter out unsupported/internal types.
• Sort columns by logical name, display name, type, or configured test count.
• Configure tests per column or apply tests in bulk to selected columns.

Implemented Tests

• Contains Data

  • Passes when the value is not null, undefined, or empty/whitespace.

• Matches Regex

  • Uses a configured regular expression against the string form of the value.

• Matches Metadata (currently implemented for selected attribute types)

  • StringType: value length must be <= MaxLength.
  • IntegerType: value must be an integer within [MinValue, MaxValue] (where bounds exist).
  • DecimalType and MoneyType: numeric value must be within [MinValue, MaxValue] (where bounds exist).
  • PicklistType, StateType, StatusType: value must exist in metadata options.

How Test Execution Works

When you click Run Selected Tests, the tool:

1. Builds a FetchXML query including only columns with configured tests.
2. Retrieves all records for the selected table (paged).
3. Executes configured tests against each record value.
4. Displays:
   • A headline pass rate across all executed checks.
   • A per-column, per-test pass/fail breakdown.

Local Development

Prerequisites

• Node.js 18+

Install

npm install

or

pnpm install

Run Dev Mode

npm run dev

Build

npm run build

Validate Package

npm run validate

Usage in Power Platform ToolBox

1. Build the tool (npm run build).
2. Package/install in ToolBox using your normal ToolBox workflow.
3. Connect to a Dataverse environment.
4. Select a table.
5. Configure tests on one or more columns.
6. Run tests and review the summary/results grid.

Notes and Limitations

• Matches Metadata is intentionally incremental and currently covers the types listed above.
• Empty values are treated as metadata-valid so that completeness checks remain the job of Contains Data.
• Very large tables may take longer due to full-table paging and in-memory evaluation.

Tech Stack

• React 18 + TypeScript
• MobX
• Fluent UI React v8
• Vite
• window.dataverseAPI and window.toolboxAPI from Power Platform ToolBox

License

MIT