Data Sources as Configuration

Where data comes from is one of the most fundamental questions in any application. The answer determines what the user sees, whether it's accurate, and how quickly it arrives. In traditional development, answering this question means writing code: queries, HTTP clients, data transformers, caching logic. It's specialized work that blocks everything downstream.

In Swifty, data sources are configuration. You define where data comes from, and components consume it without any code.

What a Data Source Definition Contains

A data source definition specifies a few key things:

The source type: Is this your application's object database, an external API integration, a static list, a monitoring service, or a composed combination of multiple sources?

The connection: Which database object type, which integration, which endpoint — the specific resource to query.

Filters and parameters: How to narrow the data — by status, by date range, by relation to another record. Parameters that come from page context (like a record ID from the URL) can be referenced directly.

Sort and pagination: Default ordering and how many records to load at once.

That's the definition. No query language, no API client code, no data transformation logic.

Configuration in the Builder

Data source definitions are configured in the builder's data panel. Select a data source type, fill in the connection details, add filters using a visual filter builder, and save. The component connected to this source immediately starts showing data.

Adjusting a data source — changing a filter, adding a sort field, switching the source type — is as fast as editing any other configuration. The change applies immediately and can be reverted if needed.

The Same Interface for Any Source

The data source configuration interface is the same regardless of the source type. Whether you're configuring a database query or an external API connection, the same builder panel applies the same conceptual operations: filter, sort, paginate. The visual representation abstracts away the differences in how each source type executes those operations.

This means learning to configure one type of data source transfers directly to any other. The skill generalizes.

Under the Hood

The platform translates data source configurations into the appropriate operations for each source type at runtime. A database source becomes an optimized query. An API source becomes an authenticated HTTP request with parameters. A monitoring source queries the relevant metrics endpoint. The configuration is source-agnostic; the execution is source-specific.

This translation layer is what makes the unified interface possible. Configuration stays simple; the platform handles the complexity.