Designing a Connector with Multiple Event Types and Independent Crawlers

Is your feature request related to a problem? Please describe.

I am currently working on a connector that needs to support multiple event types. For example:

Type A → Logs collected from one API endpoint

Type B → Logs collected from a different API endpoint

In Splunk and other observability platforms, it is common to define separate modular inputs for each event type. Each event type has its own independent data-collection configuration, and collecting one event type does not impact the others. Despite this separation, all event-type collectors can be bundled into a single add-on that is installed via the marketplace.

While reviewing the Data Prepper source plugin architecture, I noticed that most SaaS source plugins appear to expose only a single endpoint. This leads to my main question:

Can a single Data Prepper connector/plugin support multiple event types or multiple endpoints, or is it restricted to a single endpoint?
My assumption is that since each pipeline definition supports only one source, a connector is effectively limited to a single data-collection endpoint. Please correct me if this understanding is incorrect.

Additionally, if it is possible for one plugin/connector to manage multiple event types, I am looking for guidance regarding the crawler implementation. Different event types may involve distinct API behaviors, schemas, and pagination models, which could require separate crawler implementations. All example plugins in the Data Prepper codebase appear to use only a single crawler per plugin, so it is unclear how (or whether) multiple crawlers are supported within one plugin.

Review the Data Prepper source plugin architecture, pipeline definition behavior, and example plugins that use one crawler. Determine whether one connector can support multiple event types or endpoints with independent crawlers; done means maintainer-confirmed guidance or a documented design with the required scope identified.