Member Junction
    Preparing search index...

    Options controlling scope of IntrospectSchema.

    interface IntrospectSchemaOptions {
        ObjectNames?: string[];
        OnObject?: (obj: SourceObjectInfo) => Promise<void>;
        OnProgress?: (scanned: number, total: number) => void;
    }
    Index

    Properties

    ObjectNames?: string[]

    When set, restricts introspection (and any per-object describe calls) to this subset of object names. When omitted/empty, describes all objects the connector can discover — which for large systems (e.g. Salesforce with ~1,800 sobjects) is expensive. Prefer passing the user-selected subset whenever possible.

    OnObject?: (obj: SourceObjectInfo) => Promise<void>

    Optional streaming handler. When supplied, the base implementations hand each object to it as soon as that object is built, AWAIT it, and then drop the object instead of adding it to SourceSchemaInfo.Objects — so a caller that persists inside the handler holds one object (per describe in flight) instead of the whole catalog. The result carries Streamed: true.

    Opt-in on purpose. Without a handler nothing changes, and a connector override that never forwards this option simply returns a populated schema with Streamed falsy — overrides in other repositories keep working untouched.

    OnProgress?: (scanned: number, total: number) => void

    U11 — optional live progress callback: invoked after each object's describe completes (or is skipped) with (scanned, total). Lets a caller drive a DETERMINATE discovery bar ("scanned N of M objects") instead of an indeterminate spinner. Best-effort — the default IntrospectSchema invokes it; connector overrides may or may not.