Package ai.vectros.types

Class DocumentRequest

java.lang.Object
ai.vectros.types.DocumentRequest
public final class DocumentRequest extends Object

  • Method Details

    • getTitle

      public String getTitle()
      Returns:
      Human-readable document title

    • getText

      public Optional<String> getText()
      Returns:
      Raw text content to ingest. Required when creating a document via the text-ingest path. On update, supply it to replace the stored text; omit it to leave the existing text unchanged.

    • getIndexMode

      public Optional<DocumentRequestIndexMode> getIndexMode()
      Returns:
      Indexing strategy for this document. HYBRID runs both BM25 keyword and dense-vector semantic indexing (recommended for most use cases). SEMANTIC indexes only as dense vectors — best for conceptual similarity search. TEXT indexes only with BM25 — best for exact keyword matching. NONE stores the document without search indexing (store-only / archival): it remains retrievable by id and by structured-field lookup but never appears in search results. Optional: omit to inherit the bound schema's default index mode. If neither this field nor the schema specifies one, the request is rejected. When both are set, this per-document value wins.

    • getFolderId

      public Optional<String> getFolderId()
      Returns:
      ID of the folder to place this document in. On create, omit to use your account's default root folder. On update, omit to leave unchanged — this field cannot currently be cleared once set.

    • getPayload

      public Optional<Map<String,Object>> getPayload()
      Returns:
      The document's structured data, as a flat key/value object. When schemaId is set, declared fields are validated against the schema and its lookup fields become directly queryable via GET /v1/documents?type=&field=&value= (sensitive fields are blind-indexed); undeclared keys pass through as free-form and are searchable via the filters parameter on POST /v1/search. A strict schema rejects undeclared keys. On update, the supplied object replaces the stored payload in full (it is not key-merged); omit to leave it unchanged.

    • getSchemaId

      public Optional<String> getSchemaId()
      Returns:
      Optional ID of a record schema to bind this document to. When set, the document's payload is validated against the schema's fields on save, and the schema's lookup fields become directly queryable (the same behavior as records). On update, omit to leave unchanged.

    • getUserId

      public Optional<String> getUserId()
      Returns:
      Owning user ID — the Vectros-assigned UUID of a user in your account. Optional. With an API key, sets the document's owner explicitly. With a scoped token, must match the token's identity claim (if set) or fall within its data scope.

    • getScopes

      public Optional<List<String>> getScopes()
      Returns:
      The document's scope ownership, as namespace:value entries (at most 2 namespaces) — for example ["org:6ba7b810-9dad-11d1-80b4-00c04fd430c8", "group:eng-team"]. org and client are built-in namespaces; others are custom scopes you define (lowercase, 2-32 chars). Resolve a namespace's UUID from your own identifier with GET /v1/entities/{namespace}?externalId=. When supplied, this is the document's COMPLETE scope declaration: for a token that stamps identity, entries must match the token's identity values, and an empty array creates a document owned by the calling user alone (the private tier). Omit the field to inherit the token's full identity — the default. On update, omit to leave ownership unchanged, or supply the complete new selection ([] clears it). Filter lists by these values with ?scope=.

    • getExternalId

      public Optional<String> getExternalId()
      Returns:
      Stable, caller-supplied identifier for this document. Optional. Immutable after create. Unique within your account and context: posting again with the same externalId returns the existing document (idempotent ingest), and it is the key other records use to reference this one. Max 256 characters.

    • getExpectedVersion

      public Optional<Long> getExpectedVersion()
      Returns:
      Optimistic-concurrency token. Pass the version you last read (from a GET or a prior write response) to make this update conditional — it is rejected with 409 VERSION_CONFLICT if the document was modified since, leaving the stored document untouched. Omit for last-write-wins (the default). Ignored on create.

    • getStatus

      public Optional<DocumentRequestStatus> getStatus()
      Returns:
      Caller-controlled lifecycle status. ACTIVE (the default) keeps the document live and searchable; ARCHIVED soft-retracts it — the document is pulled from search/recall but kept and recoverable (set it back to ACTIVE to re-index and restore). Use this to retire superseded content without deleting it. On update, omit to leave the current lifecycle status unchanged. Distinct from the read-only indexStatus (the processing pipeline).

    • equals

      public boolean equals(Object other)
      Overrides:
      equals in class Object

    • getAdditionalProperties

      public Map<String,Object> getAdditionalProperties()

    • hashCode

      public int hashCode()
      Overrides:
      hashCode in class Object

    • toString

      public String toString()
      Overrides:
      toString in class Object

    • builder

      public static DocumentRequest.TitleStage builder()