Oracle Vector Store icon

Oracle Vector Store

Gerenciamento de vector store usando Oracle Database 23ai com suporte nativo a vetores

Actions7

Add Document

AI-generated

Summary

Adds one or more documents to the Oracle Vector Store collection.

Inputs

  • Input Data — Array of document objects. Each object must contain an embedding vector (as embedding or vector) and may optionally include id, content, and metadata. If id is missing, a timestamp-based ID is generated. The metadata field is merged with system metadata (timestamp, nodeId, workflowId) and stored as JSON.

Output shape

Array of result objects, one per input item. Each result includes: success (boolean), documentId (string), content (truncated to first 100 chars), embeddingDimension (number), rowsAffected (number), operation ("addDocument"), index (input item index). On failure, success is false and an error string is provided instead of the other fields.

The operation uses the Collection Name node parameter (default VECTOR_STORE) to determine the target table. It expects the table to already exist (created via the "Setup Collection" operation). Each insert is committed automatically (autoCommit: true). The node requires valid Oracle credentials.

Links

Discussion