@karmaniverous/jeeves-watcher
    Preparing search index...

    Class DocumentProcessor

    Core document processing pipeline.

    Handles extracting text, computing embeddings, and syncing with the vector store.

    Implements

    Index

    Constructors

    Methods

    • Process a metadata update: merge metadata, write to disk, update Qdrant payloads (no re-embed).

      Parameters

      • filePath: string

        The file whose metadata to update.

      • metadata: Record<string, unknown>

        The new metadata to merge.

      Returns Promise<Record<string, unknown> | null>

      The merged payload, or null if the file is not indexed.

    • Re-apply inference rules to a file without re-embedding. Reads file attributes, applies current rules, merges with enrichment metadata, and updates Qdrant payloads.

      Parameters

      • filePath: string

        The file to update.

      Returns Promise<Record<string, unknown> | null>

      The merged metadata, or null if the file is not indexed.