Function handleVectorOnlySearch

apps/sim/app/api/knowledge/search/utils.ts:454–513

(params: SearchParams)

}

export async function handleVectorOnlySearch(params: SearchParams): Promise<SearchResult[]> {
  const { knowledgeBaseIds, topK, queryVector, distanceThreshold } = params

  if (!queryVector || !distanceThreshold) {
    throw new Error('Query vector and distance threshold are required for vector-only search')
  }

  const strategy = getQueryStrategy(knowledgeBaseIds.length, topK)

  const distanceExpr = sql<number>`${embedding.embedding} <=> ${queryVector}::vector`.as('distance')

  if (strategy.useParallel) {
    // Parallel approach for many KBs
    const parallelLimit = Math.ceil(topK / knowledgeBaseIds.length) + 5

    const queryPromises = knowledgeBaseIds.map(async (kbId) => {
      return await db
        .select(getSearchResultFields(distanceExpr))
        .from(embedding)
        .innerJoin(document, eq(embedding.documentId, document.id))
        .where(
          and(
            eq(embedding.knowledgeBaseId, kbId),
            eq(embedding.enabled, true),
            eq(document.enabled, true),
            eq(document.processingStatus, 'completed'),
            eq(document.userExcluded, false),
            isNull(document.archivedAt),
            isNull(document.deletedAt),
            sql`${embedding.embedding} <=> ${queryVector}::vector < ${distanceThreshold}`
          )
        )
        .orderBy(sql`${embedding.embedding} <=> ${queryVector}::vector`)
        .limit(parallelLimit)
    })

    const parallelResults = await Promise.all(queryPromises)
    const allResults = parallelResults.flat()
    return allResults.sort((a, b) => a.distance - b.distance).slice(0, topK)
  }
  // Single query for fewer KBs
  return await db
    .select(getSearchResultFields(distanceExpr))
    .from(embedding)
    .innerJoin(document, eq(embedding.documentId, document.id))
    .where(
      and(
        inArray(embedding.knowledgeBaseId, knowledgeBaseIds),
        eq(embedding.enabled, true),
        eq(document.enabled, true),
        eq(document.processingStatus, 'completed'),
        eq(document.userExcluded, false),
        isNull(document.archivedAt),
        isNull(document.deletedAt),
        sql`${embedding.embedding} <=> ${queryVector}::vector < ${distanceThreshold}`
      )
    )
    .orderBy(sql`${embedding.embedding} <=> ${queryVector}::vector`)