Implementation:FlowiseAI Flowise GetFileChunks

Attribute	Value
Sources	packages/ui/src/api/documentstore.js:L13-L16
Domains	Document_Store_Ingestion
Last Updated	2026-02-12 14:00 GMT

Overview

GetFileChunks encompasses the API client functions for retrieving and editing stored document chunks. The getFileChunks function retrieves paginated chunks from the server, while editChunkFromStore updates individual chunk content and metadata.

Code Reference

Source Location

• File: packages/ui/src/api/documentstore.js, Lines 13-16
• Repository: FlowiseAI/Flowise

Signature

const getFileChunks = (storeId, fileId, pageNo) => client.get(`/document-store/chunks/${storeId}/${fileId}/${pageNo}`)

const editChunkFromStore = (storeId, loaderId, chunkId, body) =>
    client.put(`/document-store/chunks/${storeId}/${loaderId}/${chunkId}`, body)

• getFileChunks sends a GET request to /api/v1/document-store/chunks/{storeId}/{fileId}/{pageNo}.
• editChunkFromStore sends a PUT request to /api/v1/document-store/chunks/{storeId}/{loaderId}/{chunkId}.

Import

import documentStoreApi from '@/api/documentstore'

The functions are accessed as documentStoreApi.getFileChunks(storeId, fileId, pageNo) and documentStoreApi.editChunkFromStore(storeId, loaderId, chunkId, body) after importing the default export.

I/O Contract

Inputs for getFileChunks

Parameter	Type	Required	Description
storeId	string	Yes	The document store identifier
fileId	string	Yes	The file/loader identifier, or "all" to retrieve chunks from all files
pageNo	number	Yes	The page number (1-indexed). Each page contains up to 50 chunks

Outputs for getFileChunks

Field	Type	Description
data	object	The paginated chunk response
data.count	number	Total number of chunks matching the query
data.chunks	Chunk[]	Array of chunk objects for the current page
data.chunks[].pageContent	string	The text content of the chunk
data.chunks[].metadata	object	Metadata associated with the chunk (source, page, etc.)
data.currentPage	number	The current page number

The function returns a Promise<{data: {count: number, chunks: Chunk[], currentPage: number}}>.

Inputs for editChunkFromStore

Parameter	Type	Required	Description
storeId	string	Yes	The document store identifier
loaderId	string	Yes	The loader identifier that owns this chunk
chunkId	string	Yes	The unique chunk identifier
body	object	Yes	The updated chunk data
body.pageContent	string	Yes	The new text content for the chunk
body.metadata	object	Yes	The new metadata object for the chunk

Outputs for editChunkFromStore

Field	Type	Description
data	Chunk	The updated chunk object
data.pageContent	string	The updated text content
data.metadata	object	The updated metadata

The function returns a Promise<{data: Chunk}>.

Usage Examples

Paginated Chunk Browsing

import documentStoreApi from '@/api/documentstore'

const browseAllChunks = async (storeId) => {
    let page = 1
    let totalChunks = 0

    // Fetch the first page
    const response = await documentStoreApi.getFileChunks(storeId, 'all', page)
    totalChunks = response.data.count
    const totalPages = Math.ceil(totalChunks / 50)

    console.log(`Total chunks: ${totalChunks}, Pages: ${totalPages}`)
    console.log(`Page ${page}:`, response.data.chunks.length, 'chunks')

    // Display chunk summaries
    response.data.chunks.forEach((chunk, i) => {
        console.log(`  [${i + 1}] ${chunk.pageContent.substring(0, 80)}...`)
    })

    return response.data
}

Fetching Chunks for a Specific File

import documentStoreApi from '@/api/documentstore'

const getChunksForFile = async (storeId, fileId) => {
    const response = await documentStoreApi.getFileChunks(storeId, fileId, 1)
    console.log(`File has ${response.data.count} chunks`)
    return response.data
}

Editing a Chunk

import documentStoreApi from '@/api/documentstore'

const fixChunkContent = async (storeId, loaderId, chunkId) => {
    const updatedChunk = await documentStoreApi.editChunkFromStore(
        storeId,
        loaderId,
        chunkId,
        {
            pageContent: 'Corrected text content with fixed OCR errors',
            metadata: {
                source: 'product-docs.pdf',
                page: 5,
                editedBy: 'manual-review',
                editedAt: new Date().toISOString()
            }
        }
    )

    console.log('Chunk updated:', updatedChunk.data.pageContent.substring(0, 50))
    return updatedChunk.data
}

Bulk Review Workflow

import documentStoreApi from '@/api/documentstore'

const reviewChunksPage = async (storeId, pageNo) => {
    const response = await documentStoreApi.getFileChunks(storeId, 'all', pageNo)
    const { chunks, count, currentPage } = response.data

    // Flag chunks that may need attention
    const shortChunks = chunks.filter(c => c.pageContent.length < 50)
    const noMetadata = chunks.filter(c => Object.keys(c.metadata).length === 0)

    console.log(`Page ${currentPage}: ${chunks.length} chunks`)
    console.log(`  Short chunks (< 50 chars): ${shortChunks.length}`)
    console.log(`  Missing metadata: ${noMetadata.length}`)

    return { chunks, shortChunks, noMetadata }
}

Related Pages

• Principle:FlowiseAI_Flowise_Chunk_Management
• Implementation:FlowiseAI_Flowise_SaveProcessingLoader -- Processing documents to produce chunks
• Implementation:FlowiseAI_Flowise_InsertIntoVectorStore -- Upserting reviewed chunks to vector stores
• Implementation:FlowiseAI_Flowise_PreviewChunks -- Previewing chunks before full processing