Implementation:FlowiseAI Flowise SaveProcessingLoader

Attribute	Value
Sources	packages/ui/src/api/documentstore.js:L18-L19
Domains	Document_Store_Ingestion
Last Updated	2026-02-12 14:00 GMT

Overview

SaveProcessingLoader encompasses two API client functions that implement the two-phase document processing pattern: saveProcessingLoader persists the loader configuration, and processLoader triggers asynchronous background processing of the configured document source.

Code Reference

Source Location

• File: packages/ui/src/api/documentstore.js, Lines 18-19
• Repository: FlowiseAI/Flowise

Signature

const saveProcessingLoader = (body) => client.post(`/document-store/loader/save`, body)
const processLoader = (body, loaderId) => client.post(`/document-store/loader/process/${loaderId}`, body)

• saveProcessingLoader sends a POST request to /api/v1/document-store/loader/save.
• processLoader sends a POST request to /api/v1/document-store/loader/process/{loaderId}, where the loader ID is interpolated into the URL path.

Import

import documentStoreApi from '@/api/documentstore'

The functions are accessed as documentStoreApi.saveProcessingLoader(body) and documentStoreApi.processLoader(body, loaderId) after importing the default export from the documentstore API module.

I/O Contract

Inputs for saveProcessingLoader

Parameter	Type	Required	Description
body	object	Yes	The complete loader and splitter configuration (same structure as preview)
body.storeId	string	Yes	The document store identifier
body.loaderId	string	Yes	Existing loader ID to update, or empty string for new loaders
body.loaderName	string	Yes	The loader component name (e.g., pdfLoader)
body.loaderConfig	object	Yes	Loader-specific configuration
body.splitterId	string	No	The text splitter component name
body.splitterConfig	object	No	Splitter-specific configuration
body.credential	string	No	Credential identifier for authenticated loaders

Outputs for saveProcessingLoader

Field	Type	Description
data	object	The saved loader reference
data.id	string	The unique identifier of the saved loader configuration

The function returns a Promise<{data: {id: string}}> from the axios response.

Inputs for processLoader

Parameter	Type	Required	Description
body	object	Yes	Processing parameters
body.storeId	string	Yes	The document store identifier
loaderId	string	Yes	The loader identifier (from saveProcessingLoader response)

Outputs for processLoader

Field	Type	Description
(void)	--	The function returns Promise<void>; processing runs asynchronously in the background

Usage Examples

Complete Two-Phase Processing Workflow

import documentStoreApi from '@/api/documentstore'

const processDocument = async (storeId, loaderConfig) => {
    // Phase 1: Save the configuration
    const saveBody = {
        storeId,
        loaderId: '',
        loaderName: 'pdfLoader',
        loaderConfig,
        splitterId: 'recursiveCharacterTextSplitter',
        splitterConfig: {
            chunkSize: 1000,
            chunkOverlap: 200
        }
    }

    const saveResponse = await documentStoreApi.saveProcessingLoader(saveBody)
    const loaderId = saveResponse.data.id
    console.log('Configuration saved with loader ID:', loaderId)

    // Phase 2: Trigger asynchronous processing
    await documentStoreApi.processLoader({ storeId }, loaderId)
    console.log('Processing triggered for loader:', loaderId)

    return loaderId
}

Re-processing an Existing Loader

import documentStoreApi from '@/api/documentstore'

const reprocessLoader = async (storeId, existingLoaderId) => {
    // Re-trigger processing for an already-saved loader
    await documentStoreApi.processLoader({ storeId }, existingLoaderId)
    console.log('Re-processing triggered for loader:', existingLoaderId)
}

Processing with Credentials

import documentStoreApi from '@/api/documentstore'

const processConfluenceSource = async (storeId, credentialId) => {
    const saveBody = {
        storeId,
        loaderId: '',
        loaderName: 'confluenceLoader',
        loaderConfig: {
            baseUrl: 'https://company.atlassian.net/wiki',
            spaceKey: 'DOCS'
        },
        credential: credentialId,
        splitterId: 'recursiveCharacterTextSplitter',
        splitterConfig: { chunkSize: 1500, chunkOverlap: 300 }
    }

    const saveResponse = await documentStoreApi.saveProcessingLoader(saveBody)
    await documentStoreApi.processLoader({ storeId }, saveResponse.data.id)

    return saveResponse.data.id
}

Related Pages

• Principle:FlowiseAI_Flowise_Document_Processing
• Implementation:FlowiseAI_Flowise_PreviewChunks -- Previewing chunks before committing to processing
• Implementation:FlowiseAI_Flowise_GetFileChunks -- Retrieving processed chunks after completion
• Implementation:FlowiseAI_Flowise_InsertIntoVectorStore -- Upserting processed chunks to vector stores