add fixes to multiple documents

This commit is contained in:
Henry
2023-05-15 20:06:22 +01:00
parent 9f53111aac
commit 324bd3a430
6 changed files with 128 additions and 78 deletions
@@ -1,7 +1,6 @@
import { INode, INodeData, INodeParams } from '../../../src/Interface'
import { TextSplitter } from 'langchain/text_splitter'
import { DocxLoader } from 'langchain/document_loaders/fs/docx'
import { getBlob } from '../../../src/utils'
class Docx_DocumentLoaders implements INode {
label: string
@@ -49,20 +48,35 @@ class Docx_DocumentLoaders implements INode {
const docxFileBase64 = nodeData.inputs?.docxFile as string
const metadata = nodeData.inputs?.metadata
const blob = new Blob(getBlob(docxFileBase64))
const loader = new DocxLoader(blob)
let docs = []
let alldocs = []
let files: string[] = []
if (textSplitter) {
docs = await loader.loadAndSplit(textSplitter)
if (docxFileBase64.startsWith('[') && docxFileBase64.endsWith(']')) {
files = JSON.parse(docxFileBase64)
} else {
docs = await loader.load()
files = [docxFileBase64]
}
for (const file of files) {
const splitDataURI = file.split(',')
splitDataURI.pop()
const bf = Buffer.from(splitDataURI.pop() || '', 'base64')
const blob = new Blob([bf])
const loader = new DocxLoader(blob)
if (textSplitter) {
const docs = await loader.loadAndSplit(textSplitter)
alldocs.push(...docs)
} else {
const docs = await loader.load()
alldocs.push(...docs)
}
}
if (metadata) {
const parsedMetadata = typeof metadata === 'object' ? metadata : JSON.parse(metadata)
let finaldocs = []
for (const doc of docs) {
for (const doc of alldocs) {
const newdoc = {
...doc,
metadata: {
@@ -75,7 +89,7 @@ class Docx_DocumentLoaders implements INode {
return finaldocs
}
return docs
return alldocs
}
}