Feature/DocumentStore (#2106)

* datasource: initial commit

* datasource: datasource details and chunks

* datasource: Document Store Node

* more changes

* Document Store - Base functionality

* Document Store Loader Component

* Document Store Loader Component

* before merging the modularity PR

* after merging the modularity PR

* preview mode

* initial draft PR

* fixes

* minor updates and  fixes

* preview with loader and splitter

* preview with credential

* show stored chunks

* preview update...

* edit config

* save, preview and other changes

* save, preview and other changes

* save, process and other changes

* save, process and other changes

* alpha1 - for internal testing

* rerouting urls

* bug fix on new leader create

* pagination support for chunks

* delete document store

* Update pnpm-lock.yaml

* doc store card view

* Update store files to use updated storage functions, Document Store Table View and other changes

* ui changes

* add expanded chunk dialog, improve ui

* change throw Error to InternalError

* Bug Fixes and removal of subFolder, adding of view chunks for store

* lint fixes

* merge changes

* DocumentStoreStatus component

* ui changes for doc store

* add remove metadata key field, add custom document loader

* add chatflows used doc store chips

* add types/interfaces to DocumentStore Services

* document loader list dialog title bar color change

* update interfaces

* Whereused Chatflow Name and Added chunkNo to retain order of created chunks.

* use typeorm order chunkNo, ui changes

---------

Co-authored-by: Henry <hzj94@hotmail.com>
Co-authored-by: Henry Heng <henryheng@flowiseai.com>
This commit is contained in:
Vinod Kiran
2024-05-06 19:53:27 +05:30
committed by GitHub
parent af4e28aa91
commit 40e36d1b39
91 changed files with 38713 additions and 32791 deletions
@@ -1,7 +1,7 @@
import { ICommonObject, INode, INodeData, INodeOutputsValue, INodeParams } from '../../../src/Interface'
import { omit } from 'lodash'
import { ICommonObject, IDocument, INode, INodeData, INodeOutputsValue, INodeParams } from '../../../src/Interface'
import { TextSplitter } from 'langchain/text_splitter'
import { TextLoader } from 'langchain/document_loaders/fs/text'
import { Document } from '@langchain/core/documents'
import { getFileFromStorage, handleEscapeCharacters } from '../../../src'
class Text_DocumentLoaders implements INode {
@@ -40,9 +40,21 @@ class Text_DocumentLoaders implements INode {
optional: true
},
{
label: 'Metadata',
label: 'Additional Metadata',
name: 'metadata',
type: 'json',
description: 'Additional metadata to be added to the extracted documents',
optional: true,
additionalParams: true
},
{
label: 'Omit Metadata Keys',
name: 'omitMetadataKeys',
type: 'string',
rows: 4,
description:
'Each document loader comes with a default set of metadata keys that are extracted from the document. You can use this field to omit some of the default metadata keys. The value should be a list of keys, seperated by comma',
placeholder: 'key1, key2, key3.nestedKey1',
optional: true,
additionalParams: true
}
@@ -68,8 +80,14 @@ class Text_DocumentLoaders implements INode {
const txtFileBase64 = nodeData.inputs?.txtFile as string
const metadata = nodeData.inputs?.metadata
const output = nodeData.outputs?.output as string
const _omitMetadataKeys = nodeData.inputs?.omitMetadataKeys as string
let alldocs = []
let omitMetadataKeys: string[] = []
if (_omitMetadataKeys) {
omitMetadataKeys = _omitMetadataKeys.split(',').map((key) => key.trim())
}
let docs: IDocument[] = []
let files: string[] = []
//FILE-STORAGE::["CONTRIBUTING.md","LICENSE.md","README.md"]
@@ -88,11 +106,9 @@ class Text_DocumentLoaders implements INode {
const loader = new TextLoader(blob)
if (textSplitter) {
const docs = await loader.loadAndSplit(textSplitter)
alldocs.push(...docs)
docs.push(...(await loader.loadAndSplit(textSplitter)))
} else {
const docs = await loader.load()
alldocs.push(...docs)
docs.push(...(await loader.load()))
}
}
} else {
@@ -110,37 +126,42 @@ class Text_DocumentLoaders implements INode {
const loader = new TextLoader(blob)
if (textSplitter) {
const docs = await loader.loadAndSplit(textSplitter)
alldocs.push(...docs)
docs.push(...(await loader.loadAndSplit(textSplitter)))
} else {
const docs = await loader.load()
alldocs.push(...docs)
docs.push(...(await loader.load()))
}
}
}
let finaldocs: Document<Record<string, any>>[] = []
if (metadata) {
const parsedMetadata = typeof metadata === 'object' ? metadata : JSON.parse(metadata)
for (const doc of alldocs) {
const newdoc = {
...doc,
metadata: {
docs = docs.map((doc) => ({
...doc,
metadata: omit(
{
...doc.metadata,
...parsedMetadata
}
}
finaldocs.push(newdoc)
}
},
omitMetadataKeys
)
}))
} else {
finaldocs = alldocs
docs = docs.map((doc) => ({
...doc,
metadata: omit(
{
...doc.metadata
},
omitMetadataKeys
)
}))
}
if (output === 'document') {
return finaldocs
return docs
} else {
let finaltext = ''
for (const doc of finaldocs) {
for (const doc of docs) {
finaltext += `${doc.pageContent}\n`
}
return handleEscapeCharacters(finaltext, false)