Merge pull request #324 from govind-kumarr/feature/Subtitles-Loader-Integration

Feature: Subtitles Loader Integration
This commit is contained in:
Henry Heng 2023-06-15 15:05:05 +01:00 committed by GitHub
commit b79e72eea4
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
3 changed files with 99 additions and 2 deletions

View File

@ -0,0 +1,95 @@
import { INode, INodeData, INodeParams } from '../../../src/Interface'
import { TextSplitter } from 'langchain/text_splitter'
import { SRTLoader } from 'langchain/document_loaders/fs/srt'
class Subtitles_DocumentLoaders implements INode {
label: string
name: string
description: string
type: string
icon: string
category: string
baseClasses: string[]
inputs: INodeParams[]
constructor() {
this.label = 'Subtitles File'
this.name = 'subtitlesFile'
this.type = 'Document'
this.icon = 'subtitlesFile.svg'
this.category = 'Document Loaders'
this.description = `Load data from subtitles files`
this.baseClasses = [this.type]
this.inputs = [
{
label: 'Subtitles File',
name: 'subtitlesFile',
type: 'file',
fileType: '.srt'
},
{
label: 'Text Splitter',
name: 'textSplitter',
type: 'TextSplitter',
optional: true
},
{
label: 'Metadata',
name: 'metadata',
type: 'json',
optional: true,
additionalParams: true
}
]
}
async init(nodeData: INodeData): Promise<any> {
const textSplitter = nodeData.inputs?.textSplitter as TextSplitter
const subtitlesFileBase64 = nodeData.inputs?.subtitlesFile as string
const metadata = nodeData.inputs?.metadata
let alldocs = []
let files: string[] = []
if (subtitlesFileBase64.startsWith('[') && subtitlesFileBase64.endsWith(']')) {
files = JSON.parse(subtitlesFileBase64)
} else {
files = [subtitlesFileBase64]
}
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 SRTLoader(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 alldocs) {
const newdoc = {
...doc,
metadata: {
...doc.metadata,
...parsedMetadata
}
}
finaldocs.push(newdoc)
}
return finaldocs
}
return alldocs
}
}
module.exports = { nodeClass: Subtitles_DocumentLoaders }

View File

@ -0,0 +1 @@
<svg xmlns="http://www.w3.org/2000/svg" data-name="Layer 1" viewBox="0 0 24 24" id="subtitles"><path d="M20,4.25H4A2.748,2.748,0,0,0,1.25,7V17A2.748,2.748,0,0,0,4,19.75H20A2.748,2.748,0,0,0,22.75,17V7A2.748,2.748,0,0,0,20,4.25Zm-14,5H8a.75.75,0,0,1,0,1.5H6a.75.75,0,0,1,0-1.5Zm7,5.5H6a.75.75,0,0,1,0-1.5h7a.75.75,0,0,1,0,1.5Zm5,0H16a.75.75,0,0,1,0-1.5h2a.75.75,0,0,1,0,1.5Zm0-4H11a.75.75,0,0,1,0-1.5h7a.75.75,0,0,1,0,1.5Z"></path></svg>

After

Width:  |  Height:  |  Size: 436 B

View File

@ -32,16 +32,17 @@
"faiss-node": "^0.2.1",
"form-data": "^4.0.0",
"graphql": "^16.6.0",
"html-to-text": "^9.0.5",
"langchain": "^0.0.94",
"linkifyjs": "^4.1.1",
"mammoth": "^1.5.1",
"moment": "^2.29.3",
"node-fetch": "^2.6.11",
"pdf-parse": "^1.1.1",
"srt-parser-2": "^1.2.3",
"puppeteer": "^20.7.1",
"weaviate-ts-client": "^1.1.0",
"ws": "^8.9.0",
"html-to-text": "^9.0.5"
"ws": "^8.9.0"
},
"devDependencies": {
"@types/gulp": "4.0.9",