Chore/LC v0.3 (#3517)

* bump langchain version to 0.3, upgrades on all chat models

* update all docs loader to have documents and text output options

* fix pnpm lock file
This commit is contained in:
Henry Heng
2024-11-28 11:06:12 +00:00
committed by GitHub
parent 126808b62a
commit 940c8fd3b0
82 changed files with 2373 additions and 1540 deletions
@@ -8,7 +8,7 @@ import {
PlaywrightWebBaseLoaderOptions
} from '@langchain/community/document_loaders/web/playwright'
import { test } from 'linkifyjs'
import { webCrawl, xmlScrape } from '../../../src'
import { handleEscapeCharacters, INodeOutputsValue, webCrawl, xmlScrape } from '../../../src'
class Playwright_DocumentLoaders implements INode {
label: string
@@ -20,11 +20,12 @@ class Playwright_DocumentLoaders implements INode {
category: string
baseClasses: string[]
inputs: INodeParams[]
outputs: INodeOutputsValue[]
constructor() {
this.label = 'Playwright Web Scraper'
this.name = 'playwrightWebScraper'
this.version = 1.0
this.version = 2.0
this.type = 'Document'
this.icon = 'playwright.svg'
this.category = 'Document Loaders'
@@ -132,6 +133,20 @@ class Playwright_DocumentLoaders implements INode {
additionalParams: true
}
]
this.outputs = [
{
label: 'Document',
name: 'document',
description: 'Array of document objects containing metadata and pageContent',
baseClasses: [...this.baseClasses, 'json']
},
{
label: 'Text',
name: 'text',
description: 'Concatenated string from pageContent of documents',
baseClasses: ['string', 'json']
}
]
}
async init(nodeData: INodeData, _: string, options: ICommonObject): Promise<any> {
@@ -143,6 +158,7 @@ class Playwright_DocumentLoaders implements INode {
let waitUntilGoToOption = nodeData.inputs?.waitUntilGoToOption as 'load' | 'domcontentloaded' | 'networkidle' | 'commit' | undefined
let waitForSelector = nodeData.inputs?.waitForSelector as string
const _omitMetadataKeys = nodeData.inputs?.omitMetadataKeys as string
const output = nodeData.outputs?.output as string
let omitMetadataKeys: string[] = []
if (_omitMetadataKeys) {
@@ -251,7 +267,15 @@ class Playwright_DocumentLoaders implements INode {
}))
}
return docs
if (output === 'document') {
return docs
} else {
let finaltext = ''
for (const doc of docs) {
finaltext += `${doc.pageContent}\n`
}
return handleEscapeCharacters(finaltext, false)
}
}
}