diff --git a/packages/components/README-ZH.md b/packages/components/README-ZH.md index 0605812f..12cb240b 100644 --- a/packages/components/README-ZH.md +++ b/packages/components/README-ZH.md @@ -16,4 +16,4 @@ npm i flowise-components ## 许可证 -此存储库中的源代码在[MIT许可证](https://github.com/FlowiseAI/Flowise/blob/master/LICENSE.md)下提供。 \ No newline at end of file +此存储库中的源代码在[MIT许可证](https://github.com/FlowiseAI/Flowise/blob/master/LICENSE.md)下提供。 diff --git a/packages/components/credentials/MilvusAuth.credential.ts b/packages/components/credentials/MilvusAuth.credential.ts new file mode 100644 index 00000000..b94e1fc8 --- /dev/null +++ b/packages/components/credentials/MilvusAuth.credential.ts @@ -0,0 +1,31 @@ +import { INodeParams, INodeCredential } from '../src/Interface' + +class MilvusCredential implements INodeCredential { + label: string + name: string + version: number + description: string + inputs: INodeParams[] + + constructor() { + this.label = 'Milvus Auth' + this.name = 'milvusAuth' + this.version = 1.0 + this.description = + 'You can find the Milvus Authentication from here page.' + this.inputs = [ + { + label: 'Milvus User', + name: 'milvusUser', + type: 'string' + }, + { + label: 'Milvus Password', + name: 'milvusPassword', + type: 'password' + } + ] + } +} + +module.exports = { credClass: MilvusCredential } diff --git a/packages/components/nodes/chains/ApiChain/apichain.svg b/packages/components/nodes/chains/ApiChain/apichain.svg index ef62e168..3b86b905 100644 --- a/packages/components/nodes/chains/ApiChain/apichain.svg +++ b/packages/components/nodes/chains/ApiChain/apichain.svg @@ -1,3 +1,3 @@ - \ No newline at end of file diff --git a/packages/components/nodes/chains/ConversationChain/ConversationChain.ts b/packages/components/nodes/chains/ConversationChain/ConversationChain.ts index cd42a670..08663395 100644 --- a/packages/components/nodes/chains/ConversationChain/ConversationChain.ts +++ b/packages/components/nodes/chains/ConversationChain/ConversationChain.ts @@ -45,7 +45,8 @@ class ConversationChain_Chains implements INode { label: 'Document', name: 'document', type: 'Document', - description: 'Include whole document into the context window', + description: + 'Include whole document into the context window, if you get maximum context length error, please use model with higher context window like Claude 100k, or gpt4 32k', optional: true, list: true }, diff --git a/packages/components/nodes/chatmodels/ChatOpenAI/ChatOpenAI.ts b/packages/components/nodes/chatmodels/ChatOpenAI/ChatOpenAI.ts index 9512da66..ca081ff4 100644 --- a/packages/components/nodes/chatmodels/ChatOpenAI/ChatOpenAI.ts +++ b/packages/components/nodes/chatmodels/ChatOpenAI/ChatOpenAI.ts @@ -125,6 +125,13 @@ class ChatOpenAI_ChatModels implements INode { type: 'string', optional: true, additionalParams: true + }, + { + label: 'BaseOptions', + name: 'baseOptions', + type: 'json', + optional: true, + additionalParams: true } ] } @@ -139,6 +146,7 @@ class ChatOpenAI_ChatModels implements INode { const timeout = nodeData.inputs?.timeout as string const streaming = nodeData.inputs?.streaming as boolean const basePath = nodeData.inputs?.basepath as string + const baseOptions = nodeData.inputs?.baseOptions const credentialData = await getCredentialData(nodeData.credential ?? '', options) const openAIApiKey = getCredentialParam('openAIApiKey', credentialData, nodeData) @@ -156,8 +164,18 @@ class ChatOpenAI_ChatModels implements INode { if (presencePenalty) obj.presencePenalty = parseFloat(presencePenalty) if (timeout) obj.timeout = parseInt(timeout, 10) + let parsedBaseOptions: any | undefined = undefined + + if (baseOptions) { + try { + parsedBaseOptions = typeof baseOptions === 'object' ? baseOptions : JSON.parse(baseOptions) + } catch (exception) { + throw new Error("Invalid JSON in the ChatOpenAI's BaseOptions: " + exception) + } + } const model = new ChatOpenAI(obj, { - basePath + basePath, + baseOptions: parsedBaseOptions }) return model } diff --git a/packages/components/nodes/documentloaders/Cheerio/Cheerio.ts b/packages/components/nodes/documentloaders/Cheerio/Cheerio.ts index 310aa9e6..1c21c1ea 100644 --- a/packages/components/nodes/documentloaders/Cheerio/Cheerio.ts +++ b/packages/components/nodes/documentloaders/Cheerio/Cheerio.ts @@ -64,7 +64,7 @@ class Cheerio_DocumentLoaders implements INode { additionalParams: true, description: 'Only used when "Get Relative Links Method" is selected. Set 0 to retrieve all relative links, default limit is 10.', - warning: `Retreiving all links might take long time, and all links will be upserted again if the flow's state changed (eg: different URL, chunk size, etc)` + warning: `Retrieving all links might take long time, and all links will be upserted again if the flow's state changed (eg: different URL, chunk size, etc)` }, { label: 'Metadata', diff --git a/packages/components/nodes/documentloaders/Playwright/Playwright.ts b/packages/components/nodes/documentloaders/Playwright/Playwright.ts index 3399574d..eb246045 100644 --- a/packages/components/nodes/documentloaders/Playwright/Playwright.ts +++ b/packages/components/nodes/documentloaders/Playwright/Playwright.ts @@ -1,6 +1,6 @@ import { INode, INodeData, INodeParams } from '../../../src/Interface' import { TextSplitter } from 'langchain/text_splitter' -import { PlaywrightWebBaseLoader } from 'langchain/document_loaders/web/playwright' +import { Browser, Page, PlaywrightWebBaseLoader, PlaywrightWebBaseLoaderOptions } from 'langchain/document_loaders/web/playwright' import { test } from 'linkifyjs' import { webCrawl, xmlScrape } from '../../../src' @@ -64,7 +64,45 @@ class Playwright_DocumentLoaders implements INode { additionalParams: true, description: 'Only used when "Get Relative Links Method" is selected. Set 0 to retrieve all relative links, default limit is 10.', - warning: `Retreiving all links might take long time, and all links will be upserted again if the flow's state changed (eg: different URL, chunk size, etc)` + warning: `Retrieving all links might take long time, and all links will be upserted again if the flow's state changed (eg: different URL, chunk size, etc)` + }, + { + label: 'Wait Until', + name: 'waitUntilGoToOption', + type: 'options', + description: 'Select a go to wait until option', + options: [ + { + label: 'Load', + name: 'load', + description: 'Consider operation to be finished when the load event is fired.' + }, + { + label: 'DOM Content Loaded', + name: 'domcontentloaded', + description: 'Consider operation to be finished when the DOMContentLoaded event is fired.' + }, + { + label: 'Network Idle', + name: 'networkidle', + description: 'Navigation is finished when there are no more connections for at least 500 ms.' + }, + { + label: 'Commit', + name: 'commit', + description: 'Consider operation to be finished when network response is received and the document started loading.' + } + ], + optional: true, + additionalParams: true + }, + { + label: 'Wait for selector to load', + name: 'waitForSelector', + type: 'string', + optional: true, + additionalParams: true, + description: 'CSS selectors like .div or #div' }, { label: 'Metadata', @@ -81,6 +119,8 @@ class Playwright_DocumentLoaders implements INode { const metadata = nodeData.inputs?.metadata const relativeLinksMethod = nodeData.inputs?.relativeLinksMethod as string let limit = nodeData.inputs?.limit as string + let waitUntilGoToOption = nodeData.inputs?.waitUntilGoToOption as 'load' | 'domcontentloaded' | 'networkidle' | 'commit' | undefined + let waitForSelector = nodeData.inputs?.waitForSelector as string let url = nodeData.inputs?.url as string url = url.trim() @@ -91,7 +131,26 @@ class Playwright_DocumentLoaders implements INode { async function playwrightLoader(url: string): Promise { try { let docs = [] - const loader = new PlaywrightWebBaseLoader(url) + const config: PlaywrightWebBaseLoaderOptions = { + launchOptions: { + args: ['--no-sandbox'], + headless: true + } + } + if (waitUntilGoToOption) { + config['gotoOptions'] = { + waitUntil: waitUntilGoToOption + } + } + if (waitForSelector) { + config['evaluate'] = async (page: Page, _: Browser): Promise => { + await page.waitForSelector(waitForSelector) + + const result = await page.evaluate(() => document.body.innerHTML) + return result + } + } + const loader = new PlaywrightWebBaseLoader(url, config) if (textSplitter) { docs = await loader.loadAndSplit(textSplitter) } else { diff --git a/packages/components/nodes/documentloaders/Puppeteer/Puppeteer.ts b/packages/components/nodes/documentloaders/Puppeteer/Puppeteer.ts index ea6280db..4691eb94 100644 --- a/packages/components/nodes/documentloaders/Puppeteer/Puppeteer.ts +++ b/packages/components/nodes/documentloaders/Puppeteer/Puppeteer.ts @@ -1,8 +1,9 @@ import { INode, INodeData, INodeParams } from '../../../src/Interface' import { TextSplitter } from 'langchain/text_splitter' -import { PuppeteerWebBaseLoader } from 'langchain/document_loaders/web/puppeteer' +import { Browser, Page, PuppeteerWebBaseLoader, PuppeteerWebBaseLoaderOptions } from 'langchain/document_loaders/web/puppeteer' import { test } from 'linkifyjs' import { webCrawl, xmlScrape } from '../../../src' +import { PuppeteerLifeCycleEvent } from 'puppeteer' class Puppeteer_DocumentLoaders implements INode { label: string @@ -64,7 +65,45 @@ class Puppeteer_DocumentLoaders implements INode { additionalParams: true, description: 'Only used when "Get Relative Links Method" is selected. Set 0 to retrieve all relative links, default limit is 10.', - warning: `Retreiving all links might take long time, and all links will be upserted again if the flow's state changed (eg: different URL, chunk size, etc)` + warning: `Retrieving all links might take long time, and all links will be upserted again if the flow's state changed (eg: different URL, chunk size, etc)` + }, + { + label: 'Wait Until', + name: 'waitUntilGoToOption', + type: 'options', + description: 'Select a go to wait until option', + options: [ + { + label: 'Load', + name: 'load', + description: `When the initial HTML document's DOM has been loaded and parsed` + }, + { + label: 'DOM Content Loaded', + name: 'domcontentloaded', + description: `When the complete HTML document's DOM has been loaded and parsed` + }, + { + label: 'Network Idle 0', + name: 'networkidle0', + description: 'Navigation is finished when there are no more than 0 network connections for at least 500 ms' + }, + { + label: 'Network Idle 2', + name: 'networkidle2', + description: 'Navigation is finished when there are no more than 2 network connections for at least 500 ms' + } + ], + optional: true, + additionalParams: true + }, + { + label: 'Wait for selector to load', + name: 'waitForSelector', + type: 'string', + optional: true, + additionalParams: true, + description: 'CSS selectors like .div or #div' }, { label: 'Metadata', @@ -81,6 +120,8 @@ class Puppeteer_DocumentLoaders implements INode { const metadata = nodeData.inputs?.metadata const relativeLinksMethod = nodeData.inputs?.relativeLinksMethod as string let limit = nodeData.inputs?.limit as string + let waitUntilGoToOption = nodeData.inputs?.waitUntilGoToOption as PuppeteerLifeCycleEvent + let waitForSelector = nodeData.inputs?.waitForSelector as string let url = nodeData.inputs?.url as string url = url.trim() @@ -91,12 +132,26 @@ class Puppeteer_DocumentLoaders implements INode { async function puppeteerLoader(url: string): Promise { try { let docs = [] - const loader = new PuppeteerWebBaseLoader(url, { + const config: PuppeteerWebBaseLoaderOptions = { launchOptions: { args: ['--no-sandbox'], headless: 'new' } - }) + } + if (waitUntilGoToOption) { + config['gotoOptions'] = { + waitUntil: waitUntilGoToOption + } + } + if (waitForSelector) { + config['evaluate'] = async (page: Page, _: Browser): Promise => { + await page.waitForSelector(waitForSelector) + + const result = await page.evaluate(() => document.body.innerHTML) + return result + } + } + const loader = new PuppeteerWebBaseLoader(url, config) if (textSplitter) { docs = await loader.loadAndSplit(textSplitter) } else { diff --git a/packages/components/nodes/documentloaders/VectorStoreToDocument/VectorStoreToDocument.ts b/packages/components/nodes/documentloaders/VectorStoreToDocument/VectorStoreToDocument.ts new file mode 100644 index 00000000..b3f320ce --- /dev/null +++ b/packages/components/nodes/documentloaders/VectorStoreToDocument/VectorStoreToDocument.ts @@ -0,0 +1,87 @@ +import { VectorStore } from 'langchain/vectorstores/base' +import { INode, INodeData, INodeOutputsValue, INodeParams } from '../../../src/Interface' +import { handleEscapeCharacters } from '../../../src/utils' + +class VectorStoreToDocument_DocumentLoaders implements INode { + label: string + name: string + version: number + description: string + type: string + icon: string + category: string + baseClasses: string[] + inputs: INodeParams[] + outputs: INodeOutputsValue[] + + constructor() { + this.label = 'VectorStore To Document' + this.name = 'vectorStoreToDocument' + this.version = 1.0 + this.type = 'Document' + this.icon = 'vectorretriever.svg' + this.category = 'Document Loaders' + this.description = 'Search documents with scores from vector store' + this.baseClasses = [this.type] + this.inputs = [ + { + label: 'Vector Store', + name: 'vectorStore', + type: 'VectorStore' + }, + { + label: 'Minimum Score (%)', + name: 'minScore', + type: 'number', + optional: true, + placeholder: '75', + step: 1, + description: 'Minumum score for embeddings documents to be included' + } + ] + this.outputs = [ + { + label: 'Document', + name: 'document', + baseClasses: this.baseClasses + }, + { + label: 'Text', + name: 'text', + baseClasses: ['string', 'json'] + } + ] + } + + async init(nodeData: INodeData, input: string): Promise { + const vectorStore = nodeData.inputs?.vectorStore as VectorStore + const minScore = nodeData.inputs?.minScore as number + const output = nodeData.outputs?.output as string + + const topK = (vectorStore as any)?.k ?? 4 + + const docs = await vectorStore.similaritySearchWithScore(input, topK) + // eslint-disable-next-line no-console + console.log('\x1b[94m\x1b[1m\n*****VectorStore Documents*****\n\x1b[0m\x1b[0m') + // eslint-disable-next-line no-console + console.log(docs) + + if (output === 'document') { + let finaldocs = [] + for (const doc of docs) { + if (minScore && doc[1] < minScore / 100) continue + finaldocs.push(doc[0]) + } + return finaldocs + } else { + let finaltext = '' + for (const doc of docs) { + if (minScore && doc[1] < minScore / 100) continue + finaltext += `${doc[0].pageContent}\n` + } + return handleEscapeCharacters(finaltext, false) + } + } +} + +module.exports = { nodeClass: VectorStoreToDocument_DocumentLoaders } diff --git a/packages/components/nodes/documentloaders/VectorStoreToDocument/vectorretriever.svg b/packages/components/nodes/documentloaders/VectorStoreToDocument/vectorretriever.svg new file mode 100644 index 00000000..208a59f1 --- /dev/null +++ b/packages/components/nodes/documentloaders/VectorStoreToDocument/vectorretriever.svg @@ -0,0 +1,7 @@ + + + + + + + \ No newline at end of file diff --git a/packages/components/nodes/memory/ZepMemory/ZepMemory.ts b/packages/components/nodes/memory/ZepMemory/ZepMemory.ts index f2a47852..0c05563a 100644 --- a/packages/components/nodes/memory/ZepMemory/ZepMemory.ts +++ b/packages/components/nodes/memory/ZepMemory/ZepMemory.ts @@ -120,7 +120,8 @@ class ZepMemory_Memory implements INode { zep.loadMemoryVariables = async (values) => { let data = await tmpFunc.bind(zep, values)() if (autoSummary && zep.returnMessages && data[zep.memoryKey] && data[zep.memoryKey].length) { - const memory = await zep.zepClient.getMemory(zep.sessionId, parseInt(k, 10) ?? 10) + const zepClient = await zep.zepClientPromise + const memory = await zepClient.memory.getMemory(zep.sessionId, parseInt(k, 10) ?? 10) if (memory?.summary) { let summary = autoSummaryTemplate.replace(/{summary}/g, memory.summary.content) // eslint-disable-next-line no-console @@ -190,23 +191,6 @@ class ZepMemoryExtended extends ZepMemory { super(fields) this.isSessionIdUsingChatMessageId = fields.isSessionIdUsingChatMessageId } - - async clear(): Promise { - // Only clear when sessionId is using chatId - // If sessionId is specified, clearing and inserting again will error because the sessionId has been soft deleted - // If using chatId, it will not be a problem because the sessionId will always be the new chatId - if (this.isSessionIdUsingChatMessageId) { - try { - await this.zepClient.deleteMemory(this.sessionId) - } catch (error) { - console.error('Error deleting session: ', error) - } - - // Clear the superclass's chat history - await super.clear() - } - await this.chatHistory.clear() - } } module.exports = { nodeClass: ZepMemory_Memory } diff --git a/packages/components/nodes/memory/ZepMemory/zep.png b/packages/components/nodes/memory/ZepMemory/zep.png index 293be6f6..2fdb2382 100644 Binary files a/packages/components/nodes/memory/ZepMemory/zep.png and b/packages/components/nodes/memory/ZepMemory/zep.png differ diff --git a/packages/components/nodes/vectorstores/Faiss_Existing/Faiss_Existing.ts b/packages/components/nodes/vectorstores/Faiss_Existing/Faiss_Existing.ts index 8c8d03a8..15d476d8 100644 --- a/packages/components/nodes/vectorstores/Faiss_Existing/Faiss_Existing.ts +++ b/packages/components/nodes/vectorstores/Faiss_Existing/Faiss_Existing.ts @@ -2,6 +2,7 @@ import { INode, INodeData, INodeOutputsValue, INodeParams } from '../../../src/I import { FaissStore } from 'langchain/vectorstores/faiss' import { Embeddings } from 'langchain/embeddings/base' import { getBaseClasses } from '../../../src/utils' +import { Document } from 'langchain/document' class Faiss_Existing_VectorStores implements INode { label: string @@ -70,6 +71,23 @@ class Faiss_Existing_VectorStores implements INode { const vectorStore = await FaissStore.load(basePath, embeddings) + // Avoid illegal invocation error + vectorStore.similaritySearchVectorWithScore = async (query: number[], k: number) => { + const index = vectorStore.index + + if (k > index.ntotal()) { + const total = index.ntotal() + console.warn(`k (${k}) is greater than the number of elements in the index (${total}), setting k to ${total}`) + k = total + } + + const result = index.search(query, k) + return result.labels.map((id, index) => { + const uuid = vectorStore._mapping[id] + return [vectorStore.docstore.search(uuid), result.distances[index]] as [Document, number] + }) + } + if (output === 'retriever') { const retriever = vectorStore.asRetriever(k) return retriever diff --git a/packages/components/nodes/vectorstores/Faiss_Upsert/Faiss_Upsert.ts b/packages/components/nodes/vectorstores/Faiss_Upsert/Faiss_Upsert.ts index f56eccdf..c234a4f5 100644 --- a/packages/components/nodes/vectorstores/Faiss_Upsert/Faiss_Upsert.ts +++ b/packages/components/nodes/vectorstores/Faiss_Upsert/Faiss_Upsert.ts @@ -86,6 +86,23 @@ class FaissUpsert_VectorStores implements INode { const vectorStore = await FaissStore.fromDocuments(finalDocs, embeddings) await vectorStore.save(basePath) + // Avoid illegal invocation error + vectorStore.similaritySearchVectorWithScore = async (query: number[], k: number) => { + const index = vectorStore.index + + if (k > index.ntotal()) { + const total = index.ntotal() + console.warn(`k (${k}) is greater than the number of elements in the index (${total}), setting k to ${total}`) + k = total + } + + const result = index.search(query, k) + return result.labels.map((id, index) => { + const uuid = vectorStore._mapping[id] + return [vectorStore.docstore.search(uuid), result.distances[index]] as [Document, number] + }) + } + if (output === 'retriever') { const retriever = vectorStore.asRetriever(k) return retriever diff --git a/packages/components/nodes/vectorstores/Milvus/Milvus_Existing.ts b/packages/components/nodes/vectorstores/Milvus/Milvus_Existing.ts new file mode 100644 index 00000000..514fdc73 --- /dev/null +++ b/packages/components/nodes/vectorstores/Milvus/Milvus_Existing.ts @@ -0,0 +1,185 @@ +import { ICommonObject, INode, INodeData, INodeOutputsValue, INodeParams } from '../../../src/Interface' +import { DataType, ErrorCode } from '@zilliz/milvus2-sdk-node' +import { MilvusLibArgs, Milvus } from 'langchain/vectorstores/milvus' +import { Embeddings } from 'langchain/embeddings/base' +import { getBaseClasses, getCredentialData, getCredentialParam } from '../../../src/utils' +import { Document } from 'langchain/document' + +class Milvus_Existing_VectorStores implements INode { + label: string + name: string + version: number + description: string + type: string + icon: string + category: string + baseClasses: string[] + inputs: INodeParams[] + credential: INodeParams + outputs: INodeOutputsValue[] + + constructor() { + this.label = 'Milvus Load Existing collection' + this.name = 'milvusExistingCollection' + this.version = 1.0 + this.type = 'Milvus' + this.icon = 'milvus.svg' + this.category = 'Vector Stores' + this.description = 'Load existing collection from Milvus (i.e: Document has been upserted)' + this.baseClasses = [this.type, 'VectorStoreRetriever', 'BaseRetriever'] + this.credential = { + label: 'Connect Credential', + name: 'credential', + type: 'credential', + optional: true, + credentialNames: ['milvusAuth'] + } + this.inputs = [ + { + label: 'Embeddings', + name: 'embeddings', + type: 'Embeddings' + }, + { + label: 'Milvus Server URL', + name: 'milvusServerUrl', + type: 'string', + placeholder: 'http://localhost:19530' + }, + { + label: 'Milvus Collection Name', + name: 'milvusCollection', + type: 'string' + } + ] + this.outputs = [ + { + label: 'Milvus Retriever', + name: 'retriever', + baseClasses: this.baseClasses + }, + { + label: 'Milvus Vector Store', + name: 'vectorStore', + baseClasses: [this.type, ...getBaseClasses(Milvus)] + } + ] + } + + async init(nodeData: INodeData, _: string, options: ICommonObject): Promise { + // server setup + const address = nodeData.inputs?.milvusServerUrl as string + const collectionName = nodeData.inputs?.milvusCollection as string + + // embeddings + const embeddings = nodeData.inputs?.embeddings as Embeddings + const topK = nodeData.inputs?.topK as string + + // output + const output = nodeData.outputs?.output as string + + // format data + const k = topK ? parseInt(topK, 10) : 4 + + // credential + const credentialData = await getCredentialData(nodeData.credential ?? '', options) + const milvusUser = getCredentialParam('milvusUser', credentialData, nodeData) + const milvusPassword = getCredentialParam('milvusPassword', credentialData, nodeData) + + // init MilvusLibArgs + const milVusArgs: MilvusLibArgs = { + url: address, + collectionName: collectionName + } + + if (milvusUser) milVusArgs.username = milvusUser + if (milvusPassword) milVusArgs.password = milvusPassword + + const vectorStore = await Milvus.fromExistingCollection(embeddings, milVusArgs) + + // Avoid Illegal Invocation + vectorStore.similaritySearchVectorWithScore = async (query: number[], k: number, filter?: string) => { + const hasColResp = await vectorStore.client.hasCollection({ + collection_name: vectorStore.collectionName + }) + if (hasColResp.status.error_code !== ErrorCode.SUCCESS) { + throw new Error(`Error checking collection: ${hasColResp}`) + } + if (hasColResp.value === false) { + throw new Error(`Collection not found: ${vectorStore.collectionName}, please create collection before search.`) + } + + const filterStr = filter ?? '' + + await vectorStore.grabCollectionFields() + + const loadResp = await vectorStore.client.loadCollectionSync({ + collection_name: vectorStore.collectionName + }) + + if (loadResp.error_code !== ErrorCode.SUCCESS) { + throw new Error(`Error loading collection: ${loadResp}`) + } + + const outputFields = vectorStore.fields.filter((field) => field !== vectorStore.vectorField) + + const searchResp = await vectorStore.client.search({ + collection_name: vectorStore.collectionName, + search_params: { + anns_field: vectorStore.vectorField, + topk: k.toString(), + metric_type: vectorStore.indexCreateParams.metric_type, + params: vectorStore.indexSearchParams + }, + output_fields: outputFields, + vector_type: DataType.FloatVector, + vectors: [query], + filter: filterStr + }) + if (searchResp.status.error_code !== ErrorCode.SUCCESS) { + throw new Error(`Error searching data: ${JSON.stringify(searchResp)}`) + } + const results: [Document, number][] = [] + searchResp.results.forEach((result) => { + const fields = { + pageContent: '', + metadata: {} as Record + } + Object.keys(result).forEach((key) => { + if (key === vectorStore.textField) { + fields.pageContent = result[key] + } else if (vectorStore.fields.includes(key) || key === vectorStore.primaryField) { + if (typeof result[key] === 'string') { + const { isJson, obj } = checkJsonString(result[key]) + fields.metadata[key] = isJson ? obj : result[key] + } else { + fields.metadata[key] = result[key] + } + } + }) + results.push([new Document(fields), result.score]) + }) + return results + } + + if (output === 'retriever') { + const retriever = vectorStore.asRetriever(k) + return retriever + } else if (output === 'vectorStore') { + ;(vectorStore as any).k = k + return vectorStore + } + return vectorStore + } +} + +function checkJsonString(value: string): { isJson: boolean; obj: any } { + try { + const result = JSON.parse(value) + return { isJson: true, obj: result } + } catch (e) { + return { isJson: false, obj: null } + } +} + +module.exports = { nodeClass: Milvus_Existing_VectorStores } diff --git a/packages/components/nodes/vectorstores/Milvus/Milvus_Upsert.ts b/packages/components/nodes/vectorstores/Milvus/Milvus_Upsert.ts new file mode 100644 index 00000000..ca69cb39 --- /dev/null +++ b/packages/components/nodes/vectorstores/Milvus/Milvus_Upsert.ts @@ -0,0 +1,281 @@ +import { ICommonObject, INode, INodeData, INodeOutputsValue, INodeParams } from '../../../src/Interface' +import { DataType, ErrorCode, MetricType, IndexType } from '@zilliz/milvus2-sdk-node' +import { MilvusLibArgs, Milvus } from 'langchain/vectorstores/milvus' +import { Embeddings } from 'langchain/embeddings/base' +import { Document } from 'langchain/document' +import { getBaseClasses, getCredentialData, getCredentialParam } from '../../../src/utils' +import { flatten } from 'lodash' + +interface InsertRow { + [x: string]: string | number[] +} + +class Milvus_Upsert_VectorStores implements INode { + label: string + name: string + version: number + description: string + type: string + icon: string + category: string + baseClasses: string[] + inputs: INodeParams[] + credential: INodeParams + outputs: INodeOutputsValue[] + + constructor() { + this.label = 'Milvus Upsert Document' + this.name = 'milvusUpsert' + this.version = 1.0 + this.type = 'Milvus' + this.icon = 'milvus.svg' + this.category = 'Vector Stores' + this.description = 'Upsert documents to Milvus' + this.baseClasses = [this.type, 'VectorStoreRetriever', 'BaseRetriever'] + this.credential = { + label: 'Connect Credential', + name: 'credential', + type: 'credential', + optional: true, + credentialNames: ['milvusAuth'] + } + this.inputs = [ + { + label: 'Document', + name: 'document', + type: 'Document', + list: true + }, + { + label: 'Embeddings', + name: 'embeddings', + type: 'Embeddings' + }, + { + label: 'Milvus Server URL', + name: 'milvusServerUrl', + type: 'string', + placeholder: 'http://localhost:19530' + }, + { + label: 'Milvus Collection Name', + name: 'milvusCollection', + type: 'string' + } + ] + this.outputs = [ + { + label: 'Milvus Retriever', + name: 'retriever', + baseClasses: this.baseClasses + }, + { + label: 'Milvus Vector Store', + name: 'vectorStore', + baseClasses: [this.type, ...getBaseClasses(Milvus)] + } + ] + } + + async init(nodeData: INodeData, _: string, options: ICommonObject): Promise { + // server setup + const address = nodeData.inputs?.milvusServerUrl as string + const collectionName = nodeData.inputs?.milvusCollection as string + + // embeddings + const docs = nodeData.inputs?.document as Document[] + const embeddings = nodeData.inputs?.embeddings as Embeddings + const topK = nodeData.inputs?.topK as string + + // output + const output = nodeData.outputs?.output as string + + // format data + const k = topK ? parseInt(topK, 10) : 4 + + // credential + const credentialData = await getCredentialData(nodeData.credential ?? '', options) + const milvusUser = getCredentialParam('milvusUser', credentialData, nodeData) + const milvusPassword = getCredentialParam('milvusPassword', credentialData, nodeData) + + // init MilvusLibArgs + const milVusArgs: MilvusLibArgs = { + url: address, + collectionName: collectionName + } + + if (milvusUser) milVusArgs.username = milvusUser + if (milvusPassword) milVusArgs.password = milvusPassword + + const flattenDocs = docs && docs.length ? flatten(docs) : [] + const finalDocs = [] + for (let i = 0; i < flattenDocs.length; i += 1) { + finalDocs.push(new Document(flattenDocs[i])) + } + + const vectorStore = await MilvusUpsert.fromDocuments(finalDocs, embeddings, milVusArgs) + + // Avoid Illegal Invocation + vectorStore.similaritySearchVectorWithScore = async (query: number[], k: number, filter?: string) => { + const hasColResp = await vectorStore.client.hasCollection({ + collection_name: vectorStore.collectionName + }) + if (hasColResp.status.error_code !== ErrorCode.SUCCESS) { + throw new Error(`Error checking collection: ${hasColResp}`) + } + if (hasColResp.value === false) { + throw new Error(`Collection not found: ${vectorStore.collectionName}, please create collection before search.`) + } + + const filterStr = filter ?? '' + + await vectorStore.grabCollectionFields() + + const loadResp = await vectorStore.client.loadCollectionSync({ + collection_name: vectorStore.collectionName + }) + if (loadResp.error_code !== ErrorCode.SUCCESS) { + throw new Error(`Error loading collection: ${loadResp}`) + } + + const outputFields = vectorStore.fields.filter((field) => field !== vectorStore.vectorField) + + const searchResp = await vectorStore.client.search({ + collection_name: vectorStore.collectionName, + search_params: { + anns_field: vectorStore.vectorField, + topk: k.toString(), + metric_type: vectorStore.indexCreateParams.metric_type, + params: vectorStore.indexSearchParams + }, + output_fields: outputFields, + vector_type: DataType.FloatVector, + vectors: [query], + filter: filterStr + }) + if (searchResp.status.error_code !== ErrorCode.SUCCESS) { + throw new Error(`Error searching data: ${JSON.stringify(searchResp)}`) + } + const results: [Document, number][] = [] + searchResp.results.forEach((result) => { + const fields = { + pageContent: '', + metadata: {} as Record + } + Object.keys(result).forEach((key) => { + if (key === vectorStore.textField) { + fields.pageContent = result[key] + } else if (vectorStore.fields.includes(key) || key === vectorStore.primaryField) { + if (typeof result[key] === 'string') { + const { isJson, obj } = checkJsonString(result[key]) + fields.metadata[key] = isJson ? obj : result[key] + } else { + fields.metadata[key] = result[key] + } + } + }) + results.push([new Document(fields), result.score]) + }) + return results + } + + if (output === 'retriever') { + const retriever = vectorStore.asRetriever(k) + return retriever + } else if (output === 'vectorStore') { + ;(vectorStore as any).k = k + return vectorStore + } + return vectorStore + } +} + +function checkJsonString(value: string): { isJson: boolean; obj: any } { + try { + const result = JSON.parse(value) + return { isJson: true, obj: result } + } catch (e) { + return { isJson: false, obj: null } + } +} + +class MilvusUpsert extends Milvus { + async addVectors(vectors: number[][], documents: Document[]): Promise { + if (vectors.length === 0) { + return + } + await this.ensureCollection(vectors, documents) + + const insertDatas: InsertRow[] = [] + + for (let index = 0; index < vectors.length; index++) { + const vec = vectors[index] + const doc = documents[index] + const data: InsertRow = { + [this.textField]: doc.pageContent, + [this.vectorField]: vec + } + this.fields.forEach((field) => { + switch (field) { + case this.primaryField: + if (!this.autoId) { + if (doc.metadata[this.primaryField] === undefined) { + throw new Error( + `The Collection's primaryField is configured with autoId=false, thus its value must be provided through metadata.` + ) + } + data[field] = doc.metadata[this.primaryField] + } + break + case this.textField: + data[field] = doc.pageContent + break + case this.vectorField: + data[field] = vec + break + default: // metadata fields + if (doc.metadata[field] === undefined) { + throw new Error(`The field "${field}" is not provided in documents[${index}].metadata.`) + } else if (typeof doc.metadata[field] === 'object') { + data[field] = JSON.stringify(doc.metadata[field]) + } else { + data[field] = doc.metadata[field] + } + break + } + }) + + insertDatas.push(data) + } + + const descIndexResp = await this.client.describeIndex({ + collection_name: this.collectionName + }) + + if (descIndexResp.status.error_code === ErrorCode.INDEX_NOT_EXIST) { + const resp = await this.client.createIndex({ + collection_name: this.collectionName, + field_name: this.vectorField, + index_name: `myindex_${Date.now().toString()}`, + index_type: IndexType.AUTOINDEX, + metric_type: MetricType.L2 + }) + if (resp.error_code !== ErrorCode.SUCCESS) { + throw new Error(`Error creating index`) + } + } + + const insertResp = await this.client.insert({ + collection_name: this.collectionName, + fields_data: insertDatas + }) + + if (insertResp.status.error_code !== ErrorCode.SUCCESS) { + throw new Error(`Error inserting data: ${JSON.stringify(insertResp)}`) + } + + await this.client.flushSync({ collection_names: [this.collectionName] }) + } +} + +module.exports = { nodeClass: Milvus_Upsert_VectorStores } diff --git a/packages/components/nodes/vectorstores/Milvus/milvus.svg b/packages/components/nodes/vectorstores/Milvus/milvus.svg new file mode 100644 index 00000000..68dfef66 --- /dev/null +++ b/packages/components/nodes/vectorstores/Milvus/milvus.svg @@ -0,0 +1,5 @@ + + + + + diff --git a/packages/components/nodes/vectorstores/Vectara_Existing/Vectara_Existing.ts b/packages/components/nodes/vectorstores/Vectara_Existing/Vectara_Existing.ts index f344338a..3ef04f07 100644 --- a/packages/components/nodes/vectorstores/Vectara_Existing/Vectara_Existing.ts +++ b/packages/components/nodes/vectorstores/Vectara_Existing/Vectara_Existing.ts @@ -1,6 +1,6 @@ import { ICommonObject, INode, INodeData, INodeOutputsValue, INodeParams } from '../../../src/Interface' import { getBaseClasses, getCredentialData, getCredentialParam } from '../../../src/utils' -import { VectaraStore, VectaraLibArgs, VectaraFilter } from 'langchain/vectorstores/vectara' +import { VectaraStore, VectaraLibArgs, VectaraFilter, VectaraContextConfig } from 'langchain/vectorstores/vectara' class VectaraExisting_VectorStores implements INode { label: string @@ -40,9 +40,27 @@ class VectaraExisting_VectorStores implements INode { additionalParams: true, optional: true }, + { + label: 'Sentences Before', + name: 'sentencesBefore', + description: 'Number of sentences to fetch before the matched sentence. Defaults to 2.', + type: 'number', + additionalParams: true, + optional: true + }, + { + label: 'Sentences After', + name: 'sentencesAfter', + description: 'Number of sentences to fetch after the matched sentence. Defaults to 2.', + type: 'number', + additionalParams: true, + optional: true + }, { label: 'Lambda', name: 'lambda', + description: + 'Improves retrieval accuracy by adjusting the balance (from 0 to 1) between neural search and keyword-based search factors.', type: 'number', additionalParams: true, optional: true @@ -77,6 +95,8 @@ class VectaraExisting_VectorStores implements INode { const corpusId = getCredentialParam('corpusID', credentialData, nodeData) const vectaraMetadataFilter = nodeData.inputs?.filter as string + const sentencesBefore = nodeData.inputs?.sentencesBefore as number + const sentencesAfter = nodeData.inputs?.sentencesAfter as number const lambda = nodeData.inputs?.lambda as number const output = nodeData.outputs?.output as string const topK = nodeData.inputs?.topK as string @@ -92,6 +112,11 @@ class VectaraExisting_VectorStores implements INode { if (vectaraMetadataFilter) vectaraFilter.filter = vectaraMetadataFilter if (lambda) vectaraFilter.lambda = lambda + const vectaraContextConfig: VectaraContextConfig = {} + if (sentencesBefore) vectaraContextConfig.sentencesBefore = sentencesBefore + if (sentencesAfter) vectaraContextConfig.sentencesAfter = sentencesAfter + vectaraFilter.contextConfig = vectaraContextConfig + const vectorStore = new VectaraStore(vectaraArgs) if (output === 'retriever') { diff --git a/packages/components/nodes/vectorstores/Vectara_Upsert/Vectara_Upsert.ts b/packages/components/nodes/vectorstores/Vectara_Upsert/Vectara_Upsert.ts index b2ee79e7..51fb67ed 100644 --- a/packages/components/nodes/vectorstores/Vectara_Upsert/Vectara_Upsert.ts +++ b/packages/components/nodes/vectorstores/Vectara_Upsert/Vectara_Upsert.ts @@ -1,7 +1,7 @@ import { ICommonObject, INode, INodeData, INodeOutputsValue, INodeParams } from '../../../src/Interface' import { Embeddings } from 'langchain/embeddings/base' import { getBaseClasses, getCredentialData, getCredentialParam } from '../../../src/utils' -import { VectaraStore, VectaraLibArgs, VectaraFilter } from 'langchain/vectorstores/vectara' +import { VectaraStore, VectaraLibArgs, VectaraFilter, VectaraContextConfig } from 'langchain/vectorstores/vectara' import { Document } from 'langchain/document' import { flatten } from 'lodash' @@ -49,9 +49,27 @@ class VectaraUpsert_VectorStores implements INode { additionalParams: true, optional: true }, + { + label: 'Sentences Before', + name: 'sentencesBefore', + description: 'Number of sentences to fetch before the matched sentence. Defaults to 2.', + type: 'number', + additionalParams: true, + optional: true + }, + { + label: 'Sentences After', + name: 'sentencesAfter', + description: 'Number of sentences to fetch after the matched sentence. Defaults to 2.', + type: 'number', + additionalParams: true, + optional: true + }, { label: 'Lambda', name: 'lambda', + description: + 'Improves retrieval accuracy by adjusting the balance (from 0 to 1) between neural search and keyword-based search factors.', type: 'number', additionalParams: true, optional: true @@ -88,6 +106,8 @@ class VectaraUpsert_VectorStores implements INode { const docs = nodeData.inputs?.document as Document[] const embeddings = {} as Embeddings const vectaraMetadataFilter = nodeData.inputs?.filter as string + const sentencesBefore = nodeData.inputs?.sentencesBefore as number + const sentencesAfter = nodeData.inputs?.sentencesAfter as number const lambda = nodeData.inputs?.lambda as number const output = nodeData.outputs?.output as string const topK = nodeData.inputs?.topK as string @@ -103,6 +123,11 @@ class VectaraUpsert_VectorStores implements INode { if (vectaraMetadataFilter) vectaraFilter.filter = vectaraMetadataFilter if (lambda) vectaraFilter.lambda = lambda + const vectaraContextConfig: VectaraContextConfig = {} + if (sentencesBefore) vectaraContextConfig.sentencesBefore = sentencesBefore + if (sentencesAfter) vectaraContextConfig.sentencesAfter = sentencesAfter + vectaraFilter.contextConfig = vectaraContextConfig + const flattenDocs = docs && docs.length ? flatten(docs) : [] const finalDocs = [] for (let i = 0; i < flattenDocs.length; i += 1) { diff --git a/packages/components/nodes/vectorstores/Zep/Zep_Existing.ts b/packages/components/nodes/vectorstores/Zep/Zep_Existing.ts new file mode 100644 index 00000000..a2c2261f --- /dev/null +++ b/packages/components/nodes/vectorstores/Zep/Zep_Existing.ts @@ -0,0 +1,235 @@ +import { ICommonObject, INode, INodeData, INodeOutputsValue, INodeParams } from '../../../src/Interface' +import { ZepVectorStore, IZepConfig } from 'langchain/vectorstores/zep' +import { Embeddings } from 'langchain/embeddings/base' +import { Document } from 'langchain/document' +import { getBaseClasses, getCredentialData, getCredentialParam } from '../../../src/utils' +import { IDocument, ZepClient } from '@getzep/zep-js' + +class Zep_Existing_VectorStores implements INode { + label: string + name: string + version: number + description: string + type: string + icon: string + category: string + baseClasses: string[] + inputs: INodeParams[] + credential: INodeParams + outputs: INodeOutputsValue[] + + constructor() { + this.label = 'Zep Load Existing Index' + this.name = 'zepExistingIndex' + this.version = 1.0 + this.type = 'Zep' + this.icon = 'zep.png' + this.category = 'Vector Stores' + this.description = 'Load existing index from Zep (i.e: Document has been upserted)' + this.baseClasses = [this.type, 'VectorStoreRetriever', 'BaseRetriever'] + this.credential = { + label: 'Connect Credential', + name: 'credential', + type: 'credential', + optional: true, + description: 'Configure JWT authentication on your Zep instance (Optional)', + credentialNames: ['zepMemoryApi'] + } + this.inputs = [ + { + label: 'Embeddings', + name: 'embeddings', + type: 'Embeddings' + }, + { + label: 'Base URL', + name: 'baseURL', + type: 'string', + default: 'http://127.0.0.1:8000' + }, + { + label: 'Zep Collection', + name: 'zepCollection', + type: 'string', + placeholder: 'my-first-collection' + }, + { + label: 'Zep Metadata Filter', + name: 'zepMetadataFilter', + type: 'json', + optional: true, + additionalParams: true + }, + { + label: 'Embedding Dimension', + name: 'dimension', + type: 'number', + default: 1536, + additionalParams: true + }, + { + label: 'Top K', + name: 'topK', + description: 'Number of top results to fetch. Default to 4', + placeholder: '4', + type: 'number', + additionalParams: true, + optional: true + } + ] + this.outputs = [ + { + label: 'Pinecone Retriever', + name: 'retriever', + baseClasses: this.baseClasses + }, + { + label: 'Pinecone Vector Store', + name: 'vectorStore', + baseClasses: [this.type, ...getBaseClasses(ZepVectorStore)] + } + ] + } + + async init(nodeData: INodeData, _: string, options: ICommonObject): Promise { + const baseURL = nodeData.inputs?.baseURL as string + const zepCollection = nodeData.inputs?.zepCollection as string + const zepMetadataFilter = nodeData.inputs?.zepMetadataFilter + const dimension = nodeData.inputs?.dimension as number + const embeddings = nodeData.inputs?.embeddings as Embeddings + const output = nodeData.outputs?.output as string + const topK = nodeData.inputs?.topK as string + const k = topK ? parseFloat(topK) : 4 + + const credentialData = await getCredentialData(nodeData.credential ?? '', options) + const apiKey = getCredentialParam('apiKey', credentialData, nodeData) + + const zepConfig: IZepConfig & Partial = { + apiUrl: baseURL, + collectionName: zepCollection, + embeddingDimensions: dimension, + isAutoEmbedded: false + } + if (apiKey) zepConfig.apiKey = apiKey + if (zepMetadataFilter) { + const metadatafilter = typeof zepMetadataFilter === 'object' ? zepMetadataFilter : JSON.parse(zepMetadataFilter) + zepConfig.filter = metadatafilter + } + + const vectorStore = await ZepExistingVS.fromExistingIndex(embeddings, zepConfig) + + if (output === 'retriever') { + const retriever = vectorStore.asRetriever(k) + return retriever + } else if (output === 'vectorStore') { + ;(vectorStore as any).k = k + return vectorStore + } + return vectorStore + } +} + +interface ZepFilter { + filter: Record +} + +function zepDocsToDocumentsAndScore(results: IDocument[]): [Document, number][] { + return results.map((d) => [ + new Document({ + pageContent: d.content, + metadata: d.metadata + }), + d.score ? d.score : 0 + ]) +} + +function assignMetadata(value: string | Record | object | undefined): Record | undefined { + if (typeof value === 'object' && value !== null) { + return value as Record + } + if (value !== undefined) { + console.warn('Metadata filters must be an object, Record, or undefined.') + } + return undefined +} + +class ZepExistingVS extends ZepVectorStore { + filter?: Record + args?: IZepConfig & Partial + + constructor(embeddings: Embeddings, args: IZepConfig & Partial) { + super(embeddings, args) + this.filter = args.filter + this.args = args + } + + async initalizeCollection(args: IZepConfig & Partial) { + this.client = await ZepClient.init(args.apiUrl, args.apiKey) + try { + this.collection = await this.client.document.getCollection(args.collectionName) + } catch (err) { + if (err instanceof Error) { + if (err.name === 'NotFoundError') { + await this.createNewCollection(args) + } else { + throw err + } + } + } + } + + async createNewCollection(args: IZepConfig & Partial) { + if (!args.embeddingDimensions) { + throw new Error( + `Collection ${args.collectionName} not found. You can create a new Collection by providing embeddingDimensions.` + ) + } + + this.collection = await this.client.document.addCollection({ + name: args.collectionName, + description: args.description, + metadata: args.metadata, + embeddingDimensions: args.embeddingDimensions, + isAutoEmbedded: false + }) + } + + async similaritySearchVectorWithScore( + query: number[], + k: number, + filter?: Record | undefined + ): Promise<[Document, number][]> { + if (filter && this.filter) { + throw new Error('cannot provide both `filter` and `this.filter`') + } + const _filters = filter ?? this.filter + const ANDFilters = [] + for (const filterKey in _filters) { + let filterVal = _filters[filterKey] + if (typeof filterVal === 'string') filterVal = `"${filterVal}"` + ANDFilters.push({ jsonpath: `$[*] ? (@.${filterKey} == ${filterVal})` }) + } + const newfilter = { + where: { and: ANDFilters } + } + await this.initalizeCollection(this.args!).catch((err) => { + console.error('Error initializing collection:', err) + throw err + }) + const results = await this.collection.search( + { + embedding: new Float32Array(query), + metadata: assignMetadata(newfilter) + }, + k + ) + return zepDocsToDocumentsAndScore(results) + } + + static async fromExistingIndex(embeddings: Embeddings, dbConfig: IZepConfig & Partial): Promise { + const instance = new this(embeddings, dbConfig) + return instance + } +} + +module.exports = { nodeClass: Zep_Existing_VectorStores } diff --git a/packages/components/nodes/vectorstores/Zep/Zep_Upsert.ts b/packages/components/nodes/vectorstores/Zep/Zep_Upsert.ts new file mode 100644 index 00000000..0f976d2b --- /dev/null +++ b/packages/components/nodes/vectorstores/Zep/Zep_Upsert.ts @@ -0,0 +1,133 @@ +import { ICommonObject, INode, INodeData, INodeOutputsValue, INodeParams } from '../../../src/Interface' +import { ZepVectorStore, IZepConfig } from 'langchain/vectorstores/zep' +import { Embeddings } from 'langchain/embeddings/base' +import { Document } from 'langchain/document' +import { getBaseClasses, getCredentialData, getCredentialParam } from '../../../src/utils' +import { flatten } from 'lodash' + +class Zep_Upsert_VectorStores implements INode { + label: string + name: string + version: number + description: string + type: string + icon: string + category: string + baseClasses: string[] + inputs: INodeParams[] + credential: INodeParams + outputs: INodeOutputsValue[] + + constructor() { + this.label = 'Zep Upsert Document' + this.name = 'zepUpsert' + this.version = 1.0 + this.type = 'Zep' + this.icon = 'zep.png' + this.category = 'Vector Stores' + this.description = 'Upsert documents to Zep' + this.baseClasses = [this.type, 'VectorStoreRetriever', 'BaseRetriever'] + this.credential = { + label: 'Connect Credential', + name: 'credential', + type: 'credential', + optional: true, + description: 'Configure JWT authentication on your Zep instance (Optional)', + credentialNames: ['zepMemoryApi'] + } + this.inputs = [ + { + label: 'Document', + name: 'document', + type: 'Document', + list: true + }, + { + label: 'Embeddings', + name: 'embeddings', + type: 'Embeddings' + }, + { + label: 'Base URL', + name: 'baseURL', + type: 'string', + default: 'http://127.0.0.1:8000' + }, + { + label: 'Zep Collection', + name: 'zepCollection', + type: 'string', + placeholder: 'my-first-collection' + }, + { + label: 'Embedding Dimension', + name: 'dimension', + type: 'number', + default: 1536, + additionalParams: true + }, + { + label: 'Top K', + name: 'topK', + description: 'Number of top results to fetch. Default to 4', + placeholder: '4', + type: 'number', + additionalParams: true, + optional: true + } + ] + this.outputs = [ + { + label: 'Zep Retriever', + name: 'retriever', + baseClasses: this.baseClasses + }, + { + label: 'Zep Vector Store', + name: 'vectorStore', + baseClasses: [this.type, ...getBaseClasses(ZepVectorStore)] + } + ] + } + + async init(nodeData: INodeData, _: string, options: ICommonObject): Promise { + const baseURL = nodeData.inputs?.baseURL as string + const zepCollection = nodeData.inputs?.zepCollection as string + const dimension = (nodeData.inputs?.dimension as number) ?? 1536 + const docs = nodeData.inputs?.document as Document[] + const embeddings = nodeData.inputs?.embeddings as Embeddings + const topK = nodeData.inputs?.topK as string + const k = topK ? parseFloat(topK) : 4 + const output = nodeData.outputs?.output as string + + const credentialData = await getCredentialData(nodeData.credential ?? '', options) + const apiKey = getCredentialParam('apiKey', credentialData, nodeData) + + const flattenDocs = docs && docs.length ? flatten(docs) : [] + const finalDocs = [] + for (let i = 0; i < flattenDocs.length; i += 1) { + finalDocs.push(new Document(flattenDocs[i])) + } + + const zepConfig: IZepConfig = { + apiUrl: baseURL, + collectionName: zepCollection, + embeddingDimensions: dimension, + isAutoEmbedded: false + } + if (apiKey) zepConfig.apiKey = apiKey + + const vectorStore = await ZepVectorStore.fromDocuments(finalDocs, embeddings, zepConfig) + + if (output === 'retriever') { + const retriever = vectorStore.asRetriever(k) + return retriever + } else if (output === 'vectorStore') { + ;(vectorStore as any).k = k + return vectorStore + } + return vectorStore + } +} + +module.exports = { nodeClass: Zep_Upsert_VectorStores } diff --git a/packages/components/nodes/vectorstores/Zep/zep.png b/packages/components/nodes/vectorstores/Zep/zep.png new file mode 100644 index 00000000..2fdb2382 Binary files /dev/null and b/packages/components/nodes/vectorstores/Zep/zep.png differ diff --git a/packages/components/package.json b/packages/components/package.json index c48cba73..4653f744 100644 --- a/packages/components/package.json +++ b/packages/components/package.json @@ -18,7 +18,7 @@ "dependencies": { "@aws-sdk/client-dynamodb": "^3.360.0", "@dqbd/tiktoken": "^1.0.7", - "@getzep/zep-js": "^0.4.1", + "@getzep/zep-js": "^0.6.3", "@huggingface/inference": "^2.6.1", "@notionhq/client": "^2.2.8", "@opensearch-project/opensearch": "^1.2.0", @@ -28,6 +28,7 @@ "@types/js-yaml": "^4.0.5", "apify-client": "^2.7.1", "@types/jsdom": "^21.1.1", + "@zilliz/milvus2-sdk-node": "^2.2.24", "axios": "^0.27.2", "cheerio": "^1.0.0-rc.12", "chromadb": "^1.5.3", @@ -40,7 +41,7 @@ "google-auth-library": "^9.0.0", "graphql": "^16.6.0", "html-to-text": "^9.0.5", - "langchain": "^0.0.122", + "langchain": "^0.0.128", "linkifyjs": "^4.1.1", "mammoth": "^1.5.1", "moment": "^2.29.3", diff --git a/packages/components/src/utils.ts b/packages/components/src/utils.ts index bcca834a..8d06a650 100644 --- a/packages/components/src/utils.ts +++ b/packages/components/src/utils.ts @@ -520,3 +520,22 @@ export const mapChatHistory = (options: ICommonObject): ChatMessageHistory => { } return new ChatMessageHistory(chatHistory) } + +/** + * Convert incoming chat history to string + * @param {IMessage[]} chatHistory + * @returns {string} + */ +export const convertChatHistoryToText = (chatHistory: IMessage[]): string => { + return chatHistory + .map((chatMessage) => { + if (chatMessage.type === 'apiMessage') { + return `Assistant: ${chatMessage.message}` + } else if (chatMessage.type === 'userMessage') { + return `Human: ${chatMessage.message}` + } else { + return `${chatMessage.message}` + } + }) + .join('\n') +} diff --git a/packages/server/marketplaces/chatflows/Prompt Chaining with VectorStore.json b/packages/server/marketplaces/chatflows/Prompt Chaining with VectorStore.json new file mode 100644 index 00000000..9d6838eb --- /dev/null +++ b/packages/server/marketplaces/chatflows/Prompt Chaining with VectorStore.json @@ -0,0 +1,966 @@ +{ + "description": "Use chat history to rephrase user question, and answer the rephrased question using retrieved docs from vector store", + "nodes": [ + { + "width": 300, + "height": 503, + "id": "pineconeExistingIndex_0", + "position": { + "x": 1062.7418678410986, + "y": -109.27680365777141 + }, + "type": "customNode", + "data": { + "id": "pineconeExistingIndex_0", + "label": "Pinecone Load Existing Index", + "version": 1, + "name": "pineconeExistingIndex", + "type": "Pinecone", + "baseClasses": ["Pinecone", "VectorStoreRetriever", "BaseRetriever"], + "category": "Vector Stores", + "description": "Load existing index from Pinecone (i.e: Document has been upserted)", + "inputParams": [ + { + "label": "Connect Credential", + "name": "credential", + "type": "credential", + "credentialNames": ["pineconeApi"], + "id": "pineconeExistingIndex_0-input-credential-credential" + }, + { + "label": "Pinecone Index", + "name": "pineconeIndex", + "type": "string", + "id": "pineconeExistingIndex_0-input-pineconeIndex-string" + }, + { + "label": "Pinecone Namespace", + "name": "pineconeNamespace", + "type": "string", + "placeholder": "my-first-namespace", + "additionalParams": true, + "optional": true, + "id": "pineconeExistingIndex_0-input-pineconeNamespace-string" + }, + { + "label": "Pinecone Metadata Filter", + "name": "pineconeMetadataFilter", + "type": "json", + "optional": true, + "additionalParams": true, + "id": "pineconeExistingIndex_0-input-pineconeMetadataFilter-json" + }, + { + "label": "Top K", + "name": "topK", + "description": "Number of top results to fetch. Default to 4", + "placeholder": "4", + "type": "number", + "additionalParams": true, + "optional": true, + "id": "pineconeExistingIndex_0-input-topK-number" + } + ], + "inputAnchors": [ + { + "label": "Embeddings", + "name": "embeddings", + "type": "Embeddings", + "id": "pineconeExistingIndex_0-input-embeddings-Embeddings" + } + ], + "inputs": { + "embeddings": "{{openAIEmbeddings_0.data.instance}}", + "pineconeIndex": "newindex", + "pineconeNamespace": "", + "pineconeMetadataFilter": "{}", + "topK": "" + }, + "outputAnchors": [ + { + "name": "output", + "label": "Output", + "type": "options", + "options": [ + { + "id": "pineconeExistingIndex_0-output-retriever-Pinecone|VectorStoreRetriever|BaseRetriever", + "name": "retriever", + "label": "Pinecone Retriever", + "type": "Pinecone | VectorStoreRetriever | BaseRetriever" + }, + { + "id": "pineconeExistingIndex_0-output-vectorStore-Pinecone|VectorStore", + "name": "vectorStore", + "label": "Pinecone Vector Store", + "type": "Pinecone | VectorStore" + } + ], + "default": "retriever" + } + ], + "outputs": { + "output": "vectorStore" + }, + "selected": false + }, + "selected": false, + "positionAbsolute": { + "x": 1062.7418678410986, + "y": -109.27680365777141 + }, + "dragging": false + }, + { + "width": 300, + "height": 327, + "id": "openAIEmbeddings_0", + "position": { + "x": 711.3971966563331, + "y": 7.7184225021727 + }, + "type": "customNode", + "data": { + "id": "openAIEmbeddings_0", + "label": "OpenAI Embeddings", + "version": 1, + "name": "openAIEmbeddings", + "type": "OpenAIEmbeddings", + "baseClasses": ["OpenAIEmbeddings", "Embeddings"], + "category": "Embeddings", + "description": "OpenAI API to generate embeddings for a given text", + "inputParams": [ + { + "label": "Connect Credential", + "name": "credential", + "type": "credential", + "credentialNames": ["openAIApi"], + "id": "openAIEmbeddings_0-input-credential-credential" + }, + { + "label": "Strip New Lines", + "name": "stripNewLines", + "type": "boolean", + "optional": true, + "additionalParams": true, + "id": "openAIEmbeddings_0-input-stripNewLines-boolean" + }, + { + "label": "Batch Size", + "name": "batchSize", + "type": "number", + "optional": true, + "additionalParams": true, + "id": "openAIEmbeddings_0-input-batchSize-number" + }, + { + "label": "Timeout", + "name": "timeout", + "type": "number", + "optional": true, + "additionalParams": true, + "id": "openAIEmbeddings_0-input-timeout-number" + }, + { + "label": "BasePath", + "name": "basepath", + "type": "string", + "optional": true, + "additionalParams": true, + "id": "openAIEmbeddings_0-input-basepath-string" + } + ], + "inputAnchors": [], + "inputs": { + "stripNewLines": "", + "batchSize": "", + "timeout": "", + "basepath": "" + }, + "outputAnchors": [ + { + "id": "openAIEmbeddings_0-output-openAIEmbeddings-OpenAIEmbeddings|Embeddings", + "name": "openAIEmbeddings", + "label": "OpenAIEmbeddings", + "type": "OpenAIEmbeddings | Embeddings" + } + ], + "outputs": {}, + "selected": false + }, + "selected": false, + "positionAbsolute": { + "x": 711.3971966563331, + "y": 7.7184225021727 + }, + "dragging": false + }, + { + "width": 300, + "height": 473, + "id": "promptTemplate_0", + "position": { + "x": 348.2881107399286, + "y": -97.74510214137423 + }, + "type": "customNode", + "data": { + "id": "promptTemplate_0", + "label": "Prompt Template", + "version": 1, + "name": "promptTemplate", + "type": "PromptTemplate", + "baseClasses": ["PromptTemplate", "BaseStringPromptTemplate", "BasePromptTemplate", "Runnable"], + "category": "Prompts", + "description": "Schema to represent a basic prompt for an LLM", + "inputParams": [ + { + "label": "Template", + "name": "template", + "type": "string", + "rows": 4, + "placeholder": "What is a good name for a company that makes {product}?", + "id": "promptTemplate_0-input-template-string" + }, + { + "label": "Format Prompt Values", + "name": "promptValues", + "type": "json", + "optional": true, + "acceptVariable": true, + "list": true, + "id": "promptTemplate_0-input-promptValues-json" + } + ], + "inputAnchors": [], + "inputs": { + "template": "Given the following conversation and a follow up question, rephrase the follow up question to be a standalone question.\n\nChat History:\n{chat_history}\nFollow Up Input: {question}\nStandalone question:", + "promptValues": "{\"question\":\"{{question}}\",\"chat_history\":\"{{chat_history}}\"}" + }, + "outputAnchors": [ + { + "id": "promptTemplate_0-output-promptTemplate-PromptTemplate|BaseStringPromptTemplate|BasePromptTemplate|Runnable", + "name": "promptTemplate", + "label": "PromptTemplate", + "type": "PromptTemplate | BaseStringPromptTemplate | BasePromptTemplate | Runnable" + } + ], + "outputs": {}, + "selected": false + }, + "selected": false, + "positionAbsolute": { + "x": 348.2881107399286, + "y": -97.74510214137423 + }, + "dragging": false + }, + { + "width": 300, + "height": 522, + "id": "chatOpenAI_0", + "position": { + "x": 335.7621848973805, + "y": -651.7411273245009 + }, + "type": "customNode", + "data": { + "id": "chatOpenAI_0", + "label": "ChatOpenAI", + "version": 1, + "name": "chatOpenAI", + "type": "ChatOpenAI", + "baseClasses": ["ChatOpenAI", "BaseChatModel", "BaseLanguageModel", "Runnable"], + "category": "Chat Models", + "description": "Wrapper around OpenAI large language models that use the Chat endpoint", + "inputParams": [ + { + "label": "Connect Credential", + "name": "credential", + "type": "credential", + "credentialNames": ["openAIApi"], + "id": "chatOpenAI_0-input-credential-credential" + }, + { + "label": "Model Name", + "name": "modelName", + "type": "options", + "options": [ + { + "label": "gpt-4", + "name": "gpt-4" + }, + { + "label": "gpt-4-0613", + "name": "gpt-4-0613" + }, + { + "label": "gpt-4-32k", + "name": "gpt-4-32k" + }, + { + "label": "gpt-4-32k-0613", + "name": "gpt-4-32k-0613" + }, + { + "label": "gpt-3.5-turbo", + "name": "gpt-3.5-turbo" + }, + { + "label": "gpt-3.5-turbo-0613", + "name": "gpt-3.5-turbo-0613" + }, + { + "label": "gpt-3.5-turbo-16k", + "name": "gpt-3.5-turbo-16k" + }, + { + "label": "gpt-3.5-turbo-16k-0613", + "name": "gpt-3.5-turbo-16k-0613" + } + ], + "default": "gpt-3.5-turbo", + "optional": true, + "id": "chatOpenAI_0-input-modelName-options" + }, + { + "label": "Temperature", + "name": "temperature", + "type": "number", + "step": 0.1, + "default": 0.9, + "optional": true, + "id": "chatOpenAI_0-input-temperature-number" + }, + { + "label": "Max Tokens", + "name": "maxTokens", + "type": "number", + "step": 1, + "optional": true, + "additionalParams": true, + "id": "chatOpenAI_0-input-maxTokens-number" + }, + { + "label": "Top Probability", + "name": "topP", + "type": "number", + "step": 0.1, + "optional": true, + "additionalParams": true, + "id": "chatOpenAI_0-input-topP-number" + }, + { + "label": "Frequency Penalty", + "name": "frequencyPenalty", + "type": "number", + "step": 0.1, + "optional": true, + "additionalParams": true, + "id": "chatOpenAI_0-input-frequencyPenalty-number" + }, + { + "label": "Presence Penalty", + "name": "presencePenalty", + "type": "number", + "step": 0.1, + "optional": true, + "additionalParams": true, + "id": "chatOpenAI_0-input-presencePenalty-number" + }, + { + "label": "Timeout", + "name": "timeout", + "type": "number", + "step": 1, + "optional": true, + "additionalParams": true, + "id": "chatOpenAI_0-input-timeout-number" + }, + { + "label": "BasePath", + "name": "basepath", + "type": "string", + "optional": true, + "additionalParams": true, + "id": "chatOpenAI_0-input-basepath-string" + } + ], + "inputAnchors": [], + "inputs": { + "modelName": "gpt-3.5-turbo-16k", + "temperature": 0.9, + "maxTokens": "", + "topP": "", + "frequencyPenalty": "", + "presencePenalty": "", + "timeout": "", + "basepath": "" + }, + "outputAnchors": [ + { + "id": "chatOpenAI_0-output-chatOpenAI-ChatOpenAI|BaseChatModel|BaseLanguageModel|Runnable", + "name": "chatOpenAI", + "label": "ChatOpenAI", + "type": "ChatOpenAI | BaseChatModel | BaseLanguageModel | Runnable" + } + ], + "outputs": {}, + "selected": false + }, + "selected": false, + "dragging": false, + "positionAbsolute": { + "x": 335.7621848973805, + "y": -651.7411273245009 + } + }, + { + "width": 300, + "height": 522, + "id": "chatOpenAI_1", + "position": { + "x": 1765.2801848172305, + "y": -667.9261054149061 + }, + "type": "customNode", + "data": { + "id": "chatOpenAI_1", + "label": "ChatOpenAI", + "version": 1, + "name": "chatOpenAI", + "type": "ChatOpenAI", + "baseClasses": ["ChatOpenAI", "BaseChatModel", "BaseLanguageModel", "Runnable"], + "category": "Chat Models", + "description": "Wrapper around OpenAI large language models that use the Chat endpoint", + "inputParams": [ + { + "label": "Connect Credential", + "name": "credential", + "type": "credential", + "credentialNames": ["openAIApi"], + "id": "chatOpenAI_1-input-credential-credential" + }, + { + "label": "Model Name", + "name": "modelName", + "type": "options", + "options": [ + { + "label": "gpt-4", + "name": "gpt-4" + }, + { + "label": "gpt-4-0613", + "name": "gpt-4-0613" + }, + { + "label": "gpt-4-32k", + "name": "gpt-4-32k" + }, + { + "label": "gpt-4-32k-0613", + "name": "gpt-4-32k-0613" + }, + { + "label": "gpt-3.5-turbo", + "name": "gpt-3.5-turbo" + }, + { + "label": "gpt-3.5-turbo-0613", + "name": "gpt-3.5-turbo-0613" + }, + { + "label": "gpt-3.5-turbo-16k", + "name": "gpt-3.5-turbo-16k" + }, + { + "label": "gpt-3.5-turbo-16k-0613", + "name": "gpt-3.5-turbo-16k-0613" + } + ], + "default": "gpt-3.5-turbo", + "optional": true, + "id": "chatOpenAI_1-input-modelName-options" + }, + { + "label": "Temperature", + "name": "temperature", + "type": "number", + "step": 0.1, + "default": 0.9, + "optional": true, + "id": "chatOpenAI_1-input-temperature-number" + }, + { + "label": "Max Tokens", + "name": "maxTokens", + "type": "number", + "step": 1, + "optional": true, + "additionalParams": true, + "id": "chatOpenAI_1-input-maxTokens-number" + }, + { + "label": "Top Probability", + "name": "topP", + "type": "number", + "step": 0.1, + "optional": true, + "additionalParams": true, + "id": "chatOpenAI_1-input-topP-number" + }, + { + "label": "Frequency Penalty", + "name": "frequencyPenalty", + "type": "number", + "step": 0.1, + "optional": true, + "additionalParams": true, + "id": "chatOpenAI_1-input-frequencyPenalty-number" + }, + { + "label": "Presence Penalty", + "name": "presencePenalty", + "type": "number", + "step": 0.1, + "optional": true, + "additionalParams": true, + "id": "chatOpenAI_1-input-presencePenalty-number" + }, + { + "label": "Timeout", + "name": "timeout", + "type": "number", + "step": 1, + "optional": true, + "additionalParams": true, + "id": "chatOpenAI_1-input-timeout-number" + }, + { + "label": "BasePath", + "name": "basepath", + "type": "string", + "optional": true, + "additionalParams": true, + "id": "chatOpenAI_1-input-basepath-string" + } + ], + "inputAnchors": [], + "inputs": { + "modelName": "gpt-3.5-turbo-16k", + "temperature": 0.9, + "maxTokens": "", + "topP": "", + "frequencyPenalty": "", + "presencePenalty": "", + "timeout": "", + "basepath": "" + }, + "outputAnchors": [ + { + "id": "chatOpenAI_1-output-chatOpenAI-ChatOpenAI|BaseChatModel|BaseLanguageModel|Runnable", + "name": "chatOpenAI", + "label": "ChatOpenAI", + "type": "ChatOpenAI | BaseChatModel | BaseLanguageModel | Runnable" + } + ], + "outputs": {}, + "selected": false + }, + "selected": false, + "dragging": false, + "positionAbsolute": { + "x": 1765.2801848172305, + "y": -667.9261054149061 + } + }, + { + "width": 300, + "height": 473, + "id": "promptTemplate_1", + "position": { + "x": 1773.720934090435, + "y": -116.71323227575395 + }, + "type": "customNode", + "data": { + "id": "promptTemplate_1", + "label": "Prompt Template", + "version": 1, + "name": "promptTemplate", + "type": "PromptTemplate", + "baseClasses": ["PromptTemplate", "BaseStringPromptTemplate", "BasePromptTemplate", "Runnable"], + "category": "Prompts", + "description": "Schema to represent a basic prompt for an LLM", + "inputParams": [ + { + "label": "Template", + "name": "template", + "type": "string", + "rows": 4, + "placeholder": "What is a good name for a company that makes {product}?", + "id": "promptTemplate_1-input-template-string" + }, + { + "label": "Format Prompt Values", + "name": "promptValues", + "type": "json", + "optional": true, + "acceptVariable": true, + "list": true, + "id": "promptTemplate_1-input-promptValues-json" + } + ], + "inputAnchors": [], + "inputs": { + "template": "Use the following pieces of context to answer the question at the end.\n\n{context}\n\nQuestion: {question}\nHelpful Answer:", + "promptValues": "{\"context\":\"{{vectorStoreToDocument_0.data.instance}}\",\"question\":\"{{llmChain_0.data.instance}}\"}" + }, + "outputAnchors": [ + { + "id": "promptTemplate_1-output-promptTemplate-PromptTemplate|BaseStringPromptTemplate|BasePromptTemplate|Runnable", + "name": "promptTemplate", + "label": "PromptTemplate", + "type": "PromptTemplate | BaseStringPromptTemplate | BasePromptTemplate | Runnable" + } + ], + "outputs": {}, + "selected": false + }, + "selected": false, + "positionAbsolute": { + "x": 1773.720934090435, + "y": -116.71323227575395 + }, + "dragging": false + }, + { + "width": 300, + "height": 404, + "id": "llmChain_0", + "position": { + "x": 756.1670091985342, + "y": -592.5151355056942 + }, + "type": "customNode", + "data": { + "id": "llmChain_0", + "label": "LLM Chain", + "version": 1, + "name": "llmChain", + "type": "LLMChain", + "baseClasses": ["LLMChain", "BaseChain", "Runnable"], + "category": "Chains", + "description": "Chain to run queries against LLMs", + "inputParams": [ + { + "label": "Chain Name", + "name": "chainName", + "type": "string", + "placeholder": "Name Your Chain", + "optional": true, + "id": "llmChain_0-input-chainName-string" + } + ], + "inputAnchors": [ + { + "label": "Language Model", + "name": "model", + "type": "BaseLanguageModel", + "id": "llmChain_0-input-model-BaseLanguageModel" + }, + { + "label": "Prompt", + "name": "prompt", + "type": "BasePromptTemplate", + "id": "llmChain_0-input-prompt-BasePromptTemplate" + } + ], + "inputs": { + "model": "{{chatOpenAI_0.data.instance}}", + "prompt": "{{promptTemplate_0.data.instance}}", + "chainName": "QuestionChain" + }, + "outputAnchors": [ + { + "name": "output", + "label": "Output", + "type": "options", + "options": [ + { + "id": "llmChain_0-output-llmChain-LLMChain|BaseChain|Runnable", + "name": "llmChain", + "label": "LLM Chain", + "type": "LLMChain | BaseChain | Runnable" + }, + { + "id": "llmChain_0-output-outputPrediction-string|json", + "name": "outputPrediction", + "label": "Output Prediction", + "type": "string | json" + } + ], + "default": "llmChain" + } + ], + "outputs": { + "output": "outputPrediction" + }, + "selected": false + }, + "selected": false, + "positionAbsolute": { + "x": 756.1670091985342, + "y": -592.5151355056942 + }, + "dragging": false + }, + { + "width": 300, + "height": 404, + "id": "llmChain_1", + "position": { + "x": 2200.1274896215496, + "y": -144.29167974642334 + }, + "type": "customNode", + "data": { + "id": "llmChain_1", + "label": "LLM Chain", + "version": 1, + "name": "llmChain", + "type": "LLMChain", + "baseClasses": ["LLMChain", "BaseChain", "Runnable"], + "category": "Chains", + "description": "Chain to run queries against LLMs", + "inputParams": [ + { + "label": "Chain Name", + "name": "chainName", + "type": "string", + "placeholder": "Name Your Chain", + "optional": true, + "id": "llmChain_1-input-chainName-string" + } + ], + "inputAnchors": [ + { + "label": "Language Model", + "name": "model", + "type": "BaseLanguageModel", + "id": "llmChain_1-input-model-BaseLanguageModel" + }, + { + "label": "Prompt", + "name": "prompt", + "type": "BasePromptTemplate", + "id": "llmChain_1-input-prompt-BasePromptTemplate" + } + ], + "inputs": { + "model": "{{chatOpenAI_1.data.instance}}", + "prompt": "{{promptTemplate_1.data.instance}}", + "chainName": "" + }, + "outputAnchors": [ + { + "name": "output", + "label": "Output", + "type": "options", + "options": [ + { + "id": "llmChain_1-output-llmChain-LLMChain|BaseChain|Runnable", + "name": "llmChain", + "label": "LLM Chain", + "type": "LLMChain | BaseChain | Runnable" + }, + { + "id": "llmChain_1-output-outputPrediction-string|json", + "name": "outputPrediction", + "label": "Output Prediction", + "type": "string | json" + } + ], + "default": "llmChain" + } + ], + "outputs": { + "output": "llmChain" + }, + "selected": false + }, + "selected": false, + "positionAbsolute": { + "x": 2200.1274896215496, + "y": -144.29167974642334 + }, + "dragging": false + }, + { + "width": 300, + "height": 353, + "id": "vectorStoreToDocument_0", + "position": { + "x": 1407.7038120189868, + "y": -26.16468811205081 + }, + "type": "customNode", + "data": { + "id": "vectorStoreToDocument_0", + "label": "VectorStore To Document", + "version": 1, + "name": "vectorStoreToDocument", + "type": "Document", + "baseClasses": ["Document"], + "category": "Document Loaders", + "description": "Search documents with scores from vector store", + "inputParams": [ + { + "label": "Minimum Score (%)", + "name": "minScore", + "type": "number", + "optional": true, + "placeholder": "75", + "step": 1, + "description": "Minumum score for embeddings documents to be included", + "id": "vectorStoreToDocument_0-input-minScore-number" + } + ], + "inputAnchors": [ + { + "label": "Vector Store", + "name": "vectorStore", + "type": "VectorStore", + "id": "vectorStoreToDocument_0-input-vectorStore-VectorStore" + } + ], + "inputs": { + "vectorStore": "{{pineconeExistingIndex_0.data.instance}}", + "minScore": "" + }, + "outputAnchors": [ + { + "name": "output", + "label": "Output", + "type": "options", + "options": [ + { + "id": "vectorStoreToDocument_0-output-document-Document", + "name": "document", + "label": "Document", + "type": "Document" + }, + { + "id": "vectorStoreToDocument_0-output-text-string|json", + "name": "text", + "label": "Text", + "type": "string | json" + } + ], + "default": "document" + } + ], + "outputs": { + "output": "text" + }, + "selected": false + }, + "selected": false, + "positionAbsolute": { + "x": 1407.7038120189868, + "y": -26.16468811205081 + }, + "dragging": false + } + ], + "edges": [ + { + "source": "openAIEmbeddings_0", + "sourceHandle": "openAIEmbeddings_0-output-openAIEmbeddings-OpenAIEmbeddings|Embeddings", + "target": "pineconeExistingIndex_0", + "targetHandle": "pineconeExistingIndex_0-input-embeddings-Embeddings", + "type": "buttonedge", + "id": "openAIEmbeddings_0-openAIEmbeddings_0-output-openAIEmbeddings-OpenAIEmbeddings|Embeddings-pineconeExistingIndex_0-pineconeExistingIndex_0-input-embeddings-Embeddings", + "data": { + "label": "" + } + }, + { + "source": "pineconeExistingIndex_0", + "sourceHandle": "pineconeExistingIndex_0-output-vectorStore-Pinecone|VectorStore", + "target": "vectorStoreToDocument_0", + "targetHandle": "vectorStoreToDocument_0-input-vectorStore-VectorStore", + "type": "buttonedge", + "id": "pineconeExistingIndex_0-pineconeExistingIndex_0-output-vectorStore-Pinecone|VectorStore-vectorStoreToDocument_0-vectorStoreToDocument_0-input-vectorStore-VectorStore", + "data": { + "label": "" + } + }, + { + "source": "vectorStoreToDocument_0", + "sourceHandle": "vectorStoreToDocument_0-output-text-string|json", + "target": "promptTemplate_1", + "targetHandle": "promptTemplate_1-input-promptValues-json", + "type": "buttonedge", + "id": "vectorStoreToDocument_0-vectorStoreToDocument_0-output-text-string|json-promptTemplate_1-promptTemplate_1-input-promptValues-json", + "data": { + "label": "" + } + }, + { + "source": "chatOpenAI_0", + "sourceHandle": "chatOpenAI_0-output-chatOpenAI-ChatOpenAI|BaseChatModel|BaseLanguageModel|Runnable", + "target": "llmChain_0", + "targetHandle": "llmChain_0-input-model-BaseLanguageModel", + "type": "buttonedge", + "id": "chatOpenAI_0-chatOpenAI_0-output-chatOpenAI-ChatOpenAI|BaseChatModel|BaseLanguageModel|Runnable-llmChain_0-llmChain_0-input-model-BaseLanguageModel", + "data": { + "label": "" + } + }, + { + "source": "promptTemplate_0", + "sourceHandle": "promptTemplate_0-output-promptTemplate-PromptTemplate|BaseStringPromptTemplate|BasePromptTemplate|Runnable", + "target": "llmChain_0", + "targetHandle": "llmChain_0-input-prompt-BasePromptTemplate", + "type": "buttonedge", + "id": "promptTemplate_0-promptTemplate_0-output-promptTemplate-PromptTemplate|BaseStringPromptTemplate|BasePromptTemplate|Runnable-llmChain_0-llmChain_0-input-prompt-BasePromptTemplate", + "data": { + "label": "" + } + }, + { + "source": "llmChain_0", + "sourceHandle": "llmChain_0-output-outputPrediction-string|json", + "target": "promptTemplate_1", + "targetHandle": "promptTemplate_1-input-promptValues-json", + "type": "buttonedge", + "id": "llmChain_0-llmChain_0-output-outputPrediction-string|json-promptTemplate_1-promptTemplate_1-input-promptValues-json", + "data": { + "label": "" + } + }, + { + "source": "chatOpenAI_1", + "sourceHandle": "chatOpenAI_1-output-chatOpenAI-ChatOpenAI|BaseChatModel|BaseLanguageModel|Runnable", + "target": "llmChain_1", + "targetHandle": "llmChain_1-input-model-BaseLanguageModel", + "type": "buttonedge", + "id": "chatOpenAI_1-chatOpenAI_1-output-chatOpenAI-ChatOpenAI|BaseChatModel|BaseLanguageModel|Runnable-llmChain_1-llmChain_1-input-model-BaseLanguageModel", + "data": { + "label": "" + } + }, + { + "source": "promptTemplate_1", + "sourceHandle": "promptTemplate_1-output-promptTemplate-PromptTemplate|BaseStringPromptTemplate|BasePromptTemplate|Runnable", + "target": "llmChain_1", + "targetHandle": "llmChain_1-input-prompt-BasePromptTemplate", + "type": "buttonedge", + "id": "promptTemplate_1-promptTemplate_1-output-promptTemplate-PromptTemplate|BaseStringPromptTemplate|BasePromptTemplate|Runnable-llmChain_1-llmChain_1-input-prompt-BasePromptTemplate", + "data": { + "label": "" + } + } + ] +} diff --git a/packages/server/marketplaces/chatflows/Vectara LLM Chain Upload.json b/packages/server/marketplaces/chatflows/Vectara LLM Chain Upload.json index 0758ec9a..784ad240 100644 --- a/packages/server/marketplaces/chatflows/Vectara LLM Chain Upload.json +++ b/packages/server/marketplaces/chatflows/Vectara LLM Chain Upload.json @@ -1,186 +1,11 @@ { "description": "A simple LLM chain that uses Vectara to enable conversations with uploaded documents", "nodes": [ - { - "width": 300, - "height": 408, - "id": "vectaraUpsert_0", - "position": { "x": 438, "y": 214 }, - "type": "customNode", - "data": { - "id": "vectaraUpsert_0", - "label": "Vectara Upsert Document", - "version": 1, - "name": "vectaraUpsert", - "type": "Vectara", - "baseClasses": ["Vectara", "VectorStoreRetriever", "BaseRetriever"], - "category": "Vector Stores", - "description": "Upsert documents to Vectara", - "inputParams": [ - { - "label": "Connect Credential", - "name": "credential", - "type": "credential", - "credentialNames": ["vectaraApi"], - "id": "vectaraUpsert_0-input-credential-credential" - }, - { - "label": "Filter", - "name": "filter", - "type": "json", - "additionalParams": true, - "optional": true, - "id": "vectaraUpsert_0-input-filter-json" - }, - { - "label": "Lambda", - "name": "lambda", - "type": "number", - "additionalParams": true, - "optional": true, - "id": "vectaraUpsert_0-input-lambda-number" - }, - { - "label": "Top K", - "name": "topK", - "description": "Number of top results to fetch. Defaults to 4", - "placeholder": "4", - "type": "number", - "additionalParams": true, - "optional": true, - "id": "vectaraUpsert_0-input-topK-number" - } - ], - "inputAnchors": [ - { - "label": "Document", - "name": "document", - "type": "Document", - "list": true, - "id": "vectaraUpsert_0-input-document-Document" - } - ], - "inputs": { - "document": ["{{pdfFile_0.data.instance}}"], - "filter": "", - "lambda": "", - "topK": "" - }, - "outputAnchors": [ - { - "name": "output", - "label": "Output", - "type": "options", - "options": [ - { - "id": "vectaraUpsert_0-output-retriever-Vectara|VectorStoreRetriever|BaseRetriever", - "name": "retriever", - "label": "Vectara Retriever", - "type": "Vectara | VectorStoreRetriever | BaseRetriever" - }, - { - "id": "vectaraUpsert_0-output-vectorStore-Vectara|VectorStore", - "name": "vectorStore", - "label": "Vectara Vector Store", - "type": "Vectara | VectorStore" - } - ], - "default": "retriever" - } - ], - "outputs": { "output": "retriever" }, - "selected": false - }, - "selected": false, - "dragging": false, - "positionAbsolute": { "x": 438, "y": 214 } - }, - { - "width": 300, - "height": 509, - "id": "pdfFile_0", - "position": { "x": 68.3013317598369, "y": 199.60454731299677 }, - "type": "customNode", - "data": { - "id": "pdfFile_0", - "label": "Pdf File", - "version": 1, - "name": "pdfFile", - "type": "Document", - "baseClasses": ["Document"], - "category": "Document Loaders", - "description": "Load data from PDF files", - "inputParams": [ - { - "label": "Pdf File", - "name": "pdfFile", - "type": "file", - "fileType": ".pdf", - "id": "pdfFile_0-input-pdfFile-file" - }, - { - "label": "Usage", - "name": "usage", - "type": "options", - "options": [ - { "label": "One document per page", "name": "perPage" }, - { "label": "One document per file", "name": "perFile" } - ], - "default": "perPage", - "id": "pdfFile_0-input-usage-options" - }, - { - "label": "Use Legacy Build", - "name": "legacyBuild", - "type": "boolean", - "optional": true, - "additionalParams": true, - "id": "pdfFile_0-input-legacyBuild-boolean" - }, - { - "label": "Metadata", - "name": "metadata", - "type": "json", - "optional": true, - "additionalParams": true, - "id": "pdfFile_0-input-metadata-json" - } - ], - "inputAnchors": [ - { - "label": "Text Splitter", - "name": "textSplitter", - "type": "TextSplitter", - "optional": true, - "id": "pdfFile_0-input-textSplitter-TextSplitter" - } - ], - "inputs": { - "textSplitter": "", - "usage": "perPage", - "legacyBuild": "", - "metadata": "" - }, - "outputAnchors": [ - { - "id": "pdfFile_0-output-pdfFile-Document", - "name": "pdfFile", - "label": "Document", - "type": "Document" - } - ], - "outputs": {}, - "selected": false - }, - "selected": false, - "positionAbsolute": { "x": 68.3013317598369, "y": 199.60454731299677 }, - "dragging": false - }, { "width": 300, "height": 525, "id": "chatOpenAI_0", - "position": { "x": 804.3889791707068, "y": 195.11620799951592 }, + "position": { "x": 514.1088940275924, "y": 199.574479681537 }, "type": "customNode", "data": { "id": "chatOpenAI_0", @@ -211,10 +36,7 @@ { "label": "gpt-3.5-turbo", "name": "gpt-3.5-turbo" }, { "label": "gpt-3.5-turbo-0613", "name": "gpt-3.5-turbo-0613" }, { "label": "gpt-3.5-turbo-16k", "name": "gpt-3.5-turbo-16k" }, - { - "label": "gpt-3.5-turbo-16k-0613", - "name": "gpt-3.5-turbo-16k-0613" - } + { "label": "gpt-3.5-turbo-16k-0613", "name": "gpt-3.5-turbo-16k-0613" } ], "default": "gpt-3.5-turbo", "optional": true, @@ -286,7 +108,7 @@ "inputAnchors": [], "inputs": { "modelName": "gpt-3.5-turbo", - "temperature": "0.2", + "temperature": "0.5", "maxTokens": "", "topP": "", "frequencyPenalty": "", @@ -306,14 +128,14 @@ "selected": false }, "selected": false, - "positionAbsolute": { "x": 804.3889791707068, "y": 195.11620799951592 }, + "positionAbsolute": { "x": 514.1088940275924, "y": 199.574479681537 }, "dragging": false }, { "width": 300, "height": 481, "id": "conversationalRetrievalQAChain_0", - "position": { "x": 1160.4877473512795, "y": 259.2799138505109 }, + "position": { "x": 900.4793407261002, "y": 205.9476004518217 }, "type": "customNode", "data": { "id": "conversationalRetrievalQAChain_0", @@ -410,11 +232,200 @@ "selected": false }, "selected": false, - "positionAbsolute": { "x": 1160.4877473512795, "y": 259.2799138505109 }, + "positionAbsolute": { "x": 900.4793407261002, "y": 205.9476004518217 }, "dragging": false + }, + { + "width": 300, + "height": 509, + "id": "pdfFile_0", + "position": { "x": -210.44158723479913, "y": 236.6627524951051 }, + "type": "customNode", + "data": { + "id": "pdfFile_0", + "label": "Pdf File", + "version": 1, + "name": "pdfFile", + "type": "Document", + "baseClasses": ["Document"], + "category": "Document Loaders", + "description": "Load data from PDF files", + "inputParams": [ + { "label": "Pdf File", "name": "pdfFile", "type": "file", "fileType": ".pdf", "id": "pdfFile_0-input-pdfFile-file" }, + { + "label": "Usage", + "name": "usage", + "type": "options", + "options": [ + { "label": "One document per page", "name": "perPage" }, + { "label": "One document per file", "name": "perFile" } + ], + "default": "perPage", + "id": "pdfFile_0-input-usage-options" + }, + { + "label": "Use Legacy Build", + "name": "legacyBuild", + "type": "boolean", + "optional": true, + "additionalParams": true, + "id": "pdfFile_0-input-legacyBuild-boolean" + }, + { + "label": "Metadata", + "name": "metadata", + "type": "json", + "optional": true, + "additionalParams": true, + "id": "pdfFile_0-input-metadata-json" + } + ], + "inputAnchors": [ + { + "label": "Text Splitter", + "name": "textSplitter", + "type": "TextSplitter", + "optional": true, + "id": "pdfFile_0-input-textSplitter-TextSplitter" + } + ], + "inputs": { "textSplitter": "", "usage": "perPage", "legacyBuild": "", "metadata": "" }, + "outputAnchors": [ + { "id": "pdfFile_0-output-pdfFile-Document", "name": "pdfFile", "label": "Document", "type": "Document" } + ], + "outputs": {}, + "selected": false + }, + "selected": false, + "positionAbsolute": { "x": -210.44158723479913, "y": 236.6627524951051 }, + "dragging": false + }, + { + "width": 300, + "height": 408, + "id": "vectaraUpsert_0", + "position": { "x": 172.06946164914868, "y": 373.11406233089934 }, + "type": "customNode", + "data": { + "id": "vectaraUpsert_0", + "label": "Vectara Upsert Document", + "version": 1, + "name": "vectaraUpsert", + "type": "Vectara", + "baseClasses": ["Vectara", "VectorStoreRetriever", "BaseRetriever"], + "category": "Vector Stores", + "description": "Upsert documents to Vectara", + "inputParams": [ + { + "label": "Connect Credential", + "name": "credential", + "type": "credential", + "credentialNames": ["vectaraApi"], + "id": "vectaraUpsert_0-input-credential-credential" + }, + { + "label": "Vectara Metadata Filter", + "name": "filter", + "description": "Filter to apply to Vectara metadata. Refer to the documentation on how to use Vectara filters with Flowise.", + "type": "string", + "additionalParams": true, + "optional": true, + "id": "vectaraUpsert_0-input-filter-string" + }, + { + "label": "Sentences Before", + "name": "sentencesBefore", + "description": "Number of sentences to fetch before the matched sentence. Defaults to 2.", + "type": "number", + "additionalParams": true, + "optional": true, + "id": "vectaraUpsert_0-input-sentencesBefore-number" + }, + { + "label": "Sentences After", + "name": "sentencesAfter", + "description": "Number of sentences to fetch after the matched sentence. Defaults to 2.", + "type": "number", + "additionalParams": true, + "optional": true, + "id": "vectaraUpsert_0-input-sentencesAfter-number" + }, + { + "label": "Lambda", + "name": "lambda", + "description": "Improves retrieval accuracy by adjusting the balance (from 0 to 1) between neural search and keyword-based search factors.", + "type": "number", + "additionalParams": true, + "optional": true, + "id": "vectaraUpsert_0-input-lambda-number" + }, + { + "label": "Top K", + "name": "topK", + "description": "Number of top results to fetch. Defaults to 4", + "placeholder": "4", + "type": "number", + "additionalParams": true, + "optional": true, + "id": "vectaraUpsert_0-input-topK-number" + } + ], + "inputAnchors": [ + { + "label": "Document", + "name": "document", + "type": "Document", + "list": true, + "id": "vectaraUpsert_0-input-document-Document" + } + ], + "inputs": { + "document": ["{{pdfFile_0.data.instance}}"], + "filter": "", + "sentencesBefore": "", + "sentencesAfter": "", + "lambda": "", + "topK": "" + }, + "outputAnchors": [ + { + "name": "output", + "label": "Output", + "type": "options", + "options": [ + { + "id": "vectaraUpsert_0-output-retriever-Vectara|VectorStoreRetriever|BaseRetriever", + "name": "retriever", + "label": "Vectara Retriever", + "type": "Vectara | VectorStoreRetriever | BaseRetriever" + }, + { + "id": "vectaraUpsert_0-output-vectorStore-Vectara|VectorStore", + "name": "vectorStore", + "label": "Vectara Vector Store", + "type": "Vectara | VectorStore" + } + ], + "default": "retriever" + } + ], + "outputs": { "output": "retriever" }, + "selected": false + }, + "positionAbsolute": { "x": 172.06946164914868, "y": 373.11406233089934 }, + "selected": false } ], "edges": [ + { + "source": "chatOpenAI_0", + "sourceHandle": "chatOpenAI_0-output-chatOpenAI-ChatOpenAI|BaseChatModel|BaseLanguageModel", + "target": "conversationalRetrievalQAChain_0", + "targetHandle": "conversationalRetrievalQAChain_0-input-model-BaseLanguageModel", + "type": "buttonedge", + "id": "chatOpenAI_0-chatOpenAI_0-output-chatOpenAI-ChatOpenAI|BaseChatModel|BaseLanguageModel-conversationalRetrievalQAChain_0-conversationalRetrievalQAChain_0-input-model-BaseLanguageModel", + "data": { "label": "" } + }, { "source": "pdfFile_0", "sourceHandle": "pdfFile_0-output-pdfFile-Document", @@ -432,15 +443,6 @@ "type": "buttonedge", "id": "vectaraUpsert_0-vectaraUpsert_0-output-retriever-Vectara|VectorStoreRetriever|BaseRetriever-conversationalRetrievalQAChain_0-conversationalRetrievalQAChain_0-input-vectorStoreRetriever-BaseRetriever", "data": { "label": "" } - }, - { - "source": "chatOpenAI_0", - "sourceHandle": "chatOpenAI_0-output-chatOpenAI-ChatOpenAI|BaseChatModel|BaseLanguageModel", - "target": "conversationalRetrievalQAChain_0", - "targetHandle": "conversationalRetrievalQAChain_0-input-model-BaseLanguageModel", - "type": "buttonedge", - "id": "chatOpenAI_0-chatOpenAI_0-output-chatOpenAI-ChatOpenAI|BaseChatModel|BaseLanguageModel-conversationalRetrievalQAChain_0-conversationalRetrievalQAChain_0-input-model-BaseLanguageModel", - "data": { "label": "" } } ] } diff --git a/packages/server/marketplaces/chatflows/WebPage QnA.json b/packages/server/marketplaces/chatflows/WebPage QnA.json index 8197c20a..812f0bd5 100644 --- a/packages/server/marketplaces/chatflows/WebPage QnA.json +++ b/packages/server/marketplaces/chatflows/WebPage QnA.json @@ -466,7 +466,7 @@ "optional": true, "additionalParams": true, "description": "Only used when \"Get Relative Links Method\" is selected. Set 0 to retrieve all relative links, default limit is 10.", - "warning": "Retreiving all links might take long time, and all links will be upserted again if the flow's state changed (eg: different URL, chunk size, etc)", + "warning": "Retrieving all links might take long time, and all links will be upserted again if the flow's state changed (eg: different URL, chunk size, etc)", "id": "cheerioWebScraper_0-input-limit-number" }, { diff --git a/packages/server/src/index.ts b/packages/server/src/index.ts index b7289149..23b5bdf4 100644 --- a/packages/server/src/index.ts +++ b/packages/server/src/index.ts @@ -36,7 +36,6 @@ import { isSameOverrideConfig, replaceAllAPIKeys, isFlowValidForStream, - isVectorStoreFaiss, databaseEntities, getApiKey, transformToCredentialEntity, @@ -811,11 +810,17 @@ export class App { } } + /*** Get chatflows and prepare data ***/ + const flowData = chatflow.flowData + const parsedFlowData: IReactFlowObject = JSON.parse(flowData) + const nodes = parsedFlowData.nodes + const edges = parsedFlowData.edges + /* Reuse the flow without having to rebuild (to avoid duplicated upsert, recomputation) when all these conditions met: * - Node Data already exists in pool * - Still in sync (i.e the flow has not been modified since) * - Existing overrideConfig and new overrideConfig are the same - * - Flow doesn't start with nodes that depend on incomingInput.question + * - Flow doesn't start with/contain nodes that depend on incomingInput.question ***/ const isFlowReusable = () => { return ( @@ -826,16 +831,10 @@ export class App { this.chatflowPool.activeChatflows[chatflowid].overrideConfig, incomingInput.overrideConfig ) && - !isStartNodeDependOnInput(this.chatflowPool.activeChatflows[chatflowid].startingNodes) + !isStartNodeDependOnInput(this.chatflowPool.activeChatflows[chatflowid].startingNodes, nodes) ) } - /*** Get chatflows and prepare data ***/ - const flowData = chatflow.flowData - const parsedFlowData: IReactFlowObject = JSON.parse(flowData) - const nodes = parsedFlowData.nodes - const edges = parsedFlowData.edges - if (isFlowReusable()) { nodeToExecuteData = this.chatflowPool.activeChatflows[chatflowid].endingNodeData isStreamValid = isFlowValidForStream(nodes, nodeToExecuteData) @@ -884,6 +883,7 @@ export class App { depthQueue, this.nodesPool.componentNodes, incomingInput.question, + incomingInput.history, chatId, this.AppDataSource, incomingInput?.overrideConfig @@ -894,7 +894,12 @@ export class App { if (incomingInput.overrideConfig) nodeToExecute.data = replaceInputsWithConfig(nodeToExecute.data, incomingInput.overrideConfig) - const reactFlowNodeData: INodeData = resolveVariables(nodeToExecute.data, reactFlowNodes, incomingInput.question) + const reactFlowNodeData: INodeData = resolveVariables( + nodeToExecute.data, + reactFlowNodes, + incomingInput.question, + incomingInput.history + ) nodeToExecuteData = reactFlowNodeData const startingNodes = nodes.filter((nd) => startingNodeIds.includes(nd.id)) @@ -905,7 +910,6 @@ export class App { const nodeModule = await import(nodeInstanceFilePath) const nodeInstance = new nodeModule.nodeClass() - isStreamValid = isStreamValid && !isVectorStoreFaiss(nodeToExecuteData) logger.debug(`[server]: Running ${nodeToExecuteData.label} (${nodeToExecuteData.id})`) if (nodeToExecuteData.instance) checkMemorySessionId(nodeToExecuteData.instance, chatId) diff --git a/packages/server/src/utils/index.ts b/packages/server/src/utils/index.ts index cf4a9c3f..9be26987 100644 --- a/packages/server/src/utils/index.ts +++ b/packages/server/src/utils/index.ts @@ -18,8 +18,15 @@ import { IComponentCredentials, ICredentialReqBody } from '../Interface' -import { cloneDeep, get, omit, merge, isEqual } from 'lodash' -import { ICommonObject, getInputVariables, IDatabaseEntity, handleEscapeCharacters } from 'flowise-components' +import { cloneDeep, get, isEqual } from 'lodash' +import { + ICommonObject, + getInputVariables, + IDatabaseEntity, + handleEscapeCharacters, + IMessage, + convertChatHistoryToText +} from 'flowise-components' import { scryptSync, randomBytes, timingSafeEqual } from 'crypto' import { lib, PBKDF2, AES, enc } from 'crypto-js' @@ -30,6 +37,7 @@ import { Tool } from '../entity/Tool' import { DataSource } from 'typeorm' const QUESTION_VAR_PREFIX = 'question' +const CHAT_HISTORY_VAR_PREFIX = 'chat_history' const REDACTED_CREDENTIAL_VALUE = '_FLOWISE_BLANK_07167752-1a71-43b1-bf8f-4f32252165db' export const databaseEntities: IDatabaseEntity = { ChatFlow: ChatFlow, ChatMessage: ChatMessage, Tool: Tool, Credential: Credential } @@ -199,6 +207,7 @@ export const buildLangchain = async ( depthQueue: IDepthQueue, componentNodes: IComponentNodes, question: string, + chatHistory: IMessage[], chatId: string, appDataSource: DataSource, overrideConfig?: ICommonObject @@ -231,7 +240,7 @@ export const buildLangchain = async ( let flowNodeData = cloneDeep(reactFlowNode.data) if (overrideConfig) flowNodeData = replaceInputsWithConfig(flowNodeData, overrideConfig) - const reactFlowNodeData: INodeData = resolveVariables(flowNodeData, flowNodes, question) + const reactFlowNodeData: INodeData = resolveVariables(flowNodeData, flowNodes, question, chatHistory) logger.debug(`[server]: Initializing ${reactFlowNode.data.label} (${reactFlowNode.data.id})`) flowNodes[nodeIndex].data.instance = await newNodeInstance.init(reactFlowNodeData, question, { @@ -315,7 +324,13 @@ export const clearSessionMemory = async ( * @param {boolean} isAcceptVariable * @returns {string} */ -export const getVariableValue = (paramValue: string, reactFlowNodes: IReactFlowNode[], question: string, isAcceptVariable = false) => { +export const getVariableValue = ( + paramValue: string, + reactFlowNodes: IReactFlowNode[], + question: string, + chatHistory: IMessage[], + isAcceptVariable = false +) => { let returnVal = paramValue const variableStack = [] const variableDict = {} as IVariableDict @@ -345,6 +360,10 @@ export const getVariableValue = (paramValue: string, reactFlowNodes: IReactFlowN variableDict[`{{${variableFullPath}}}`] = handleEscapeCharacters(question, false) } + if (isAcceptVariable && variableFullPath === CHAT_HISTORY_VAR_PREFIX) { + variableDict[`{{${variableFullPath}}}`] = handleEscapeCharacters(convertChatHistoryToText(chatHistory), false) + } + // Split by first occurrence of '.' to get just nodeId const [variableNodeId, _] = variableFullPath.split('.') const executedNode = reactFlowNodes.find((nd) => nd.id === variableNodeId) @@ -374,25 +393,6 @@ export const getVariableValue = (paramValue: string, reactFlowNodes: IReactFlowN return returnVal } -/** - * Temporarily disable streaming if vectorStore is Faiss - * @param {INodeData} flowNodeData - * @returns {boolean} - */ -export const isVectorStoreFaiss = (flowNodeData: INodeData) => { - if (flowNodeData.inputs && flowNodeData.inputs.vectorStoreRetriever) { - const vectorStoreRetriever = flowNodeData.inputs.vectorStoreRetriever - if (typeof vectorStoreRetriever === 'string' && vectorStoreRetriever.includes('faiss')) return true - if ( - typeof vectorStoreRetriever === 'object' && - vectorStoreRetriever.vectorStore && - vectorStoreRetriever.vectorStore.constructor.name === 'FaissStore' - ) - return true - } - return false -} - /** * Loop through each inputs and resolve variable if neccessary * @param {INodeData} reactFlowNodeData @@ -400,13 +400,13 @@ export const isVectorStoreFaiss = (flowNodeData: INodeData) => { * @param {string} question * @returns {INodeData} */ -export const resolveVariables = (reactFlowNodeData: INodeData, reactFlowNodes: IReactFlowNode[], question: string): INodeData => { +export const resolveVariables = ( + reactFlowNodeData: INodeData, + reactFlowNodes: IReactFlowNode[], + question: string, + chatHistory: IMessage[] +): INodeData => { let flowNodeData = cloneDeep(reactFlowNodeData) - if (reactFlowNodeData.instance && isVectorStoreFaiss(reactFlowNodeData)) { - // omit and merge because cloneDeep of instance gives "Illegal invocation" Exception - const flowNodeDataWithoutInstance = cloneDeep(omit(reactFlowNodeData, ['instance'])) - flowNodeData = merge(flowNodeDataWithoutInstance, { instance: reactFlowNodeData.instance }) - } const types = 'inputs' const getParamValues = (paramsObj: ICommonObject) => { @@ -415,13 +415,13 @@ export const resolveVariables = (reactFlowNodeData: INodeData, reactFlowNodes: I if (Array.isArray(paramValue)) { const resolvedInstances = [] for (const param of paramValue) { - const resolvedInstance = getVariableValue(param, reactFlowNodes, question) + const resolvedInstance = getVariableValue(param, reactFlowNodes, question, chatHistory) resolvedInstances.push(resolvedInstance) } paramsObj[key] = resolvedInstances } else { const isAcceptVariable = reactFlowNodeData.inputParams.find((param) => param.name === key)?.acceptVariable ?? false - const resolvedInstance = getVariableValue(paramValue, reactFlowNodes, question, isAcceptVariable) + const resolvedInstance = getVariableValue(paramValue, reactFlowNodes, question, chatHistory, isAcceptVariable) paramsObj[key] = resolvedInstance } } @@ -474,13 +474,17 @@ export const replaceInputsWithConfig = (flowNodeData: INodeData, overrideConfig: * @param {IReactFlowNode[]} startingNodes * @returns {boolean} */ -export const isStartNodeDependOnInput = (startingNodes: IReactFlowNode[]): boolean => { +export const isStartNodeDependOnInput = (startingNodes: IReactFlowNode[], nodes: IReactFlowNode[]): boolean => { for (const node of startingNodes) { for (const inputName in node.data.inputs) { const inputVariables = getInputVariables(node.data.inputs[inputName]) if (inputVariables.length > 0) return true } } + const whitelistNodeNames = ['vectorStoreToDocument'] + for (const node of nodes) { + if (whitelistNodeNames.includes(node.data.name)) return true + } return false } @@ -791,7 +795,7 @@ export const isFlowValidForStream = (reactFlowNodes: IReactFlowNode[], endingNod isValidChainOrAgent = whitelistAgents.includes(endingNodeData.name) } - return isChatOrLLMsExist && isValidChainOrAgent && !isVectorStoreFaiss(endingNodeData) + return isChatOrLLMsExist && isValidChainOrAgent } /** diff --git a/packages/ui/README-ZH.md b/packages/ui/README-ZH.md index 5d33c07f..c6307935 100644 --- a/packages/ui/README-ZH.md +++ b/packages/ui/README-ZH.md @@ -16,4 +16,4 @@ npm i flowise-ui ## 许可证 -本仓库中的源代码在[MIT许可证](https://github.com/FlowiseAI/Flowise/blob/master/LICENSE.md)下提供。 \ No newline at end of file +本仓库中的源代码在[MIT许可证](https://github.com/FlowiseAI/Flowise/blob/master/LICENSE.md)下提供。 diff --git a/packages/ui/src/assets/images/chathistory.png b/packages/ui/src/assets/images/chathistory.png new file mode 100644 index 00000000..52f496a8 Binary files /dev/null and b/packages/ui/src/assets/images/chathistory.png differ diff --git a/packages/ui/src/ui-component/dialog/NodeInfoDialog.js b/packages/ui/src/ui-component/dialog/NodeInfoDialog.js index 054353fc..74c45a1a 100644 --- a/packages/ui/src/ui-component/dialog/NodeInfoDialog.js +++ b/packages/ui/src/ui-component/dialog/NodeInfoDialog.js @@ -123,7 +123,7 @@ const NodeInfoDialog = ({ show, dialogProps, onCancel }) => { )} {getNodeConfigApi.data && getNodeConfigApi.data.length > 0 && ( - + )} diff --git a/packages/ui/src/ui-component/json/SelectVariable.js b/packages/ui/src/ui-component/json/SelectVariable.js index 1b891ed1..7a482bae 100644 --- a/packages/ui/src/ui-component/json/SelectVariable.js +++ b/packages/ui/src/ui-component/json/SelectVariable.js @@ -2,14 +2,15 @@ import { useSelector } from 'react-redux' import PropTypes from 'prop-types' import { Box, List, ListItemButton, ListItem, ListItemAvatar, ListItemText, Typography, Stack } from '@mui/material' import PerfectScrollbar from 'react-perfect-scrollbar' - +import robotPNG from 'assets/images/robot.png' +import chatPNG from 'assets/images/chathistory.png' import { baseURL } from 'store/constant' const SelectVariable = ({ availableNodesForVariable, disabled = false, onSelectAndReturnVal }) => { const customization = useSelector((state) => state.customization) - const onSelectOutputResponseClick = (node, isUserQuestion = false) => { - let variablePath = isUserQuestion ? `question` : `${node.id}.data.instance` + const onSelectOutputResponseClick = (node, prefix) => { + let variablePath = node ? `${node.id}.data.instance` : prefix const newInput = `{{${variablePath}}}` onSelectAndReturnVal(newInput) } @@ -32,7 +33,7 @@ const SelectVariable = ({ availableNodesForVariable, disabled = false, onSelectA mb: 1 }} disabled={disabled} - onClick={() => onSelectOutputResponseClick(null, true)} + onClick={() => onSelectOutputResponseClick(null, 'question')} > @@ -52,13 +53,52 @@ const SelectVariable = ({ availableNodesForVariable, disabled = false, onSelectA objectFit: 'contain' }} alt='AI' - src='https://raw.githubusercontent.com/zahidkhawaja/langchain-chat-nextjs/main/public/parroticon.png' + src={robotPNG} /> + onSelectOutputResponseClick(null, 'chat_history')} + > + + +
+ chatHistory +
+ + + + {availableNodesForVariable && availableNodesForVariable.length > 0 && availableNodesForVariable.map((node, index) => { diff --git a/packages/ui/src/views/chatmessage/ChatMessage.js b/packages/ui/src/views/chatmessage/ChatMessage.js index 98eef72a..506f02da 100644 --- a/packages/ui/src/views/chatmessage/ChatMessage.js +++ b/packages/ui/src/views/chatmessage/ChatMessage.js @@ -346,7 +346,9 @@ export const ChatMessage = ({ open, chatflowid, isDialog }) => { key={index} label={ URL - ? `${URL.pathname.substring(0, 15)}...` + ? URL.pathname.substring(0, 15) === '/' + ? URL.host + : `${URL.pathname.substring(0, 15)}...` : `${source.pageContent.substring(0, 15)}...` } component='a'