anything-llm/server/utils/vectorDbProviders/pinecone/index.js

const { PineconeClient } = require("@pinecone-database/pinecone");
const { PineconeStore } = require("langchain/vectorstores/pinecone");
const { OpenAI } = require("langchain/llms/openai");
const { VectorDBQAChain, LLMChain } = require("langchain/chains");
const { OpenAIEmbeddings } = require("langchain/embeddings/openai");
const { VectorStoreRetrieverMemory } = require("langchain/memory");
const { PromptTemplate } = require("langchain/prompts");
const { RecursiveCharacterTextSplitter } = require("langchain/text_splitter");
const { storeVectorResult, cachedVectorInformation } = require("../../files");
const { Configuration, OpenAIApi } = require("openai");
const { v4: uuidv4 } = require("uuid");
const { toChunks, curateSources } = require("../../helpers");

const Pinecone = {
  name: "Pinecone",
  connect: async function () {
    if (process.env.VECTOR_DB !== "pinecone")
      throw new Error("Pinecone::Invalid ENV settings");

    const client = new PineconeClient();
    await client.init({
      apiKey: process.env.PINECONE_API_KEY,
      environment: process.env.PINECONE_ENVIRONMENT,
    });
    const pineconeIndex = client.Index(process.env.PINECONE_INDEX);
    const { status } = await client.describeIndex({
      indexName: process.env.PINECONE_INDEX,
    });

    if (!status.ready) throw new Error("Pinecode::Index not ready.");
    return { client, pineconeIndex, indexName: process.env.PINECONE_INDEX };
  },
  embedder: function () {
    return new OpenAIEmbeddings({ openAIApiKey: process.env.OPEN_AI_KEY });
  },
  openai: function () {
    const config = new Configuration({ apiKey: process.env.OPEN_AI_KEY });
    const openai = new OpenAIApi(config);
    return openai;
  },
  getChatCompletion: async function (
    openai,
    messages = [],
    { temperature = 0.7 }
  ) {
    const model = process.env.OPEN_MODEL_PREF || "gpt-3.5-turbo";
    const { data } = await openai.createChatCompletion({
      model,
      messages,
      temperature,
    });

    if (!data.hasOwnProperty("choices")) return null;
    return data.choices[0].message.content;
  },
  embedChunk: async function (openai, textChunk) {
    const {
      data: { data },
    } = await openai.createEmbedding({
      model: "text-embedding-ada-002",
      input: textChunk,
    });
    return data.length > 0 && data[0].hasOwnProperty("embedding")
      ? data[0].embedding
      : null;
  },
  llm: function ({ temperature = 0.7 }) {
    const model = process.env.OPEN_MODEL_PREF || "gpt-3.5-turbo";
    return new OpenAI({
      openAIApiKey: process.env.OPEN_AI_KEY,
      modelName: model,
      temperature,
    });
  },
  totalIndicies: async function () {
    const { pineconeIndex } = await this.connect();
    const { namespaces } = await pineconeIndex.describeIndexStats1();
    return Object.values(namespaces).reduce(
      (a, b) => a + (b?.vectorCount || 0),
      0
    );
  },
  similarityResponse: async function (index, namespace, queryVector) {
    const result = {
      contextTexts: [],
      sourceDocuments: [],
    };
    const response = await index.query({
      queryRequest: {
        namespace,
        vector: queryVector,
        topK: 4,
        includeMetadata: true,
      },
    });

    response.matches.forEach((match) => {
      result.contextTexts.push(match.metadata.text);
      result.sourceDocuments.push(match);
    });

    return result;
  },
  namespace: async function (index, namespace = null) {
    if (!namespace) throw new Error("No namespace value provided.");
    const { namespaces } = await index.describeIndexStats1();
    return namespaces.hasOwnProperty(namespace) ? namespaces[namespace] : null;
  },
  hasNamespace: async function (namespace = null) {
    if (!namespace) return false;
    const { pineconeIndex } = await this.connect();
    return await this.namespaceExists(pineconeIndex, namespace);
  },
  namespaceExists: async function (index, namespace = null) {
    if (!namespace) throw new Error("No namespace value provided.");
    const { namespaces } = await index.describeIndexStats1();
    return namespaces.hasOwnProperty(namespace);
  },
  deleteVectorsInNamespace: async function (index, namespace = null) {
    await index.delete1({ namespace, deleteAll: true });
    return true;
  },
  addDocumentToNamespace: async function (
    namespace,
    documentData = {},
    fullFilePath = null
  ) {
    const { DocumentVectors } = require("../../../models/vectors");
    try {
      const { pageContent, docId, ...metadata } = documentData;
      if (!pageContent || pageContent.length == 0) return false;

      console.log("Adding new vectorized document into namespace", namespace);
      const cacheResult = await cachedVectorInformation(fullFilePath);
      if (cacheResult.exists) {
        const { pineconeIndex } = await this.connect();
        const { chunks } = cacheResult;
        const documentVectors = [];

        for (const chunk of chunks) {
          // Before sending to Pinecone and saving the records to our db
          // we need to assign the id of each chunk that is stored in the cached file.
          const newChunks = chunk.map((chunk) => {
            const id = uuidv4();
            documentVectors.push({ docId, vectorId: id });
            return { ...chunk, id };
          });

          // Push chunks with new ids to pinecone.
          await pineconeIndex.upsert({
            upsertRequest: {
              vectors: [...newChunks],
              namespace,
            },
          });
        }

        await DocumentVectors.bulkInsert(documentVectors);
        return true;
      }

      // If we are here then we are going to embed and store a novel document.
      // We have to do this manually as opposed to using LangChains `PineconeStore.fromDocuments`
      // because we then cannot atomically control our namespace to granularly find/remove documents
      // from vectordb.
      // https://github.com/hwchase17/langchainjs/blob/2def486af734c0ca87285a48f1a04c057ab74bdf/langchain/src/vectorstores/pinecone.ts#L167
      const textSplitter = new RecursiveCharacterTextSplitter({
        chunkSize: 1000,
        chunkOverlap: 20,
      });
      const textChunks = await textSplitter.splitText(pageContent);

      console.log("Chunks created from document:", textChunks.length);
      const documentVectors = [];
      const vectors = [];
      const openai = this.openai();
      for (const textChunk of textChunks) {
        const vectorValues = await this.embedChunk(openai, textChunk);

        if (!!vectorValues) {
          const vectorRecord = {
            id: uuidv4(),
            values: vectorValues,
            // [DO NOT REMOVE]
            // LangChain will be unable to find your text if you embed manually and dont include the `text` key.
            // https://github.com/hwchase17/langchainjs/blob/2def486af734c0ca87285a48f1a04c057ab74bdf/langchain/src/vectorstores/pinecone.ts#L64
            metadata: { ...metadata, text: textChunk },
          };
          vectors.push(vectorRecord);
          documentVectors.push({ docId, vectorId: vectorRecord.id });
        } else {
          console.error(
            "Could not use OpenAI to embed document chunk! This document will not be recorded."
          );
        }
      }

      if (vectors.length > 0) {
        const chunks = [];
        const { pineconeIndex } = await this.connect();
        console.log("Inserting vectorized chunks into Pinecone.");
        for (const chunk of toChunks(vectors, 100)) {
          chunks.push(chunk);
          await pineconeIndex.upsert({
            upsertRequest: {
              vectors: [...chunk],
              namespace,
            },
          });
        }
        await storeVectorResult(chunks, fullFilePath);
      }

      await DocumentVectors.bulkInsert(documentVectors);
      return true;
    } catch (e) {
      console.error("addDocumentToNamespace", e.message);
      return false;
    }
  },
  deleteDocumentFromNamespace: async function (namespace, docId) {
    const { DocumentVectors } = require("../../../models/vectors");
    const { pineconeIndex } = await this.connect();
    if (!(await this.namespaceExists(pineconeIndex, namespace))) return;

    const knownDocuments = await DocumentVectors.where(`docId = '${docId}'`);
    if (knownDocuments.length === 0) return;

    const vectorIds = knownDocuments.map((doc) => doc.vectorId);
    await pineconeIndex.delete1({
      ids: vectorIds,
      namespace,
    });

    const indexes = knownDocuments.map((doc) => doc.id);
    await DocumentVectors.deleteIds(indexes);
    return true;
  },
  "namespace-stats": async function (reqBody = {}) {
    const { namespace = null } = reqBody;
    if (!namespace) throw new Error("namespace required");
    const { pineconeIndex } = await this.connect();
    if (!(await this.namespaceExists(pineconeIndex, namespace)))
      throw new Error("Namespace by that name does not exist.");
    const stats = await this.namespace(pineconeIndex, namespace);
    return stats
      ? stats
      : { message: "No stats were able to be fetched from DB" };
  },
  "delete-namespace": async function (reqBody = {}) {
    const { namespace = null } = reqBody;
    const { pineconeIndex } = await this.connect();
    if (!(await this.namespaceExists(pineconeIndex, namespace)))
      throw new Error("Namespace by that name does not exist.");

    const details = await this.namespace(pineconeIndex, namespace);
    await this.deleteVectorsInNamespace(pineconeIndex, namespace);
    return {
      message: `Namespace ${namespace} was deleted along with ${details.vectorCount} vectors.`,
    };
  },
  query: async function (reqBody = {}) {
    const { namespace = null, input, workspace = {} } = reqBody;
    if (!namespace || !input) throw new Error("Invalid request body");

    const { pineconeIndex } = await this.connect();
    if (!(await this.namespaceExists(pineconeIndex, namespace))) {
      return {
        response: null,
        sources: [],
        message: "Invalid query - no documents found for workspace!",
      };
    }

    const vectorStore = await PineconeStore.fromExistingIndex(this.embedder(), {
      pineconeIndex,
      namespace,
    });

    const model = this.llm({
      temperature: workspace?.openAiTemp,
    });
    const chain = VectorDBQAChain.fromLLM(model, vectorStore, {
      k: 5,
      returnSourceDocuments: true,
    });
    const response = await chain.call({ query: input });
    return {
      response: response.text,
      sources: curateSources(response.sourceDocuments),
      message: false,
    };
  },
  // This implementation of chat uses the chat history and modifies the system prompt at execution
  // this is improved over the regular langchain implementation so that chats do not directly modify embeddings
  // because then multi-user support will have all conversations mutating the base vector collection to which then
  // the only solution is replicating entire vector databases per user - which will very quickly consume space on VectorDbs
  chat: async function (reqBody = {}) {
    const {
      namespace = null,
      input,
      workspace = {},
      chatHistory = [],
    } = reqBody;
    if (!namespace || !input) throw new Error("Invalid request body");

    const { pineconeIndex } = await this.connect();
    if (!(await this.namespaceExists(pineconeIndex, namespace)))
      throw new Error(
        "Invalid namespace - has it been collected and seeded yet?"
      );

    const queryVector = await this.embedChunk(this.openai(), input);
    const { contextTexts, sourceDocuments } = await this.similarityResponse(
      pineconeIndex,
      namespace,
      queryVector
    );
    const prompt = {
      role: "system",
      content: `Given the following conversation, relevant context, and a follow up question, reply with an answer to the current question the user is asking. Return only your response to the question given the above information following the users instructions as needed.
Context:
${contextTexts
  .map((text, i) => {
    return `[CONTEXT ${i}]:\n${text}\n[END CONTEXT ${i}]\n\n`;
  })
  .join("")}`,
    };
    const memory = [prompt, ...chatHistory, { role: "user", content: input }];

    const responseText = await this.getChatCompletion(this.openai(), memory, {
      temperature: workspace?.openAiTemp ?? 0.7,
    });

    return {
      response: responseText,
      sources: curateSources(sourceDocuments),
      message: false,
    };
  },
};

module.exports.Pinecone = Pinecone;