const express = require('express');
const { MongoClient } = require("mongodb");
const axios = require('axios');
const app = express();

/** In case you require CORS for Browser */
//const cors = require('cors');
//app.use(cors());

/** OpenAI Embedding Function */
async function openaiEmbedding(query) {

  // OpenAI Embeddings
  const url = 'https://api.openai.com/v1/embeddings';
  const openai_key = "YOUR-API-TOKEN"; // Replace with your OpenAI key.
  
  // OpenAI embeddings APIs
  let response = await axios.post(url, {
      input: query,
      model: "text-embedding-ada-002"
  }, {
      headers: {
          'Authorization': `Bearer ${openai_key}`,
          'Content-Type': 'application/json'
      }
  });
  
  if(response.status === 200) {
      console.log(response.data.data[0].embedding)
      return response.data.data[0].embedding;
  } else {
      throw new Error(`Failed to get embedding with code: ${response.status}`);
  }
}

app.get("/vectorSearch/:query", async (req,res)=>{  

  try {

    // Transform query into embedding
    const embedding = await openaiEmbedding(req.params.query);

    // Change these constants:
    const URI = "mongodb+srv:https://username:[email protected]";
    const databaseName = "databaseDemo";
    const collectionName = "collectionDemo"


    const client = new MongoClient(URI);
    await client.connect();
    
    const db = client.db(databaseName); 
    const collection = db.collection(collectionName); 
    
    // Query for similar documents.
    const documents = await collection.aggregate([
        {
          "$search": {
            "index": "vectorIndex", // Name of Vector Search Index
            "knnBeta": {
            "vector": embedding,
            "path": "embedding", // Name of the 'embedding' field
            "k": 5
            }
          }
        }
        ]).toArray();      
      
    res.send(documents);
    
  } catch(err) {
      console.error(err);
  }  
});

/** PORT */ 
const port = process.env.PORT || 8000;

/** PORT LISTENER **/
app.listen(port, () => {
  console.log(`Listening to port ${port}`);
}); 

exports = async function(changeEvent) {

    // Gets the full document that was changed
    const changedDocument = changeEvent.fullDocument;
    const url = 'https://api.openai.com/v1/embeddings';
    
    // OpenAI API to change
    const openai_key = "<YOUR-OPENAI-KEY>";

    try {
        // HTTP call to OpenAI API
        let response = await context.http.post({
            url: url,
             headers: {
                'Authorization': [`Bearer ${openai_key}`],
                'Content-Type': ['application/json']
            },
            body: JSON.stringify({
                input: changedDocument.description, //You can change the 'description' field to another one that you wish to convert into vector embedding
                model: "text-embedding-ada-002"
            })
        });

        // Parse the JSON response
        let responseData = EJSON.parse(response.body.text());

        if(response.statusCode === 200) {
            console.log("Successfully received embedding.");

            const responseEmbedding = responseData.data[0].embedding;

            // MongoDB Atlas Cluster / Database / Collection
            const collection = context.services.get("<CLUSTER_NAME>").db("databaseDemo").collection("collectionDemo");

            // Update the document in MongoDB.
            const result = await collection.updateOne(
                { _id: changedDocument._id },
                // Adds the embedding field
                { $set: { embedding: responseEmbedding }}
            );

            if(result.modifiedCount === 1) {
                console.log("Document successfully Updated.");
            } else {
                console.log("Failed to modify document.");
            }
        } else {
            console.log(`Failed embedding with code: ${response.statusCode}`);
        }

    } catch(err) {
        console.error(err);
    }
};