Vectara GSheet Integration

gsheet-vectara.js

/**
 * Calculates the sale price of a value at a given discount.
 * The sale price is formatted as US dollars.
 *
 * @param {string} question The RFI question to answer.
 * @param {number} discount The discount to apply, such as .5 or 50%.
 * @return The answer according to Vectara.
 * @customfunction
 */
function VectaraRFIAnswer(question) {
  var vectaraCorpusID = 1; // your corpus ID
  var vectaraCustomerID = "YOUR_CUSTOMER_ID";
  var vectaraApiKey = 'YOUR_PERSONAL_API_KEY';
  var vectaraPrompt = `[
    {"role": "system", "content": "You are an RFI answering assistant acting on behalf of the company Vectara.  You are provided with search results from previously responded-to RFIs that may help answer the given question. You must summarize these results as a coherent answer. Give slight preference to search results that appear earlier in the chat. Only use information provided in this chat."},
    #foreach ($qResult in $vectaraQueryResults)
      #if ($foreach.first)
        {"role": "user", "content": "Search for \\"$esc.java(\\\${vectaraQuery})\\", and give me the first search result."},
        {"role": "assistant", "content": "$esc.java(\\\${qResult.getText()})" },
      #else
        {"role": "user", "content": "Give me the $vectaraIdxWord[$foreach.index] search result."},
        {"role": "assistant", "content": "$esc.java(\\\${qResult.getText()})" },
      #end
    #end
    {"role": "user", "content": "Generate a comprehensive and informative answer (but no more than $vectaraOutChars characters) for the question \\"$esc.java(\\\${vectaraQuery})\\" solely based on the search results in this chat. You must only use information from the provided results. Combine search results together into a coherent answer. Do not repeat text. Only use the most relevant results that answer the question accurately. If a result does not answer the question, do not use it. If the search results are not valid, respond with \\"The returned results did not contain sufficient information to the question.\\"."}
]`;

  var headers = {
     'x-api-key': vectaraApiKey,
     'Content-Type': 'application/json',
     "customer-id": vectaraCustomerID
   };

   var url = 'https://api.vectara.io/v1/query';
   
   var corpus_key = {
     'customer_id': vectaraCustomerID,
     'corpus_id': vectaraCorpusID,
     'lexicalInterpolationConfig': { 'lambda': 0.025 }
   };

   var query = {
    'query': question,
    'num_results': 10,
    'corpus_key': [ corpus_key ],
    'summary': [{
      'promptText': vectaraPrompt,
      'responseLang': 'en',
      'maxSummarizedResults': 5
    }]
   };
   
   var payload = {
    "query": [ query ]
   };

   Logger.log(vectaraPrompt);

  var options = {
    'method' : 'post',
    'contentType': 'application/json',
    'headers': headers,
    'payload': JSON.stringify(payload)
  };
  var response = JSON.parse(UrlFetchApp.fetch(url, options));
  var summary = response['responseSet'][0]['summary'][0]['text'];
  return summary;
}