The following section shows some best practices for using the LivePerson Functions platform.

• Using Async/Await in Functions
• Making Functions configurable

Using Async/Await in Functions

There’s a special syntax to work with promises in a more comfortable fashion, called “async/await”. It's recommended to use this notation, because it provides more readable code and bypasses then..catch-chains. It’s surprisingly easy to understand and use. The word async before a function means one simple thing: a function always returns a promise. Other values are wrapped in a resolved promise automatically. The keyword await makes JavaScript wait until that promise settles and returns its result.

It is possible to use the async/await functionality within LivePerson Functions. Proper error handling is important when using this. If an async call inside of a LivePerson function has no proper error handling, it could be executed longer than the maximum execution time. In such a case a Lambda Execution is taking too long error will be thrown.

If a promise resolves normally, then await promise returns the result. But in the case of a rejection, it throws the error, just as if there were a throw statement at that line. In real situations, the promise may take some time before it rejects. In that case there will be a delay before await throws an error.

We can catch that error using try..catch, the same way as a regular throw:

try {
  await myPromise();
} catch(error) {
  // Handle error or pass it to the callback
  callback(error, null)
} 

The example shows the HTTP-Template with async/await.

async function lambda(input, callback) {
  // Importing the FaaS Toolbelt
  const { Toolbelt } = require("lp-faas-toolbelt");
  // Obtain an HTTPClient Instance from the Toolbelt
  const httpClient = Toolbelt.HTTPClient(); // For API Docs look @ https://www.npmjs.com/package/request-promise

  const URL = "https://www.liveperson.com";

  try {
    const response = await httpClient(URL, {
      method: "GET", // HTTP VERB
      headers: {}, // Your Headers
      simple: false, // IF true => Status Code != 2xx & 3xx will throw
      resolveWithFullResponse: true //IF true => Includes Status Code, Headers etc.
    });
      
    const { statusCode, body } = response;
    // Perform Action based on Status Code
    switch (statusCode) {
      case 200: 
        return callback(null, statusCode);
      // If not Whitelisted Proxy will reject with 403. Body contains also a message indicating that
      case 403:
        return callback(new Error("Domain is not whitelisted"), null);
      default:
        return callback(
          new Error(`Recieved unexpected status code ${statusCode}`)
        );
    }
  } catch (err) {
    callback(err, null)
  }
}

Making Functions configurable

It's suggested to use Environment Variables to make a function more configurable. With this approach you can separate static variables and configurations from the code, to make it more clean and readable.

For example if you have a function which is triggered by the Messaging Conversation End-Event and you want to process the code only if the closeReason is equal to AGENT or CONSUMER then it's a good use case to save this configuration to an Environment Variable (e.g. ['AGENT', 'CONSUMER']).

function lambda(input, callback) {
  const conversation = input.payload;
  const closeReason = conversation.end.closeReason
  if (process.env['CUSTOM_CLOSE_REASON'].includes(closeReason)) {
    // Some processing...
  }
  callback(null, "OK");
}

See Environment Variables for more information about it.