How to connect Bot Framework dialog with api.ai client

Question

I'm creating a bot using Bot Framework in C#

I have this piece of code :

  var faq = await result;

  if (faq == "Faq with menu")
  {
      await context.PostAsync("Under construction");
  }
  else if (faq == "Faq with dialog")
  {
      context.Call(new FaqDialog(), this.ResumeAfterOptionDialog);
  }

Faq with dialog I have connected with a dialog class.

I want to connect Faq with menu with my client in Api.ai. Do you have any idea how to do it?

Answer 1

What I would do is to create an enum with the Faq values:

Public enum Faq{
 Undefined,
 Menu,
 Dialog
}

Then create a method that will call Api.ai with the user message and map the intent response to the enum:

  public T MatchAiIntent<T>(string message) where T : struct, IConvertible
    {
        if (!typeof(T).IsEnum)
        {
            throw new ArgumentException("T must be an enum type!");
        }

    T result = default(T);
    try
    {
        var response = apiAi.TextRequest(message);
        var intentName = response?.Result?.Metadata?.IntentName;

        if (intentName == null)
        {
            return result;
        }

        Enum.TryParse<T>(intentName, true, out result);

        return result;
    }
    catch (Exception exception)
    {
        //logit
        throw;
    }
} 

Then you can use it in your code:

var response = MatchAiIntent(faq);
if (response == Faq.Menu)
  {
      await context.PostAsync("Under construction");
  }

Answer 2

[UPDATE]

CONNECTING TO Dialogflow (previously known as API.AI) FROM C#

Follow these steps (working example in C#)

1. After you create a Dialogflow agent go to the agent's settings --> General --> click on the Service Account link
2. You will be sent to to google cloud platform where you can create a service account
3. After you create a service account, there will be an option to create a KEY, create it and download the (JSON) format of it
4. This key will be used to connect from your C# project to the Dialogflow agent
5. Install Google.Cloud.Dialogflow.V2 package in your project

6. Create for example a Dialogflow manager class (check below for an example)

       public class DialogflowManager {
       private string _userID;
       private string _webRootPath;
       private string _contentRootPath;
       private string _projectId;
       private SessionsClient _sessionsClient;
       private SessionName _sessionName;
   
       public DialogflowManager(string userID, string webRootPath, string contentRootPath, string projectId) {
   
           _userID = userID;
           _webRootPath = webRootPath;
           _contentRootPath = contentRootPath;
           _projectId = projectId;
           SetEnvironmentVariable();
   
       }
   
       private void SetEnvironmentVariable() {
           try {
               Environment.SetEnvironmentVariable("GOOGLE_APPLICATION_CREDENTIALS", _contentRootPath + "\\Keys\\{THE_DOWNLOADED_JSON_FILE_HERE}.json");
           } catch (ArgumentNullException) {
               throw;
           } catch (ArgumentException) {
               throw;
           } catch (SecurityException) {
               throw;
           }
       }
   
       private async Task CreateSession() {
           // Create client
           _sessionsClient = await SessionsClient.CreateAsync();
           // Initialize request argument(s)
           _sessionName = new SessionName(_projectId, _userID);
   
       }
   
       public async Task < QueryResult > CheckIntent(string userInput, string LanguageCode = "en") {
           await CreateSession();
           QueryInput queryInput = new QueryInput();
           var queryText = new TextInput();
           queryText.Text = userInput;
           queryText.LanguageCode = LanguageCode;
           queryInput.Text = queryText;
   
           // Make the request
           DetectIntentResponse response = await _sessionsClient.DetectIntentAsync(_sessionName, queryInput);
           return response.QueryResult;
       }
   }
   

7. And then this can be called like this for example to get detect Intents

        DialogflowManager dialogflow = new DialogflowManager("{INSERT_USER_ID}",
       _hostingEnvironment.WebRootPath,
       _hostingEnvironment.ContentRootPath,
       "{INSERT_AGENT_ID");
   
   var dialogflowQueryResult = await dialogflow.CheckIntent("{INSERT_USER_INPUT}");