I am implementing a dialogRootBot that will call a skillDialogBot.
My dialogRootBot is able to invoke the skillDialogBot. The conversation can progress until the point where the skillDialogBot is supposed to return the results to the dialogRootBot.
The skillDialogBot has the following setup
this.addDialog(new TextPrompt(TEXT_PROMPT))
.addDialog(new ConfirmPrompt(CONFIRM_PROMPT))
.addDialog(new WaterfallDialog(WATERFALL_DIALOG, [
this.processStep.bind(this)
]));
The processStep is laid out like this
async processStep(stepContext) {
const details = stepContext.options;
details.result = {
status: 'success'
};
return await stepContext.endDialog(stepContext.options);
}
I was expecting the dialogRootBot to get the result from the skillDialogBot after processStep has called endDialog, but that never happens. Instead, the user is stuck with the skillDialogBot until the user manually types in "abort", which is the command the dialogRootBot is monitoring to cancel all dialogs in its onContinueDialog() implementation
Here is how the onContinueDialog() looks like
async onContinueDialog(innerDc) {
const activeSkill = await this.activeSkillProperty.get(innerDc.context, () => null);
const activity = innerDc.context.activity;
if (activeSkill != null && activity.type === ActivityTypes.Message && activity.text) {
if (activity.text.toLocaleLowerCase() === 'abort') {
// Cancel all dialogs when the user says abort.
// The SkillDialog automatically sends an EndOfConversation message to the skill to let the
// skill know that it needs to end its current dialogs, too.
await innerDc.cancelAllDialogs();
return await innerDc.replaceDialog(this.initialDialogId, { text: 'Request canceled!' });
}
}
return await super.onContinueDialog(innerDc);
}
I modeled this after the botbuilder-samples\samples\javascript_nodejs\81.skills-skilldialog sample. If I were to change the skillDialogBot and have it do a ConfirmPrompt() before the finalStep()'s endDialog(), then the conversation ends correctly with the skillDialogBot() posting the dialog's results to the rootDialogBot.
For the sake of clarity, this is how the bookingDialog in the skills-skilldialog sample looks like
/**
* Confirm the information the user has provided.
*/
async confirmStep(stepContext) {
const bookingDetails = stepContext.options;
// Capture the results of the previous step.
bookingDetails.travelDate = stepContext.result;
const messageText = `Please confirm, I have you traveling to: ${ bookingDetails.destination } from: ${ bookingDetails.origin } on: ${ bookingDetails.travelDate }. Is this correct?`;
const msg = MessageFactory.text(messageText, messageText, InputHints.ExpectingInput);
// Offer a YES/NO prompt.
return await stepContext.prompt(CONFIRM_PROMPT, { prompt: msg });
}
/**
* Complete the interaction and end the dialog.
*/
async finalStep(stepContext) {
if (stepContext.result === true) {
const bookingDetails = stepContext.options;
return await stepContext.endDialog(bookingDetails);
}
return await stepContext.endDialog();
}
Is it not possible to endDialog() without a prompt? Why is my skillDialogBot unable to endDialog and pass the results to the rootDialogBot?
Thank You
For the sake of reference, when using stepContext.beginDialog(), it appears that we cannot reliably endDialog() if the waterfall dialog does not have a prompt step added. If there is a use case where we want to use a skillDialogBot and call a specific dialog in the skillDialogBot via stepContext.beginDialog(), and the called dialog is only doing processing (eg. call REST APIs) without the need to prompt users for further information, at the end of the processing, we can opt to end the conversation instead.
Hence, assuming we have finalStep() bound to the waterfall dialog as the last step, just use the following in finalStep()
return await stepContext.context.sendActivity({
type: ActivityTypes.EndOfConversation,
code: EndOfConversationCodes.CompletedSuccessfully,
value: stepContext.options
});
At least for my use case, I was able to terminate the dialog without needing a user prompt and got the results back to the rootDialog.
Related
I am mocking navigator functions for simple clipboard functionality. Here is the relevant code:
// FUNCTION
/**
* Adds a click event to the button which will save a string to the navigator clipboard. Checks for
* clipboard permissions before copying.
*/
function loader(): void {
async function copyUrl(): Promise<void> {
const permission = await navigator.permissions.query({ name: "clipboard-write" });
if (permission.state == "granted" || permission.state == "prompt" ) {
await navigator.clipboard.writeText("the url");
} else {
console.error('Permission not supported');
}
}
const button = document.querySelector('button') as HTMLElement;
button.addEventListener('click', async () => {
await copyUrl();
});
}
// TEST
it('works', () => {
// mock navigator functions
Object.assign(navigator, {
permissions: {
query: jest.fn(async () => ({ state: "granted" }))
},
clipboard: {
writeText: jest.fn(async () => {})
}
});
// initialize DOM
document.body.innerHTML = '<button></button>';
loader(); // adds the event listener
// click the button!
const button = document.querySelector('button') as HTMLElement;
button.click();
expect(navigator.permissions.query).toHaveBeenCalledTimes(1);
expect(navigator.clipboard.writeText).toHaveBeenCalledWith('the url');
});
The test fails on expect(navigator.clipboard.writeText).toHaveBeenCalledWith('the url') with:
Expected: "the url" Number of calls: 0
Defeats the purpose of permissions, yes, but for the sake of debugging:
Try adding a clipboard call before permissions call like so?
// FUNCTION
// ...
async function copyUrl(): Promise<void> {
// add this
await navigator.clipboard.writeText('the url');
// keep the rest still
const permission = await navigator.permissions.query({ name: "clipboard-write" });
// ...
}
This fails on the first assertion now, expect(navigator.permissions.query).toHaveBeenCalledTimes(1) with
Expected number of calls: 1 Received number of calls: 0
With the addition above, I also changed the assertions to be:
expect(navigator.clipboard.writeText).toHaveBeenCalledWith('the url');
expect(navigator.clipboard.writeText).toHaveBeenCalledTimes(2);
expect(navigator.permissions.query).toHaveBeenCalledTimes(1);
... which failed on the second assertion because it expected 2 calls but only received 1.
I have been testing in a VSCode devcontainer and tried out the extension firsttris.vscode-jest-runner to debug the test. With breakpoints in the loader function, I'm able to see that every single line executes perfectly with my mockup but still fails at the end of debug.
I even changed the mock navigator.permissions.query function to return { state: 'denied' } instead. Both running and debugging, it did not satisfy the permission check and gave an error to the console as expected but the test still failed at expect(navigator.permissions.query).toHaveBeenCalledTimes(1) (with the added writeText call before it).
It seems to me that after the first call of a mock function, the others just don't work.
Am I missing something? Send help pls lol
EDITS
Using jest.spyOn as in this answer has the same issues.
Using an async test with an expect.assertions(n) assertion still produces the exact same issue.
I have a chatbot that will eventually be deployed on multiple websites, and there are a number or variables that need to change based on the site (e.g. language, QnA Database, Dialog, etc.). I'd like to do this with a single bot, and just pass a variable so that it knows which page it is being rendered on (for a simple example, let's assume country pages: us, fr, de, etc.). I have been unsuccessful in passing this information to the bot.
Ideally this would be before the welcome message fires, but I can't even get it to send at all. I have a custom store set up:
const store = window.WebChat.createStore({}, function(dispatch) { return function(next) { return function(action) {
if (action.type === 'WEB_CHAT/SEND_MESSAGE') {
// Message sent by the user
PageTitleNotification.Off();
clearTimeout(interval);
} else if (action.type === 'DIRECT_LINE/INCOMING_ACTIVITY' && action.payload.activity.name !== "inactive") {
// Message sent by the bot
clearInterval(interval);
interval = setTimeout(function() {
// Change title to flash the page
PageTitleNotification.On('Are you still there?');
// Notify bot the user has been inactive
dispatch.dispatch({
type: 'WEB_CHAT/SEND_EVENT',
payload: {
name: 'inactive',
value: ''
}
});
}, 300000)
}
return next(action);
}}});
But for my use case I don't think what's in there actually matters, only that it is defined. The functions here just 1) clear an interval when the user sends a message and 2) set a new interval and send an inactivity message to the bot.
I also have a send message activity that is on a button click for a transcript. It looks like this:
document.querySelector('#transcriptButton').addEventListener('click', function() {
return store.dispatch({
type: 'WEB_CHAT/SEND_MESSAGE',
payload: { text: 'Email me a transcript' }
});
/*return store.dispatch({
type: 'WEB_CHAT/SEND_EVENT',
payload: {
name: 'siteContext',
value: 'eatonchatbot indexBackup.html'
}
});*/
});
This sends a "front channel" message (that I can see in the bot) to request a transcript, which kicks off a dialog. That works. The commented out section alludes to what I'm trying to do. I have a separate dispatch statement as shown below, which has the exact same SEND_EVENT code as is commented out above. The SEND_EVENT does work as expected when it keys off the button click.
Here is the additional code I added. This is the piece that is NOT working. What I want is, when the bot has been rendered (but ideally before the welcome message), send this siteContext event to the bot so that I know where the bot is being rendered. I do not get any activity in the bot with this code. I also tried replacing it with SEND_MESSAGE instead of SEND_EVENT in a sort of reverse test from above, but that didn't work either.
// Test setting site context
store.dispatch({
type: 'WEB_CHAT/SEND_EVENT',
payload: {
name: 'siteContext',
value: 'eatonchatbot indexBackup.html'
}
});
/*store.dispatch({
type: 'WEB_CHAT/SEND_MESSAGE',
payload: {
text: 'eatonchatbot indexBackup.html'
}
});*/
It just occurred to me that this statement is probably running before the bot is rendered. So I put it in an interval and this DOES work. However, it does not fire the message until after the welcome message has been sent.
setTimeout(function() {
store.dispatch({
type: 'WEB_CHAT/SEND_EVENT',
payload: {
name: 'siteContext',
value: 'eatonchatbot indexBackup.html'
}
});
}, 5000);
So this kind of works, but if this siteContext value was needed to determine the language of the welcome message, this would obviously fail. So my main ask here is, is there a better way to try to pass in a siteContext value like this, or is there some way to ensure that the context is received and can be used by the bot before the welcome message fires? I do see that there is a locale setting in the renderWebChat method, but I can't figure out if and how I could access that in the bot, and besides it may not be granular enough depending on the business need. But it seems if I could send some sort of value in that renderWebChat object, that might avoid all of the other crazy stuff I'm trying to do.
With some help from #Hessel and this issue I found on GitHub, I was able to come up with a solution. Just setting the values being passed in via onEvent (which I am now using in place of onTurn to reduce an if statement) isn't good enough if you need to alter content in the welcome message (e.g. language, user name, or an altogether different message). The onMembersAdded still fires before the values can be set, at least if you're setting them in userState. The key is to set up separate welcome messages in onEvent for directline and onMembersAdded for all other channels (I didn't include webchat as in the example as I'm not sending any event for that channel).
Here is the onEvent function I used:
this.onEvent(async (context, next) => {
// Check for inactivity
if (context.activity.name && context.activity.name === 'inactive') {
await context.sendActivity({
text: 'Are you still there? Is there anything else I can help you with?',
name: 'inactive'
});
}
// Check for webchat/join event (directline conversation initiation)
if (context.activity.name && context.activity.name === 'webchat/join') {
const userData = await this.userDialogStateAccessor.get(context, {});
userData.siteContext = context.activity.value;
// Debug
console.log(`The current language is: ${userData.siteContext.language}`);
console.log(`The current page is: ${userData.siteContext.page}`);
//await context.sendActivity(`The current language is: ${userData.siteContext.language}`);
//await context.sendActivity(`The current page is: ${userData.siteContext.page}`);
if (!userData.accountNumber) {
const dc = await this.dialogs.createContext(context);
await dc.beginDialog(AUTH_DIALOG);
await this.conversationState.saveChanges(context);
await this.userState.saveChanges(context);
} else {
if (context.activity.channelId == 'msteams') {
var welcomeCard = CardHelper.GetMenuCardTeams(welcomeMessage,'Y','Y');
} else {
var welcomeCard = CardHelper.GetMenuCard(welcomeMessage,'Y','Y');
}
await context.sendActivity(welcomeCard);
this.appInsightsClient.trackEvent({name:'conversationStart', properties:{accountNumber:userData.accountNumber}});
}
await this.userState.saveChanges(context);
}
// By calling next() you ensure that the next BotHandler is run.
await next();
});
The event I used in the custom store is pretty much the same as above, except I updated it to pull in the most preferred language and current url (was hard coded above).
store = window.WebChat.createStore({}, function (dispatch) {
return function (next) {
return function (action) {
if (action.type === 'DIRECT_LINE/CONNECT_FULFILLED') {
dispatch.dispatch({
type: 'WEB_CHAT/SEND_EVENT',
payload: {
name: 'webchat/join',
value: {
language: navigator.languages[0],
page: window.location.href
}
}
});
}
return next(action);
};
};
});
If you have your renderWebChat method inside a function that you can call so that your bot doesn't automatically start (I have a floating icon that causes the bot to load onclick) this should go outside that function.
Azure BotFramework
SDK4
NodeJS
In dialog state I am using 'return await step.prompt' inside async function. But once user enters the value it is not considering the user input value as a input for prompt instead it is going to luisrecognizer for the match.
I have written the similar code for different dialog where it works fine.
Kindly requesting you provide some valuable input.
Azure BotFramework
SDK4
NodeJS
this.addDialog(new WaterfallDialog('OrderStatusDialog', [
async function(step) {
return await step.context.sendActivity(reply);
},
async function(step) {
var cardValue = step.context.activity.value;
if (cardValue) {
if (cardValue.action == "Order_Number") {
return await step.prompt('textPrompt', 'enter order number');
} else if (cardValue.action == "Serial_Number") {
return await step.prompt('textPrompt', 'enter serial number');
} else {
}
}
return await step.next();
// return await step.endDialog();
},
async function (step) {
var cardValue = step.context.activity;
console.log("****** cardValue");
console.log(cardValue);
console.log("****** step After");
console.log(step);
return await step.endDialog();
}
]));
at prompt step it should read the value what user is entering. Also stack is empty when i console the step ..
Unfortunately, you have a few issues with your code. But, hopefully this will help iron them out. I had no issues running the below code.
One, double check which version of Botbuilder you have installed. At least one call ('this.addDialog') I am unfamiliar with, doesn't appear to be a part of the current SDK, and didn't work for me when testing.
Two, configure your bot like below. Technically, you should be able pass the various steps in as you have them in your question. However, something about yours was not work and I couldn't figure out what. The below setup DOES work, however, and conforms to already accepted practices.
Three, your first step calls 'step.context.sendActivity(reply)'. In the next step you are trying to read the returned value of that call. That isn't going to work as sendActivity simply sends a statement from the bot to the user ("Welcome to my bot!", for example). You need to perform a prompt to capture the user's input response (see below).
It looks like you are trying to read a value from a card. Since you didn't supply that bit of code, I faked the value by supplying the 'Order_Number' and 'Serial_Number' via a text prompt from the user.
You should also take advantage of the bot state options. Rather than use the variable 'cardValue', consider using UserState or DialogState to store user values important to the conversation.
Lastly, in this simple example, order and serial numbers will overwrite each other if you pass thru multiple times.
const START_DIALOG = 'starting_dialog';
const DIALOG_STATE_PROPERTY = 'dialogState';
const USER_PROFILE_PROPERTY = 'user';
const ACTION_PROMPT = 'action_prompt';
const ORDER_PROMPT= 'order_prompt';
const SERIAL_PROMPT= 'serial_prompt';
...
class ExampleBot {
constructor(conversationState, userState) {
this.conversationState = conversationState;
this.userState = userState;
this.dialogState = this.conversationState.createProperty(DIALOG_STATE_PROPERTY);
this.userData = this.userState.createProperty(USER_PROFILE_PROPERTY);
this.dialogs = new DialogSet(this.dialogState);
this.dialogs
.add(new TextPrompt(ACTION_PROMPT))
.add(new TextPrompt(ORDER_PROMPT))
.add(new TextPrompt(SERIAL_PROMPT));
this.dialogs.add(new WaterfallDialog(START_DIALOG, [
this.promptForAction.bind(this),
this.promptForNumber.bind(this),
this.consoleLogResult.bind(this)
]));
}
async promptForAction(step) {
return await step.prompt(ACTION_PROMPT, `Action?`,
{
retryPrompt: 'Try again. Action??'
}
);
}
async promptForNumber(step) {
const user = await this.userData.get(step.context, {});
user.action = step.result;
if (user) {
if (user.action == 'Order_Number') {
return await step.prompt(ORDER_PROMPT, 'enter order number');
} else if (user.action == 'Serial_Number') {
return await step.prompt(SERIAL_PROMPT, 'enter serial number');
} else {
}
}
return await step.next();
}
async consoleLogResult(step) {
const user = await this.userData.get(step.context, {});
user.orderNumber = step.result;
console.log('****** cardValue');
console.log(user);
console.log('****** step After');
console.log(step);
return await step.endDialog();
}
async onTurn(turnContext) {
... //Other code
// Save changes to the user state.
await this.userState.saveChanges(turnContext);
// End this turn by saving changes to the conversation state.
await this.conversationState.saveChanges(turnContext);
}
}
Hope of help!
somehow it worked for me after saving conversation state where ever i am replacing the dialog.
await this.conversationState.saveChanges(context);
I've created a chatbot using Node.js and the flow of dialogs works fine til the endDialog. Im having issues implementing a back option so it can jump back only to previous dialog. Can anyone please guide me how to solve that?
.reloadAction(
"restartBenefits", "Ok. Let's start over.",
{
matches: /^start over$|^restart$/i
}
)
.cancelAction(
"cancelRequest", "Thank you for reaching out, Good bye!",
{
matches: /^nevermind$|^cancel$|^cancel.*request/i,
confirmPrompt: "This will cancel your request. Are you sure?"
}
);
Use a customAction
In this case you can listen for whatever you want to be a keyword for "back" and then simply route the user back to that dialog using replaceDialog
bot.customAction({
matches: /back|last/gi, //whatever prompts you want to match.
onSelectAction: (session, args, next) => {
session.replaceDialog(PreviousDialog); //variable with the last dialog in it, set that somewhere, such as the end of your previous dialog
}
})
I think that at the final step of the dialog waterfall you need to add the last lines in this sample step:
/**
* This is the final step in the main waterfall dialog.
* It wraps up the sample "book a flight" interaction with a simple confirmation.
*/
async finalStep(stepContext) {
// If the child dialog ("bookingDialog") was cancelled or the user failed to confirm, the Result here will be null.
if (stepContext.result) {
const result = stepContext.result;
// Now we have all the booking details.
// This is where calls to the booking AOU service or database would go.
// If the call to the booking service was successful tell the user.
const timeProperty = new TimexProperty(result.travelDate);
const travelDateMsg = timeProperty.toNaturalLanguage(new Date(Date.now()));
await stepContext.context.sendActivity(ActivityFactory.fromObject(this.lgTemplates.evaluate('BookingConfirmation', {
Destination: result.destination,
Origin: result.origin,
DateMessage: travelDateMsg
})));
}
// =====> HERE: Restart the main dialog with a different message the second time around
return await stepContext.replaceDialog(this.initialDialogId, { restartMsg: 'What else can I do for you?' });
}
Just change the this.initialDialogId accordingly.
The context is MS Bot Framework on Node. I need to custom validate an input from a Prompt.text or Prompt.Number. I have been taking a look at examples and docs. I have implemented a quick example with DialogAction.validatePrompt(), which works ok but has the problem that I cannot (or at least I don't know how) customize the message in case the validation fails.
The validation could fail for many reasons and it would be great to be able to choose a response message based on the failure reason.
Also I have seen the basics-custom-prompt example in: (https://github.com/Microsoft/BotBuilder/tree/master/Node/examples/basics-customPrompt) and it uses and IntentDialog to get the desired behavior. It also mentions it is a replacement for the basics-validatedPrompt example.
My questions are:
Which one is recommend to use, validatedPrompt or the IntentDialog
approach?
Is the validatedPrompt() going to be deprecated?
Does the validatedPrompt() provide a mechanism for custom message?
Microsoft Bot Framework Version 4 Prompt Validation Example link is specified in below :
https://github.com/Microsoft/BotBuilder-Samples/tree/master/samples/javascript_nodejs/10.prompt-validations
For Proper Text Prompt Validation is you create separate file which perform validation task
Code: namePrompt/index.js
const { TextPrompt } = require('botbuilder-dialogs');
module.exports.NamePrompt = class NamePrompt extends TextPrompt {
constructor(dialogId) {
super(dialogId, async (prompt) => {
if (!prompt.recognized.succeeded) {
await prompt.context.sendActivity('Please tell me your name.');
return false;
} else {
const value = prompt.recognized.value;
if (value.length < 1) {
await prompt.context.sendActivity('Your name has to include at least one character.');
return false;
} else if (value.length > 50) {
await prompt.context.sendActivity(`Sorry, but I can only handle names of up to 50 characters. Yours was ${ value.length }.`);
return false;
} else {
return true;
}
}
});
}
};
Now inside your current dialog you can import above file:
const { NamePrompt } = require('../../prompts/namePrompt');
const GET_NAME_PROMPT = 'namePrompt';
this.addDialog(new NamePrompt(GET_NAME_PROMPT));
inside your dialog step you can prompt such like this:
async promptForName(step) {
return await step.prompt(GET_NAME_PROMPT, `What is your name, human?`);
}