I am building a bot with the botbuilder framework using node, and I am now trying to use the CardAction.dialogAction:
builder.CardAction.dialogAction(session, 'help', 'topic:mytopic', 'Click me')
It generates a message that looks like this:
action?help=topic:mytopic
Now I need to route that action to the correct dialog to handle it, but I can't figure out where and how to do that. Seeing as this is a builtin feature, I figured there should be easy ways of doing this already?
Appreciate the help.
For a lack of any other options at this point I resorted to writing my own simple action recognizer. It is no beauty, but it does the trick.
function action_recognizer() {
return {
recognize: function (context, done) {
let intent = { score: 0.0 }
const text = context.message.text
if (text) {
const m = text.match(/action\?(\w+)=?(.+)/)
logger.debug(m)
if (m) {
switch (m[1]) {
case 'help':
intent = { score: 1.0, intent: 'help' }
if (m.length > 2) {
intent.entities = [{entity: m[2], type: 'custom_intent'}]
}
break
}
}
}
done(null, intent)
}
}
}
This will basically direct to my help:/ dialog. Which in turn will read the entities list (if custom_intent types are available).
Related
I'm using Bot Framework SDK with nodejs to implement a disamibuation flow.
I want that if two intents predicted by Luis are close to each other, ask the user from which of them are the one they want. I have done the validator but, I have a problem with the flow.
It is a waterfall Dialog with 3 steps:
FirstStep: Calls Orchestrator and Luis to get intents and entities. It pass the data with return await step.next({...})
Disamiguation Step: Checks if it is necessary to disambiguate, and, in that case, prompts the options. If not, it pass the data like the first step.
Answer step: If it has a disambiguation flag in the data it receives in step.result, it prompts the answer acordingly with the user response. Elsewhere, it uses the data in step.result that comes from the first step.
The problem is that, when it prompts user to say the intent, I lost the data of the FirstStep since I cannot use step.next({...})
¿How can I maintain both the data from the first step and the user answer in the prompt?
Here are the basic code:
async firstStep(step) {
logger.info(`FinalAnswer Dialog: firstStep`);
let model_dispatch = await this.bot.get_intent_dispatch(step.context);
let result = await this.bot.dispatchToTopIntentAsync(step.context, model_dispatch.model)
// model_dispatch = orchestrator_model
// result = {topIntent: String, entities: Array, disamibiguation: Array}
return await step.next({ model_dispatch: model_dispatch, result: result})
}
async disambiguationStep(step) {
logger.info(`FinalAnswer Dialog: disambiguationStep`);
if (step.result.result.disambiguation) {
logger.info("We need to disambiguate")
let disambiguation_options = step.result.result.disambiguation
const message_text = "What do you need";
const data = [
{
"title": "TEXT",
"value": disambiguation_option[0]
},
{
"title": "TEXT",
"value": disambiguation_option[1]
},
]
let buttons = data.map(function (d) {
return {
type: ActionTypes.PostBack,
title: d.title,
value: d.value
}
});
const msg = MessageFactory.suggestedActions(buttons, message_text);
return await step.prompt(TEXT_PROMPT, { prompt: msg });
return step.next(step.result) //not working
}
else {
logger.info("We dont desambiguate")
return step.next(step.result)
}
}
async answerStep(step) {
logger.info(`FinalAnswer Dialog: answerStep`);
let model_dispatch = step.result.model_dispatch
let result = step.result.result
//Show answer
return await step.endDialog();
}
You can use the step dictionary to store your values. The complex dialogs sample on GitHub is excellent for demonstrating this. https://github.com/microsoft/BotBuilder-Samples/blob/main/samples/javascript_nodejs/43.complex-dialog/dialogs/topLevelDialog.js
You can save data in the context with whatever name you want:
step.values['nameProperty'] = {}
This will be accessible within the entire execution context of the waterfall dialog:
const data = step.values['nameProperty'] // {}
Using TipTap, I'm trying to avoid adding a <br />, but create a <p></p> instead, with the focus inside that <p>|</p> when the user hit shift-Enter but I can't make it work.
Here's what I did so far:
new (class extends Extension {
keys () {
return {
'Shift-Enter' (state, dispatch, view) {
const { schema, tr } = view.state
const paragraph = schema.nodes.paragraph
console.log(tr.storedMarks)
const transaction = tr.deleteSelection().replaceSelectionWith(paragraph.create(), true).scrollIntoView()
view.dispatch(transaction)
return true
}
}
}
})()
How can I do this?
I don't know if this is still relevant but as I was looking for the same thing, I found two ways to make this work.
NOTE:
I'm using tiptap v2, if that's not a problem, then:
I overrode the HardBreak extension, since it's the one that use the Shift-Enter keybinding. It looks something like;
const CustomHardBreak = HardBreak.extend({
addKeyboardShortcuts() {
return {
"Mod-Enter": () => this.editor.commands.setHardBreak(),
"Shift-Enter": () => this.editor.commands.addNewline(),
};
},
});
And used it like so;
editor = new Editor({
extensions: [
customNewline,
CustomHardBreak,
]
});
Use the default editor command createParagraphNear. E.g this.editor.commands.createParagraphNear()
I tried creating a custom extension from your code and ended up with something similar to the command above, i.e;
export const customNewline = Extension.create({
name: "newline",
priority: 1000, // Optional
addCommands() {
return {
addNewline:
() =>
({ state, dispatch }) => {
const { schema, tr } = state;
const paragraph = schema.nodes.paragraph;
const transaction = tr
.deleteSelection()
.replaceSelectionWith(paragraph.create(), true)
.scrollIntoView();
if (dispatch) dispatch(transaction);
return true;
},
};
},
addKeyboardShortcuts() {
return {
"Shift-Enter": () => this.editor.commands.addNewline(),
};
},
});
And added this as an extension in my editor instance.
PS:
They both work, almost exactly the same, I haven't found a difference yet. But there's somewhat of a 'catch' if you would call it that; Both these methods don't work on empty lines/nodes, a character has to be added before the cursor for it to work, any character, even a space.
In TipTap 2.0 I am able to use this custom extension:
const ShiftEnterCreateExtension = Extension.create({
addKeyboardShortcuts() {
return {
"Shift-Enter": ({ editor }) => {
editor.commands.enter();
return true;
},
};
},
});
To make shift + enter behave like enter.
In my case I actually wanted enter to do something different. So I use prosemirror events to set a ref flag on whether shift was pressed. Than I check that flag under the "Enter" keyboard event -- which could be triggered normally or through the shift + enter extension.
My code has a simple card carousel which has action button like below:
actions = [
{
"type": "Action.Submit",
"title": "Qualify",
"data": { "action" : "qualify_lead" }
},
{
"type": "Action.OpenUrl",
"title": "Retire",
"url": "{viewUrl}"
},
{
"type": "Action.ShowCard",
"title": "Add Note",
"card": this.noteCard(item.LeadId, "leads")
}
]
I am having a method to handle qualify_lead action like below
async qualifyLead(context:any){
console.log("in qualifyLead:" + JSON.stringify(context.activity))
await context.updateActivity(this.successCard('Lead is qualified'))
}
All I am doing on purpose is to replace entire carousel with a simple text message. But it fails with error:
Error: BotFrameworkAdapter.updateActivity(): missing activity.id
Where do i get this ?
I am using google firebase for this and the wrapper code is like below
const {
TurnContext,
TeamsActivityHandler,
CardFactory,
AttachmentLayoutTypes,
ActionTypes
} = require('botbuilder');
class TeamsConversationBot extends TeamsActivityHandler {
constructor() {
super();
this.leadState =
this.conversationState.createProperty('leadCarouselState');
this.onMessage(async (context:any, next:any) => {
TurnContext.removeRecipientMention(context.activity);
let msg = context.activity.text
const action = context.activity.value
let objNum = ''
let keyword = ''
if(msg === undefined && action === undefined)
msg = 'help'
else if(msg !== undefined){
msg = msg.trim().toLowerCase()
if(msg.indexOf("help") > -1)
msg = 'help'
else{
if(msg.startsWith('lead')){
msg = 'lead'
}
}
}
switch (msg) {
case 'lead':
await this.lead(context, userKey, platform)
break;
case 'qualify_lead':
await this.qualifyLead(context)
break;
}
await next();
});
}
I'm not sure exactly what this.successCard('Lead is qualified') does, but presumably it returns an Activity. To my knowledge, in order for this Activity to replace another one, you need to set it's Id property to match the previous message. That means that, when you send the previous message (i.e. the card), you need to capture the reference that's returned from the send method on the context (e.g. into bot state), and then use it for this new activity.
As I explained in my answer to your other question, you need to save the activity ID in bot state and then apply it to the update that you're sending. The Bot Framework can't update an activity unless you tell it which activity to update, and you do that using an activity ID.
This was the part that saves the ID:
const dict = await this.carouselState.get(turnContext, {});
dict[batchId] = {
[KEYACTIVITYID]: response.id,
[KEYLEADS]: leads
};
And this was the part that applies it to the updated activity:
const update = this.createCarousel(batchId, leads);
update.id = info[KEYACTIVITYID];
I am trying to do begindialog from another another dialog js file. I am getting error.
<b>[onTurnError]: Error: DialogContext.beginDialog(): A dialog with an id of 'FollowUpDialog' wasn't found. </b>
this is dialog structure-
dialogs
orderstatus
orderstatus.js
index.js
FollowUp
followUp.js
index.js
i am trying to include FollowUp dialog in OrderStatus Dialog, similary i have other dialogs where i want to begin followup or orderstatus dialog. trying to reuse the dialogs.
One way to do use how we include the file in botjs amd to do adddialog same way i can include in otherfile. But it is redundant work. I am trying to avoid that. Can some one tell me better approach to include the dialog in different dialogs.
code:
Below code is from greeting.js
If you see line where i am doing begindialog.
return await step.beginDialog(ORDER_STATUS_DIALOG);
return await step.beginDialog(ENTITLEMENT_CHECK_DIALOG);
It is error. I am trying to include the dialog which is part of different JS files.
async step => {
if (step.context.activity.channelId == 'directline') {
const buttons = [{
type: ActionTypes.ImBack,
title: 'Repair Order Status',
value: symbolicString.ZEB_GR_STR_013
}, {
type: ActionTypes.ImBack,
title: 'Entitlement Check',
value: symbolicString.ZEB_GR_STR_014
}];
const card = CardFactory.heroCard('', undefined, buttons, {
text: symbolicString.ZEB_GR_STR_015
});
const reply = {type: ActivityTypes.Message};
reply.attachments = [card];
return await step.context.sendActivity(reply);
} else {
return await step.prompt(MENU_PROMPT, symbolicString.ZEB_GR_STR_028);
}
},
async step => {
if (step.context.activity.channelId != 'directline'){
console.log("step greeting dialog next step");
console.log(step);
console.log(step.context.activity);
if (step.context.activity.text == '1'){
return await step.beginDialog(ORDER_STATUS_DIALOG);
}else if (step.context.activity.text == '2'){
return await step.beginDialog(ENTITLEMENT_CHECK_DIALOG);
}
}
return await step.endDialog();
}
]));
I need to use a WCF API to save data into a DB. Ordinarily, I'd use chaining, like the example below:
IClientBroker clientBroker = UIContext.CreateWcfInterface<IClientBroker>("Data/ClientBroker.svc");
clientBroker.BeginSetClientBusinessName(_client.ID, businessName, (result) =>
{
_client = ((IClientBroker)result.AsyncState).EndSetClientBusinessName(result);
clientBroker.BeginSetClientAddress(_client.ID, addressObservableCollection, postcodeZip, (result2) =>
{
_client = ((IClientBroker)result2.AsyncState).EndSetClientAddress(result2);
clientBroker.BeginSetClientTelephone(_client.ID, telephone, (result3) =>
{
_client = ((IClientBroker)result3.AsyncState).EndSetClientTelephone(result3);
clientBroker.BeginSetClientFax(_client.ID, fax, (result4) =>
{
_client = ((IClientBroker)result4.AsyncState).EndSetClientFax(result4);
if (customFields.Save(validationSummaryBridge))
{
CloseWindow(true, "ClientID=" + _client.ID.ToString());
}
else
{
validationSummary.Errors.Add(new ValidationSummaryItem("Failed to save Custom Fields"));
}
}, clientBroker);
}, clientBroker);
}, clientBroker);
}, clientBroker);
}
This gives me faux-synchronous behaviour which I need so exceptions are thrown in a timely fashion and I can react on validation events.
This doesn't map well, however, when I have a loop of fields to save. For example, what pattern would be best to save the following list of "Custom Fields", where each Custom Field must be saved using a single WCF call?
ICustomFieldsBroker customFieldsBroker = UIContext.CreateWcfInterface<ICustomFieldsBroker>("Data/CustomFieldsBroker.svc");
foreach (CustomField customField in _customFields)
{
string newValue=_customFieldControlDictionary[customField].CustomField.Value;
customFieldsBroker.BeginSetCustomFieldValueForItem(DataTypeID, DataItemID, customField.Key, newValue, (result) =>
{
((ICustomFieldsBroker)result.AsyncState).EndSetCustomFieldValueForItem(result);
}, customFieldsBroker);
}
In the above example, this would just set off, say, 5 requests to the WCF API/threads which would potentially return AFTER the form has closed. I need them to "line up", so I can list their status and return to the form.
Thanks very much.
Don't let the WCF distract you, but if you have any comments, do let me know. :)
This is the answer I was looking for:
http://www.netfxharmonics.com/2008/11/Understanding-WCF-Services-in-Silverlight-2#WCFSilverlightThreadWaiting