How retrieve intent slot value, drive me crazy - aws-lambda

I simply want to retrieve to slot value on code, I wanna try to do a simple skill who respond in different ways according to day sopeak from users.
This is my sample code, I'm trying in blank project for don't have other issue.
The intent is "HelloWorldIntent" the slot is "day"
JSON:
{
"interactionModel": {
"languageModel": {
"invocationName": "try",
"intents": [
{
"name": "AMAZON.CancelIntent",
"samples": []
},
{
"name": "AMAZON.HelpIntent",
"samples": []
},
{
"name": "AMAZON.StopIntent",
"samples": []
},
{
"name": "HelloWorldIntent",
"slots": [
{
"name": "day",
"type": "AMAZON.DayOfWeek"
}
],
"samples": [
"good {day}",
"hello",
"how are you",
"say hi world",
"say hi",
"hi",
"say hello world",
"say hello"
]
},
{
"name": "AMAZON.NavigateHomeIntent",
"samples": []
}
],
"types": []
}
}
}
The index.js is:
const HelloWorldIntentHandler = {
canHandle(handlerInput) {
return handlerInput.requestEnvelope.request.type === 'IntentRequest'
&& handlerInput.requestEnvelope.request.intent.name === 'HelloWorldIntent';
},
handle(handlerInput) {
var app = this.event.request.intent.slots.day.value;
const speechText = app;
return handlerInput.responseBuilder
.speak(speechText)
//.reprompt('add a reprompt if you want to keep the session open for the user to respond')
.getResponse();
}
};
When I ask good Monday (or any other day) the result is Sorry, I couldn't understand what you said. Please try again.
Any suggestion?

This is your issue:
var app = this.event.request.intent.slots.day.value;
That looks like code for the v1 of the SDK.
It looks like you're actually using v2 - if you have the most recent version, to get a slot value you can do this:
// Require the SDK at the top of your file
const Alexa = require('ask-sdk');
// Then in your handle function where you want the slot value
handle (handlerInput) {
const { requestEnvelope } = handlerInput;
const speechText = Alexa.getSlotValue(requestEnvelope, 'day');
return handlerInput.responseBuilder.speak(speechText).getResponse();
}
Alternatively, you could also get it like this:
const speechText = handlerInput.requestEnvelope.request.intent.slots['day'].value
Request Envelope Utils - ASK SDK v2

Related

"Key" on a BucketAggregate is missing when the response is serialized

I am working on upgrading a service from ES 5.0 to 6.8. I have a bucket aggregate that in v5 serializes to this:
"items": [
{
"key": "random+topic",
"docCount": 27919,
"aggregations": {
"ParentReference": {
"docCount": 24992,
"aggregations": {
"Popularity": {
"value": 25223
}
}
}
}
},
{
"key": "unknown problem+latency",
"docCount": 24566,
"aggregations": {
"ParentReference": {
"docCount": 23419,
"aggregations": {
"Popularity": {
"value": 23931
}
}
}
}
},
With the v6 of Elasticsearch.Net and Nest, when serialized, I end up with:
"items": [
{
"ParentReference": {
"Popularity": {
"value": 25223
}
}
},
{
"ParentReference": {
"Popularity": {
"value": 23931
}
}
},
I had previously encountered the issue where the "aggregations" property is no longer returned (though I would love a pointer to the breaking changes announcement on that), and have updated my code accordingly. I can't do much without the Key and docCount, however. I figure there must be something related to the Json parsing changes.
I have already tried the steps in:
Custom Serialization | Elasticsearch.Net and NEST: the .NET clients [6.x] | Elastic
I have tried with the default serializer, as well as a custom one using the JsonNetSerializer.Default to no effect.
Can anyone provide a suggestion on what I should be doing?
note that this is how I am getting my BucketAggregate:
var childAgg = response.Aggregations[ss.Type] as SingleBucketAggregate;
var nestedAgg = childAgg.Aggregations[ss.Path] as SingleBucketAggregate;
var countAgg = nestedAgg.Aggregations[ssTermsName] as BucketAggregate;
return new ProviderResult<BucketAggregate>
{
Result = countAgg,
};
Thank!
~john
additional:
ElasticClient elasticClient_BuiltInSerializer = new ElasticClient();
var source = elasticClient_BuiltInSerializer.SourceSerializer.SerializeToString(o);
var response = elasticClient_BuiltInSerializer.RequestResponseSerializer.SerializeToString(o);
ConnectionSettings connectionSettings = new ConnectionSettings(new SingleNodeConnectionPool(new Uri("http://fake")), JsonNetSerializer.Default);
ElasticClient elasticClient_JsonNetSerializer = new ElasticClient(connectionSettings);
var source2 = elasticClient_JsonNetSerializer.SourceSerializer.SerializeToString(o);
var response2 = elasticClient_JsonNetSerializer.RequestResponseSerializer.SerializeToString(o);

How do i connect two types in GraphQL?

So I have two types in GraphQL:
article.mdx
---
title: "Post n.1"
category: "Category"
---
Post Content
categories.json
[
{
"name": "Category",
"description": "This is a description",
"order": 1
}
]
I want to query my post type in order to have this kind of result:
{
"node": {
"title": Post n.1
"category": {
"name": "Category",
"description": "This is a description",
"order": 1
}
}
}
How can i do this? I'm currently using GatsbyJS! Thanks.
its pretty easy as you know you should use gatsby-transformer-remark to read md files , so for the json files you should use gatsby-transformer-json , add it in the gatsby-config.js file under plugins. then you need to query your data , unfortunatly i realy dont think you can combile two files to get data as you ask , but you can try this
first in gatsby-node.js file you need to reference the variables you gonna use for filter the query data , pass those fields in to the context
exports.createPages = async function({ actions, graphql }) {
const { data } = await graphql(`
query {
allMarkdownRemark {
edges {
node {
fields {
slug
}
}
}
}
}
`)
data.allMarkdownRemark.edges.forEach(edge => {
const slug = edge.node.fields.slug
actions.createPage({
path: slug,
component: require.resolve(`./src/templates/article.js`),
context: { category: category},
})
})
}
then in your page query you can accesss the filterd query by this read more in Creating Pages from Data Pro grammatically
export const pageQuery = graphql`
query MyQuery($category: String!) {
allMarkdownRemark(filter: {frontmatter: {category: {eq: $category}}}) {
edges {
node {
frontmatter {
title
}
}
}
}
allDataJson(filter: {name: {eq: $category}}) {
edges {
node {
nodes {
name,
description,
order
}
}
}
}
}`
then access you can access the your data by const {allMarkdownRemark , allDataJson} = data
then combine those two data as you prefer
const item = {node : { title: allMarkdownRemark.edges.node[0].frontmatter }};
item.node.category = allDataJson.edges.node[0].nodes
note this was assuming that edges.node is an array so we need to exact the 1st element of your data by node[0] , please check whether this method is working .
and the structure for the json data was
{ "nodes": [ {
"name": "Category",
"description": "This is a description",
"order": 1
}
]
}

loopback REST API filter by nested data

I would like to filter from REST API by nested data. For example this object:
[
{
"name": "Handmade Soft Fish",
"tags": "Rubber, Rubber, Salad",
"categories": [
{
"name": "women",
"id": 2,
"parent_id": 0,
"permalink": "/women"
},
{
"name": "kids",
"id": 3,
"parent_id": 0,
"permalink": "/kids"
}
]
},
{
"name": "Tasty Rubber Soap",
"tags": "Granite, Granite, Chair",
"categories": [
{
"name": "kids",
"id": 3,
"parent_id": 0,
"permalink": "/kids"
}
]
}
]
is comming by GET /api/products?filter[include]=categories
and i would like to get only products which has category name "women". How do this?
LoopBack does not support filters based on related models.
This is a limitation that we have never had bandwidth to solve, unfortunately :(
For more details, see the discussion and linked issues here:
Filter on level 2 properties: https://github.com/strongloop/loopback/issues/517
Filter by properties of related models (use SQL JOIN in queries): https://github.com/strongloop/loopback/issues/683
Maybe you want to get this data by the Category REST API. For example:
GET /api/categories?filter[include]=products&filter[where][name]=woman
The result will be a category object with all products related. To this, will be necessary declare this relation on the models.
Try like this.It has worked for me.
const filter = {
where: {
'categories.name': {
inq: ['women']**strong text**
}
}
};
Pass this filter to request as path parameters and the request would be like bellow
GET /api/categoriesfilter=%7B%22where%22:%7B%categories.name%22:%7B%22inq%22:%5B%women%22%5D%7D%7D%7D
Can you share how it looks like without filter[include]=categorie, please ?
[edit]
after a few questions in comment, I'd build a remote method : in common/models/myModel.js (inside the function) :
function getItems(filter, categorieIds = []) {
return new Promise((resolve, reject) => {
let newInclude;
if (filter.hasOwnProperty(include)){
if (Array.isArray(filter.include)) {
newInclude = [].concat(filter.include, "categories")
}else{
if (filter.include.length > 0) {
newInclude = [].concat(filter.include, "categories");
}else{
newInclude = "categories";
}
}
}else{
newInclude = "categories";
}
myModel.find(Object.assign({}, filter, {include: newInclude}))
.then(data => {
if (data.length <= 0) return resolve(data);
if (categoriesIds.length <= 0) return resolve(data);
// there goes your specific filter on categories
const tmp = data.filter(
item => item.categories.findIndex(
categorie => categorieIds.indexOf(categorie.id) > -1
) > -1
);
return resolve(tmp);
})
}
}
myModel.remoteMethod('getItems', {
accepts: [{
arg: "filter",
type: "object",
required: true
}, {
arg: "categorieIds",
type: "array",
required: true
}],
returns: {arg: 'getItems', type: 'array'}
});
I hope it answers your question...

How to use the post_poll method in Zapier CLI

According to the docs, I should use a post_poll function to add the missing id field in the response.
How do I add the post_poll function ?
Here's my error:
Results must be an array, got: object,
({"totalevents":83,"events":[{"eventid":10266033,"c)
- Got a result missing the "id" property (83)
Tried following this but it is not clear to me, I'm very new to Zapier-CLI
Update - adding code
This is the function that returns the data:
const listEvents = (z, bundle) => {
console.log('listing events.. ');
let client_id = bundle.inputData.client_id;
const requestOpts = {
url: `https://wccqa.on24.com/wcc/api/v2/client/${client_id}/event`
};
return z.request(requestOpts)
.then((response) => {
return z.JSON.parse(response.content);
});
};
The sample response is the following, with the distiction that I added the id param manually to avoid errors when zapier test|push:
{
"id": 9964513,
"eventid": 9964513,
"archivestart": "2017-09-21T10:30:00-07:00",
"archiveend": "2018-09-21T10:30:00-07:00",
"description": "Zapier Event Test",
"iseliteexpired": "N",
"displaytimezonecd": "America/Bogota",
"eventtype": "Live Webcam ",
"regrequired": true,
"clientid": 22921,
"liveend": "2017-09-21T10:00:00-07:00",
"createtimestamp": "2017-09-21T09:47:44-07:00",
"audienceurl": "https://localhost.on24.com/wcc/r/9964513/C49755A02229BD48E6010848D7C81EF8",
"lastmodified": "2017-09-21T09:47:44-07:00",
"livestart": "2017-09-21T08:45:00-07:00",
"goodafter": "2017-09-21T09:00:00-07:00",
"regnotificationrequired": true,
"isactive": true,
"localelanguagecd": "en"
}
The ACTUAL response from the endpoint the following which is used in the app created in the Web Builder App instead of CLI and works fine:
{
"events": [
{
"eventid": 9964513,
"archivestart": "2017-09-21T10:30:00-07:00",
"archiveend": "2018-09-21T10:30:00-07:00",
"description": "Zapier Event Test",
"iseliteexpired": "N",
"displaytimezonecd": "America/Bogota",
"eventtype": "Live Webcam ",
"regrequired": true,
"clientid": 22921,
"liveend": "2017-09-21T10:00:00-07:00",
"createtimestamp": "2017-09-21T09:47:44-07:00",
"audienceurl": "https://localhost.on24.com/wcc/r/9964513/C49755A02229BD48E6010848D7C81EF8",
"lastmodified": "2017-09-21T09:47:44-07:00",
"livestart": "2017-09-21T08:45:00-07:00",
"goodafter": "2017-09-21T09:00:00-07:00",
"regnotificationrequired": true,
"isactive": true,
"localelanguagecd": "en"
}
],
"totalevents": 1
}
I was thinking something along the line of the following, but how do I register this ?
const postPoll = (event,z,bundle) => {
if(event.key === 'events'){
var results = z.JSON.parse(bundle.request.data).results;
var events = results.events.map(function(event){
event.id = event.eventid;
return event;
});
bundle.request.data = events;
}
};
module.exports = postPoll;
Nice, so you're almost there! CLI apps don't have pre_ and post_ poll methods. Instead, you put any manipulation after the response comes in.
const listEvents = (z, bundle) => {
console.log('listing events.. ');
let client_id = bundle.inputData.client_id;
const requestOpts = {
url: `https://wccqa.on24.com/wcc/api/v2/client/${client_id}/event`
};
return z.request(requestOpts)
.then((response) => {
return z.JSON.parse(response.content);
})
.then(data => {
const events = data.events; // array of events
return events.map(function(e){ // returns array of objects with `id` defined
e.id = e.event_id
return e
})
})
};

How to create *static* property in can.Model

I need somehow to store metadata in the can.Model
I use findAll method and receive such JSON:
{
"metadata": {
"color": "red"
},
"data": [
{ "id": 1, "description": "Do the dishes." },
{ "id": 2, "description": "Mow the lawn." },
{ "id": 3, "description": "Finish the laundry." }
]
}
I can work with data like can.Model.List, but I need metadata like a static property or something.
You can use can.Model.parseModels to adjust your response JSON before it's turned into a can.Model.List.
parseModels: function(response, xhr) {
var data = response.data;
var metadata = response.metadata;
var properties;
if(data && data.length && metadata) {
properties = Object.getOwnPropertyNames(metadata);
can.each(data, function(datum) {
can.each(properties, function(property) {
datum[property] = metadata[property];
});
});
}
return response;
}
Here's a functional example in JS Bin: http://jsbin.com/qoxuju/1/edit?js,console

Resources