Authorization
API Resources
NL Interpreter
Search
Assist
SDK JS-Client
Introduction
Instead of loading the SDK, you can use this helper in javascript to use the API.
Setup
Before you use the js-client, see Finding your API credentials in the Help Center to obtain your credentials. These credentials are project-specific. For more information, see the Authorization section.
Please review the API documentation carefully before using the helper.
Note
All examples shown are set in a Production environment.
Integration
-
First, include the javascript file somewhere on your page:
-
If you need compatibility with IE11 browsers, use the version with polyfills:
<script src="https://sdk.inbenta.io/chatbot/1/inbenta-chatbot-sdk.js"></script> -
If you do not need compatibility with IE11 browsers, you can use the version without polyfills:
<script src="https://sdk.inbenta.io/chatbot/1/inbenta-chatbot-sdk-no-polyfills.js"></script>
-
-
Then, proceed with the authorization, just as if you were using the SDK.
Caution
Your Secret must remain confidential. To prevent third parties from accessing your API, always perform this request in a server-side environment. Never expose your secret in client-side integrations.
After performing the authorization, you get an access token with the following format:
AuthResponse={ <your_access_token> <expiration> };
Creation
If you completed the authorization server-side, you can now initiate the API client:
You can now define the API Configuration, which sets the headers, environment, userType and bot configuration. For more information, see the /conversation endpoint section.
The following example shows how to define which Chat App attributes the API returns after a user asks a question, and the maximum number of related contents that should appear:
APIConfig = {
answers:{
answerAttributes: ['ANSWER_TEXT'],
sideBubbleAttributes: ['SIDEBUBBLE_TEXT'],
maxRelatedContents: 3
},
}; APIPromise = InbentaChatbotSDK.apiClient(<AuthResponse>,<APIConfig>,<your_API_key>);Using the client
Now that the API client is created, you need to wait for the API Promise to be ready, which will complete the set-up returning the chatbotAPI object:
APIPromise.then((chatbotAPI) => {
//API calls start here
}You can then execute startConversation, which will use the bot configuration used to create the API Client. For more information on available configurations, see the payload/conversation endpoint section.
chatbotAPI.startConversation().then(function()
{
//start asking the API with conversation method
}Finally, after completing the session set-up, you can start asking the API questions:
ChatbotAPI.conversation('My first question');Full Application example code
<!DOCTYPE html>
<html>
<head>
<meta charset="UTF-8">
<title>Demo</title>
<meta name="viewport" content="width=device-width, initial-scale=1.0, maximum-scale=1.0, user-scalable=0" />
<script src="https://sdk.inbenta.io/chatbot/1/inbenta-chatbot-sdk.js"></script>
</head>
<body>
</body>
</html>
<script>
var accessToken={
accessToken:<your_access_token> //Response form server-side Authorization
};
botConfig = {
environment: "development",
lang:'en',
answers:{
answerAttributes: ['ANSWER_TEXT'],
sideBubbleAttributes: ['SIDEBUBBLE_TEXT'],
maxRelatedContents: 3
},
};
var queryForPolarQuestion = 'baggage';
var queryForMultipleChoiceQuestion = 'special meals';
var queryForSimpleAnswer = 'vegan meals';
APIPromise = InbentaChatbotSDK.apiClient(accessToken,botConfig,"<?echo $key?>");
APIPromise.then((chatbotAPI)=> {
chatbotAPI.startConversation().then(function(response){
// set a variable
chatbotAPI.addVariable('first_name', 'NewName', true);
// send a message
chatbotAPI.conversation(queryForSimpleAnswer).then(function(response){
console.log(response.data);
// get the last answer
var lastAnswer = response.data;
// if there are many answers choose only the last one
if(response.data.answers) {
lastAnswer = response.data.answers[response.data.answers.length-1];
}
// console.log(lastAnswer);
switch (lastAnswer.type) {
case "multipleChoiceQuestion":
console.log("multiple choice question received");
// choose one option randomly
var options = lastAnswer.options;
var choosedOption = options[Math.floor(Math.random() * options.length) % options.length];
console.log(choosedOption);
chatbotAPI.conversation(choosedOption.label, choosedOption.value).then(function(response){
console.log("'"+choosedOption.label+"' selected");
console.log(response.data);
// rate the content
if(response.data.parameters.contents.trackingCode) {
chatbotAPI.rateContent(response.data.parameters.contents.trackingCode.rateCode, 1);
}
});
break;
case "polarQuestion":
console.log("polar question received");
// choose yes
var choosedOption = "yes";
chatbotAPI.conversation(choosedOption, choosedOption).then(function(response){
console.log("'yes' selected");
console.log(response.data);
});
break;
case "answer":
console.log("simple answer received");
break;
default:
break;
}
});
});
});
</script>
API Config
The client configuration will affect all the methods.
| Name | Type | Default | Description |
|---|---|---|---|
| userType | integer | 0 | Defines the profile identifier from the Chat App's Knowledge Base. It will be used in the client as a x-inbenta-user-type header to make API requests. For more information about this header, see the API Routes. |
| environment | String | production | Defines source environment from the Chat App's Knowledge Base. It will be used in the client as a x-inbenta-env header. For more information about this header, see the API Routes. |
Methods
The methods page is available here.
Libraries and Browser Support
The JavaScript SDK depends on:
- vue 2.7.14
- axios 0.26.1
- CORS
The following browsers are supported:
- IE 11+ (only in the Chat SDK version with polyfills)
- Last version of self-updating browsers:
- Chrome
- Firefox
- Safari
- Edge