Botium-connector-watson NPM

Botium Connector for IBM Watson Assistant

This is a Botium connector for testing your IBM Watson Assistant chatbots.

Did you read the Botium in a Nutshell articles ? Be warned, without prior knowledge of Botium you won't be able to properly use this library!

How it works ?

Botium uses the IBM Watson Assistant API to run conversations.

It can be used as any other Botium connector with all Botium Stack components:

This connector processes info about NLP. So Intent/Entity asserters can be used.

Requirements

Node.js and NPM
a IBM Watson Assistant chatbot, and user account with administrative rights
a project directory on your workstation to hold test cases and Botium configuration

Install Botium and Watson Connector

When using Botium CLI:

> npm install -g botium-cli
> npm install -g botium-connector-watson
> botium-cli init
> botium-cli run

When using Botium Bindings:

> npm install -g botium-bindings
> npm install -g botium-connector-watson
> botium-bindings init mocha
> npm install && npm run mocha

When using Botium Box:

Already integrated into Botium Box, no setup required

Connecting IBM Watson Assistant to Botium

You need IBM Cloud credentials (Username/Password or API Key) - see IBM Docs on how to get it.

Open the file botium.json in your working directory and add the secret:

{
  "botium": {
    "Capabilities": {
      "PROJECTNAME": "<whatever>",
      "CONTAINERMODE": "watson",
      "WATSON_WORKSPACE_ID": "<watson workspace id>",
      "WATSON_APIKEY": "<ibm cloud api key>"
    }
  }
}

To check the configuration, run the emulator (Botium CLI required) to bring up a chat interface in your terminal window:

> botium-cli emulator

Botium setup is ready, you can begin to write your BotiumScript files.

Using the botium-connector-watson-cli

This connector provides a CLI interface for importing convos and utterances from your Watson workspace and convert it to BotiumScript.

Intents and user examples are converted to BotiumScript utterances and convo files (using the import command and the --buildconvos or --buildentities option)
Entities and synonyms are converted to BotiumScript utterances and convo files (using the import command and the --buildentities option)
User Conversations are downloaded and converted to BotiumScript convos or just a plain list for analytics (using the importlogs command)

You can either run the CLI with botium-cli (it is integrated there), or directly from this connector (see samples/convoV1/package.json for some examples):

> botium-connector-watson-cli import
> botium-connector-watson-cli importlogs --watsonformat convo
> botium-connector-watson-cli importlogs --watsonformat intent

Please note that a botium-core installation is required

For getting help on the available CLI options and switches, run:

> botium-connector-watson-cli import --help
> botium-connector-watson-cli importlogs --help

Watson Assistant Context Handling

When using BotiumScript, you can do assertions on and manipulation of the Watson Assistant context variables.

Asserting context variables

For asserting context variables, you can use the JSON_PATH asserter:

#bot
JSON_PATH $.context.skills['main skill'].user_defined.lightonoff|off

Depending on your Watson Assistant skill structure, this may different - but by default, this should work

Initializing context variables

You can use the WATSON_INITIAL_CONTEXT capability to initialize the context in your botium.json:

"WATSON_INITIAL_CONTEXT"{
  "initcontext1": "initcontext1value",
  "initcontext2": "initcontext2value"
}

Adding context variables

For adding a context variable within a test case, you have to use the UPDATE_CUSTOM logic hook. This example will set two context variables (when using V2 API), one to a plain string, the other one to a JSON object:

#me
play some jazz music
UPDATE_CUSTOM SET_WATSON_CONTEXT|skills['main skill'].user_defined.mycontext1|botium
UPDATE_CUSTOM SET_WATSON_CONTEXT|skills['main skill'].user_defined.mycontext2|{"nested": "botium"}

The parameters are: 1. SET_WATSON_CONTEXT 2. The path to the context variable 3. The value of the context variable

Removing context variables

For removing a context variable, the same logic hook is used (when using V2 API):

#me
play some jazz music
UPDATE_CUSTOM UNSET_WATSON_CONTEXT|skills['main skill'].user_defined.mycontext1
UPDATE_CUSTOM UNSET_WATSON_CONTEXT|skills['main skill'].user_defined.mycontext2

The parameters are: 1. UNSET_WATSON_CONTEXT 2. The path to the context variable

Usage behind a corporate proxy

In case you have an HTTPS proxy, set the HTTPS_PROXY environment variable

> HTTPS_PROXY=my-proxy-address:port npm test

If you have an HTTP proxy, Botium has to tunnel the HTTPS traffic to Watson Assistant services over HTTP. Set the WATSON_HTTP_PROXY_HOST and WATSON_HTTP_PROXY_PORT capabilities in botium.json (see below).

Supported Capabilities

Set the capability CONTAINERMODE to watson to activate this connector.

WATSON_ASSISTANT_VERSION

Default: V1

Watson supports two Assistant SDK versions, V1 and V2.

With V1, Botium accesses a workspace (or Skill) directly
With V2, Botium accesses an assistant wrapping a versioned skill

WATSON_URL

Default: "https://api.us-south.assistant.watson.cloud.ibm.com"

Has to be set to the URL shown in the Skill API details page (e.g. https://api.us-east.assistant.watson.cloud.ibm.com) - for a list of valid IBM Cloud URLs see IBM Docs.

WATSON_HTTP_PROXY_HOST / WATSON_HTTP_PROXY_PORT

Hostname/IP Address and port of your HTTP proxy

This is only required if you have a HTTP proxy only. For HTTPS proxies, you can use the HTTPS_PROXY environment variable

WATSON_VERSION

Default: "2018-09-20"

WATSON_APIKEY *

IAM API Key for IBM Watson - see IBM Docs how to create it for your IBM Watson account. Either the IAM API Key or the Service credentials (see below) are required.

WATSON_BEARER *

IBM Watson instances using the Cloud Pak For Data Platform do not have IAM API Keys, but can instead be authenticated using the bearer token found within the service instance details.

WATSON_USER and WATSON_PASSWORD

Service credentials for your IBM Watson instance - see here how to create them for your IBM Watson account.

WATSON_WORKSPACE_ID *

The Skill ID to use (Workspace ID). You can find it in the IBM Watson Assistant Dashboard when clicking on "View API Details" in the popup menu of a skill in the Skills overview list.

This is only supported for Assistant SDK V1

WATSON_ASSISTANT_ID *

The Assistant ID to use. You can find it in the IBM Watson Assistant Dashboard when clicking on "Settings" in the popup menu of an assistant in the Assistants overview list.

This is only supported for Assistant SDK V2

WATSON_FORCE_INTENT_RESOLUTION

Default: false If this capability is disabled, then a response will be dropped if the connector does not recognizes any component like text or button in it. But the dropped message has NLP recognition info like intent and entities, which could be checked.

WATSON_COPY_WORKSPACE

Default: false

This capability will copy the Watson workspace and run the Botium script on the new workspace (and delete it afterwards). Typically, when running a large amount of tests on production conversation service, the Watson workspace should not get "polluted" with test data - enabling this capability will prevent that.

This is only supported for Assistant SDK V1

Attention: as the copied workspace will run through Watson training session, it could take some time until the copied workspace is available. Botium will only continue after training is complete