Skip to content

A JS adapter to query Intento API – a single API to Cognitive AI services from many vendors

License

Notifications You must be signed in to change notification settings

intento/intento-nodejs

Repository files navigation

Intento SDK for Node.js

An adapter to query Intento API.

Intento provides a single API to Cognitive AI services from many vendors.

To get more information check out the site.

API User Manual

In case you don't have a key to use Intento API, please register here inten.to

Installation

npm install intento-nodejs

or

yarn add intento-nodejs

Command line interface

See CLI example as one of the ways to use this SDK.

Try it in a browser

Requires serve to be installed globally in your system

yarn global add serve
yarn build # it will prepare minified files, see ./dist folder
yarn test-in-browser
# Visit http://localhost:5000/samples/browser-app

Go to an example browser app at http://localhost:5000/samples/browser-app

Basic usage

Initialise the client

const IntentoConnector = require('intento-nodejs')
const client = new IntentoConnector({ apikey: YOUR_INTENTO_KEY })

Translation

This is an intent to translate text from one language to another.

Related documentation

More examples in the sample app

Simple translate text text to language to:

  • source language will be detected automatically
  • provider for the translation will be smart-selected based on the Smart routing feature
client.ai.text.translate
    .fulfill({ text: "How's it going?", to: 'es' })
    .then(data => {
        console.log('Translation results:\n', data, '\n\n')
    })

Sentiment analysis

This is an intent to analyze the sentiment of the provided text.

Related documentation

More examples in the sample app

client.ai.text.sentiment
    .fulfill({
        text: 'We love this place',
        lang: 'en',
        provider: 'ai.text.sentiment.ibm.natural_language_understanding',
    })
    .then(data => {
        console.log('Sentiment analysis results:\n', data, '\n\n')
    })

Text meanings

This is an intent to get meanings of text in selected language.

Related documentation

More examples in the sample app

client.ai.text.dictionary
    .fulfill({
        text: 'meaning',
        from: 'en',
        to: 'ru',
    })
    .then(data => {
        console.log('Dictionary results:\n', JSON.stringify(data, null, 4), '\n')
    })

Explore providers (basics)

Response structure for provider-related requests

In all cases a response object is a list of objects. Each object in that list describes one provider. The structure of the description is following:

{
    id: 'provider.id',
    name: 'Provider Name',
    score: 0,
    price: 0,
    symmetric: [...],
    pairs: [...],
}

symmetric - is a list of language codes for which translation in both directions is available.

pairs - is a list of plain objects with structure { from: 'lang-code-1', to: 'lang-code-2' }. It means that for current provider translation from lang-code-1 to lang-code-2 is available.

List all available providers

Translation providers

client.ai.text.translate
    .providers()
    .then(data => data.forEach(p => console.info(p.name)))

Sentiment analysis providers

client.ai.text.sentiment
    .providers()
    .then(data => data.forEach(p => console.info(p.name)))

Text meanings providers

client.ai.text.dictionary
    .providers()
    .then(data => data.forEach(p => console.info(p.name)))

Translation capabilities

More information in ai.text.translate.md

Sentiment analysis capabilities

More information in ai.text.sentiment.md

Text meanings capabilities (dictionary)

More information in ai.text.dictionary.md

Smart routing

Intento provides the smart routing feature, so that the translation request is automatically routed to the best provider. The best provider is determined based on the following information:

  • apriori benchmark on the standard dataset
  • provider usage statistics, collected by Intento, including user feedback signals (the post-editing complexity for Machine Translation).

Basic smart routing

To use the smart routing, just omit the provider parameter:

client.ai.text.translate
    .fulfill({
        text: "How's it going?",
        to: 'es'
    })
    .then(console.log)

Response:

{
    "id": "XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX",
    "done": true,
    "response": [
        {
            "results": [
                "¿Qué tal va todo?"
            ],
            "meta": {
                "detected_source_language": [
                    "en"
                ]
            },
            "service": {
                "provider": {
                    "id": "ai.text.translate.deepl.api",
                    "name": "DeepL API",
                    "vendor": "DeepL",
                    "description": "API",
                    "logo": "https://inten.to/static/img/api/deepl_translate.png"
                }
            }
        }
    ],
    "meta": {
        // information about initial request
    },
    "error": null
}

Specifying a custom routing strategy

By default, when the provider is missing, requests are routed to a provider with the best expected price/performance ratio. This behavior may be controlled by specifying the desired routing strategy in the routing parameter. To set up routing for your account contact us at [email protected].

client.ai.text.translate
    .fulfill({
        text: 'A sample text',
        to: 'es',
        routing: 'best-quality',
    })
    .then(console.log)

Failover mode

Both for smart routing mode and basic mode, a failover is supported. By default, the failover is off, thus when the selected provider fails, an error is returned. To enable the failover mode, set the failover to true. By default, failover is governed by the default routing strategy (best). To control this behavior, another routing strategy may be specified via routing parameter. Alternatively, you may specify a list of providers to consider for the failover (failover_list). This option overrides the routing strategy for the failover procedure.

In the following example we set the provider, but specify the list of alternatives to control the failover:

client.ai.text.translate
    .fulfill({
        text: 'A sample text',
        to: 'es',
        provider: 'ai.text.translate.microsoft.translator_text_api.2-0',
        failover: true,
        failover_list: [
            'ai.text.translate.google.translate_api.2-0',
            'ai.text.translate.yandex.translate_api.1-5'
        ]
    })
    .then(console.log)

Using a service provider with your own keys

Intento supports two modes of using 3rd party services:

  • full proxy: 3rd party service used via Intento and paid to the Intento (available for some of the services).
  • tech proxy: 3rd party service used via our user's own credentials and billed towards the user's account at the third-party service (available for all services).

In the tech proxy mode, the custom credentials are passed in the auth service field. auth field is a dictionary, it's keys are provider IDs. For each ID specify your own key(s) you want to use and values set to a list of keys for the specified provider. There could be more than one key for the provider in the case you want to work with a pool of keys (more on this advanced feature later).

client.ai.text.translate
    .fulfill({
        text: "A sample text",
        to: 'es',
        provider: 'some-provider-id',
        auth: {
            'some-provider-id': [
                { ... custom auth structure with 'YOUR_KEY_TO_SOME_PROVIDER_1' ... },
                { ... custom auth structure with 'YOUR_KEY_TO_SOME_PROVIDER_2' ... },
            ],
            'another-provider-id': [
                { ... another custom auth structure with 'YOUR_KEY_TO_ANOTHER_PROVIDER_1' ... },
                { ... another custom auth structure with 'YOUR_KEY_TO_ANOTHER_PROVIDER_2' ... },
            ]

    })
    .then(console.log)

Auth object structure is different for different providers and may be obtained together with other provider details by requesting info for this provider:

client.ai.text.translate
    .provider('ai.text.translate.google.translate_api.2-0')
    .then(console.log)

For example for google translate custom auth structure is { key: YOUR_GOOGLE_KEY }.

client.ai.text.translate
    .fulfill({
        text: "A sample text",
        to: 'es',
        provider: 'ai.text.translate.google.translate_api.2-0',
        auth: {
            'ai.text.translate.google.translate_api.2-0': [
                {
                    key: process.env.YOUR_GOOGLE_KEY
                }
            ]
        }
    })
    .then(console.log)

Response:

{
    "id": "XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX",
    "done": true,
    "response": [
        {
            "results": [
                "Un texto de muestra"
            ],
            "meta": {
                "detected_source_language": [
                    "en"
                ]
            },
            "service": {
                "provider": {
                    "id": "ai.text.translate.google.translate_api.2-0",
                    "name": "Google Cloud Basic Translation API",
                    "vendor": "Google Cloud",
                    "description": "Basic Translation API",
                    "logo": "https://inten.to/static/img/api/ggl_translate.png"
                }
            }
        }
    ],
    "meta": {
        // information about initial request
    },
    "error": null
}

Advanced Examples

Dynamic parameters

Describe data dynamicly through content

For example we'd like to translate the same text for different languages. One can make similar requests for each language like this:

const IntentoConnector = require('intento-nodejs')
const client = new IntentoConnector({ apikey: process.env.INTENTO_API_KEY })

const options = {
    path: '/ai/text/translate',
    method: 'POST',
    // content: ...
}

['es', 'ru', 'da'].forEach(lang => {
    client
        .makeRequest({
            ...options, // path, method
            content: {
                context: {
                    text: 'A sample text',
                    to: lang,
                },
            },
        })
        .then(console.info)
})

So you can pass a plain javascript object as a content parameter.

Using data argument from a curl request directly

One can pass request parameters as raw json specified as a data parameter. Make sure your json is valid. For example one can validate json online

For a curl instruction from the docs

curl -XPOST -H 'apikey: YOUR_API_KEY' 'https://api.inten.to/ai/text/translate' -d '{
    "context": {
        "text": "Validation is the king",
        "to": "es"
    },
    "service": {
        "provider": "ai.text.translate.microsoft.translator_text_api.2-0"
    }
}'

run following:

const IntentoConnector = require('intento-nodejs')
const client = new IntentoConnector({ apikey: process.env.INTENTO_API_KEY })

client
    .makeRequest({
        path: '/ai/text/translate',
        method: 'POST',
        data: `{
            "context": {
                "text": "Validation is the king",
                "to": "es"
            },
            "service": {
                "provider": "ai.text.translate.microsoft.translator_text_api.2-0"
            }
        }`,
    })
    .then(console.info)

More examples

To get more examples:

git clone [email protected]:intento/intento-nodejs.git
cd intento-nodejs
INTENTO_API_KEY=YOUR_SECRET_KEY node example.js

Though for the latter you need to have INTENTO_API_KEY in your environment.

How to pass your API keys to your environment

zero option (dev only)

Hardcode your keys in the script your are experimenting with :)

1st option

For Unix-like machines run:

INTENTO_API_KEY=YOUR_SECRET_KEY GOOGLE_API_KEY=YOUR_GOOGLE_API_KEY node example.js

For Windows machines run:

SET INTENTO_API_KEY=YOUR_SECRET_KEY GOOGLE_API_KEY=YOUR_GOOGLE_API_KEY node example.js

2nd option

Assuming your are in intento-nodejs folder run:

cp .env.example .env

Then edit .env, put your api keys there.

Make that environmental variables available in the current context

export $(cat .env) # once per terminal

Run example.js script

node example.js

3rd option

Use your favorite "env" package like dotenv or node-env-file.