DocumentDetector

Importing SDK

To use DocumentDetector, you can either remotely import the .js file or download it locally.

Remotely

Include the .js file directly from the CDN:

<script
  src="https://repo.combateafraude.com/javascript/release/document-detector/<VERSION>.umd.js"
  type="text/javascript"
></script>

You can retrieve the class from the SDK using the following code:

const { DocumentDetectorSdk } = window["@combateafraude/document-detector"];

Locally

Download the .js file and import it as an ES6 module:

import { DocumentDetectorSdk } from "../assets/js/document-detector-<VERSION>.js";

Construction

In the builder, the SDK receives a single parameter with the settings:

const sdk = new DocumentDetectorSdk(options);

Supported parameters

Parameter	Required?
token Authentication token for consuming the SDK.	Yes.
language Default message language, valid values: en_US, en_BR, es_MX.	No. The default is pt_BR
analyticsSettings Analytics Configuration Objects.	No.
analyticsSettings.disableAnalytics Parameter responsible for enabling or disabling analytics.	No.
analyticsSettings.trackingId Unique id that we are going to save the information from this SDK execution.	No.
analyticsSettings.trackingInfo We accept an information object.	No.
environmentSettings.disableDesktopExecution Flag indicating whether execution on desktops should be blocked.	No. The default is false
capturerSettings.disableAdvancedCapturing Flag indicating whether advanced capture should be disabled*.	No. The default is false
appearenceSettings.captureButtonIcon Customization of the capture icon accepts values such as image URL or base64 SVGs.	No
appearenceSettings.captureIconSize Customization of the icon size for the captureButtonIcon field.	No
appearenceSettings.captureButtonColor Customization of the default image capture button color.	No
appearenceSettings.switchButtonIcon Customization of the camera switch icon accepts values such as image URL or base64 SVGs.	No
appearenceSettings.switchIconSize Customization of the icon size for the switchButtonIcon field.	No
appearenceSettings.switchIconColor Customization of the color of the default camera switching icon.	No
appearenceSettings.fontFamily Changes the font for all elements contained in the SDK.	No. The pattern is inherited from the page
textSettings.messages.processMessage Customization of the image processing message.	No. The default is "Processando sua foto, aguarde um momento"
textSettings.messages.wrongDocumentMessage Customization of the invalid document message.	No. The default is "Este não é o documento esperado"
textSettings.messages.bothWrongSideMessage Customization of the incorrect side message by selecting a front or back side and capturing both sides.	No. The default is "Por favor, realize a captura com o documento fechado e com o lado correto para a câmera"
textSettings.messages.wrongSideMessage Customization of the incorrect side message.	No. The default is "Este não é o lado esperado para este documento"
textSettings.messages.lowQualityMessage Customization of the API return message for poor image quality.	No. The default is "A qualidade da captura não ficou legal. Certifique-se que está em um ambiente iluminado e tente novamente"
textSettings.messages.captureFailedMessage Customization of the capture failure message.	No. The default is "Ops! Tivemos um problema ao processar sua imagem."

* Advanced capture consists of using more complex and not-so-stable APIs in browsers that support them (e.g. ImageCapture)

Example

const sdk = new DocumentDetectorSdk({
    token: `my-sdk-token`,

    language: `pt_BR`,

    analyticsSettings: {
        disableAnalytics: false,
        trackingId: '',
        trackingInfo: '',
    },

    environmentSettings: {
        disableDesktopExecution: false,
    },

    appearenceSettings: {
        hideCaptureTitle: false,
        hideCaptureMask: false,
        hideCameraSwitchButton: false,
        useGenericMask: false,
    },

    capturerSettings: {
        disableAdvancedCapturing: false,
    },

    textSettings: {
        messages: {
            processMessage: '',
            wrongDocumentMessage: '',
            bothWrongSideMessage: '',
            wrongSideMessage: '',
            lowQualityMessage: '',
            captureFailedMessage: '',
        }
});

CaptureStage

CaptureStage allows the client to configure the stages. To do this, we offer the CaptureStage object, where you can set the following parameters:

Parameter
mode Desired capture mode. It can be used manual, automatic or upload. In the manual capture, a button will be enabled for the user to do the capture, in the upload, the functionality of uploading documents will be displayed instead of capturing.
attempts The number of attempts of the current stage. If it is the only stage, the value 0 must be passed.
duration Duration time of the current stage. If more than one stage is set, the total time for each stage can be set, and when the total time is reached, the stage will move on to the next one. Set it to 0 if you don't want to set a time for the stage.

Example CaptureStage

const stages = [
  { mode: "automatic", attempts: 3, duration: 60 },
  { mode: "manual", attempts: 3, duration: 60 },
  { mode: "upload", attempts: 0, duration: 0 },
];

Initialization

initialize(): Promise<void>

The SDK has a separate method of initialization, to allow greater control over when it occurs.

During this initialization, the SDK will initialize its internal variables and download the resources it needs to run.

[!] You must call this method before using other SDK methods.

[!] The initialization of the SDK can take a few seconds. We recommend that you call this function as early as possible in your flow so that opening the SDK is smooth for the user.

Example

await sdk.initialize();

Utilization

Opening and capturing documents

capture(container: HTMLElement, stages, {type: SupportedDocumentType, side: DocumentSide, totalAttempts?: Number}): Promise<Result>

The method used to load the SDK onto the screen and perform document capture.

It will initialize the video stream (requesting permissions if needed) and load it into the container.

Parameters

Parameter	Type	Valid Values
type Type of document to be captured.	string	any¹, rg, cnh, crlv, rne, passaporte, ctps, other²
side⁶ Type of document to be captured.	string	front, back, both⁴, not_applicable

¹ The type any does not have the validations for quality and document type, thus accepting any side or document - recommended for unsupported documents such as the OAB card.

² The type other corresponds to other documents (not including those already specified), such as vaccination cards, etc.

³ If not specified, a default value of 30 seconds is used

⁴ The side both correspond to the document showing both sides in the same photo (e.g. open CNH)

⁵ The side parameter depends on the type of document entered in type. See the table below.

Example

// div or another element on DOM
const sdkContainer = document.getElementById("sdk-displayer");
await sdk.capture(sdkContainer, stages, {
  documentType,
  documentSide,
  totalAttempts,
});

Accepted sides for each document type

Document type	Accepted sides
rg	front, back, both
cnh	front, back, both
rne	front, back
rnm	front, back
crlv	not_applicable
ctps	front, back
passport	both
any	not_applicable
other	front, back, both

Return

The return consists of an object with the following fields:

Field	Type	Description
imageUrl	string	Temporary link to the image, generated by our API
imageKey	string	Temporary key for the image
blob	Blob	Blob of the captured image
documentType	SupportedDocumentType	Type of captured document
documentSide	DocumentSide	Side of the captured document

Example

const result = sdk.capture(sdkContainer, stages, {
  documentType,
  documentSide,
  totalAttempts,
});
// { imageUrl: '[link of image]', blob: Blob, documentType: 'rg', documentSide: 'front' }

Close SDK

close(): Promise<void>

The method used to remove the SDK on the screen, removing the visual elements of the SDK from the DOM.

Will remove the visual elements of the SDK from the DOM.

De-initialize the SDK

dispose(): Promise<void>

The method used to remove the SDK on the screen.

Will de-initialize the video stream and clear the SDK's internal variables

Complete Example

Soon.