Collecting the Responses

Warning

The startOcr Promise resolves with a JSON string, on both Android and iOS. You must call JSON.parse() before accessing the fields. The declared TypeScript type is OCR_RETURN_VALUES | string.

Response structure

type OCR_RETURN_VALUES = {
  DocumentRecognitionResponse: {
    ocr_key?: string;
    ocr_front_key?: string;
    ocr_back_key?: string;
    document_type: string;
  }[];
};

Attribute	Type	Description
ocr_key	string	Image identification key, present for single-capture documents (cnh_full, cnh_digital, proof_of_address, rg_cin_digital). It can be used in any other service of the QI Tech system.
ocr_front_key	string	Identification key of the front image, present for dual-capture documents (cnh, rg, rne, crnm).
ocr_back_key	string	Identification key of the back image, present for dual-capture documents (cnh, rg, rne, crnm).
document_type	string	Identifies which document that key refers to (e.g. "cnh", "rg", "proof_of_address").

Important

Store the returned keys — they are the image identifier in the other products of the QI Tech system, such as the Onboarding API.

Single-capture document

{
  "DocumentRecognitionResponse": [
    { "ocr_key": "5d0f0e1c-8f9e-4b6c-9c3d-2f8a1b4e7c10", "document_type": "cnh_full" }
  ]
}

Dual-capture document

{
  "DocumentRecognitionResponse": [
    { "ocr_front_key": "5d0f0e1c-8f9e-4b6c-9c3d-2f8a1b4e7c10", "document_type": "cnh" },
    { "ocr_back_key": "a1b2c3d4-5e6f-4a7b-8c9d-0e1f2a3b4c5d", "document_type": "cnh" }
  ]
}

Reading the response

startOcr(CAAS_ENVIRONMENT.SANDBOX, mobileToken, CAAS_DOCUMENT_TYPE.CNH)
  .then((response) => {
    const { DocumentRecognitionResponse } = JSON.parse(response as string);

    if (DocumentRecognitionResponse.length === 1) {
      const { ocr_key, document_type } = DocumentRecognitionResponse[0];
      console.log('Ocr key: ' + ocr_key + ' document type: ' + document_type);
    } else if (DocumentRecognitionResponse.length === 2) {
      const { ocr_front_key } = DocumentRecognitionResponse[0];
      const { ocr_back_key } = DocumentRecognitionResponse[1];
      console.log('Ocr front key: ' + ocr_front_key + ' / Ocr back key: ' + ocr_back_key);
    }
  })
  .catch((error) => {
    console.log('Error executing Ocr. Error: ' + error);
  });

Error handling

The Promise is rejected when the user interrupts the flow or when the native SDK encounters an error.

Situation	Message
User canceled the flow	User canceled OCR
Error in the native iOS SDK	Error executing OCR: <detail>
Native module not linked	The package 'react-native-qi-tech-module' doesn't seem to be linked...

Warning

If you get the linking error, check that you ran pod install on iOS, that you rebuilt the app after installing the package, and that you are not in the Expo managed workflow without prebuild.