Document Understanding Cloud APIs error response codes

Depending on your use case, you can use two types of calls for Document Understanding^TM Cloud APIs: synchronous (sync) and asynchronous (async).

Common issues concerning Document Understanding^TM Cloud APIs are outlined in this page.

Handling errors for sync calls

If errors occur, the synchronous call will return an appropriate HTTP status code, accompanied by:

• Error Code: specific code that identifies the type of error.
• Error Message: detailed explanation that describes what the error is about.

Handling errors for async calls

In case of errors, the resulting URL provides the following:

• Error Code: specific code that identifies the type of error.
• Error Message: detailed explanation that describes what the error is about.

If the operationId (returned by the start operation) is found, the results route will persistently return a 202 status (applicable even to unsuccesful requests) and the details of the error will be in the body of the response.

For additional details and examples, check the Swagger documentation.

General error codes

Chec the list of error codes and their corresponding messages for general failures:

OperationIdNotFoundError - Operation Id not found.
DatabaseConcurrencyError - Unable to update job status for the given input.
RequestAbortedByClient - Request aborted by client
TooManyRequests - Rate limit exceeded. Try again later. For details around the applicable limitations, check out the official documentation.
DeploymentUnavailableError - The project version this resource is part of is not available. Please check the project deployment version status and try again.
DeploymentTagNotFoundError - There is no project version tag that matches your request.
DeploymentDocumentTypeNotFoundError - The requested project version deployment does not contain an extraction model for the requested document type.
UnexpectedInternalServerError - Internal Server Error. Please contact the UiPath support team.
ServiceUnavailableError - Service Unavailable. Please retry in a few moments.

Discovery

Document Understanding projects error codes

In some cases, a failure related to Document Understanding projects may result in an error message. The error code and the corresponding message can have one of the following values:

DiscoveryResourceNotFoundError - Resource not found.
DuCenterProjectNotFound - Du Center project not found.
DocumentIdNotFound - Cannot perform the operation for the given documentId: Ensure it is correct, the digitization is successful (retrieving the digitization result), and not more than 7 days since the digitization call passed (case in which, it expired).
DocumentIdInvalid - Required input DocumentId is missing or invalid.
DocumentTypeIdNotFound - Document Type Id not found in the given project.
ProjectVersionNotSupportedError - Project Version is not supported for classic projects.
ModernProjectExtractorRequestTooLargeError - Maximum number of pages per document exceeded for the given custom trained extractor.

Digitization

Digitization failed due to client error

In some cases, a failed digitization due to a client error may result in an error message. This is a 400 error, which is displayed as follows: Code: [DigitizationErrorCode], Message: "DigitizationErrorMessage". The error code and the corresponding message can have one of the following values:

[UnsupportedContentTypeError]", "Content type of the input document is not supported."
[UnexpectedPdfStructureError]", "Invalid or corrupt PDF structure."
[InvalidImageSizeError]", "Image size of the input document is not supported."
[UnableToProcessContentError]", "Unable to process document contents."
[ContentTypeMismatchError]", "Declared content-type of the input document does not match the binary content type."
[PasswordProtectedPdfError]", "Password protected PDFs are not supported."
[MaximumNumberOfPagesPerDocumentLimitExceededError]", "Maximum number of pages for digitization exceeded."
[InvalidRequestData]", "The form data in the request is invalid. Expected is a multi-part form data, with either one part consisting of the document to be digitized, or two named parts: File - the document, DigitizationResult - the digitization result, with content type application/json."
[UnexpectedDigitizationResultStructure]", "The digitization result object is invalid and non-serializable."
[InvalidDom]", "The provided DOM is invalid. Make sure the DOM is correctly built, including valid non-overlapping indices, well formed boxes and polygons and valid values for all properties."
[MismatchingDomAndContent]", "The provided DOM and content do not match. Make sure the DOM was generated on the provided document."
[MismatchingDomAndText]", "The provided DOM and text do not match. Make sure the text and DOM were generated on the same document."
[PreprocessingOptionIncompatibleWithDigitizationResult]", "Using the preprocessing option while also providing a digitization result input is not supported."
[InvalidOcrApiKeyError]", "OCR Api key is invalid."
[OcrTooManyRequestsError]", "OCR request quota exceeded."
[ExternalOcrTooManyRequestsError]", "OCR request quota exceeded."
[GoogleBillingNotEnabled]", "Google OCR billing is not enabled. Please enable billing in your Google Cloud Platform account."
[GoogleApiKeyExpired]", "Google OCR Api Key Expired."
[InvalidOcrUrlError]", "The provided OCR URL is invalid or malformed."
[InvalidResponseFromOcrEngineError]", "Invalid response received from the OCR engine. Please set another OCR engine for the project you are using."
[DigitizationFileRequired]", "Required input file(s) are missing or not packaged as a valid multipart/form-data."

Digitization failed due to number of pages exceeded

If digitization failed for synchronous calls failed due to the document having more than five pages, an error message is generated. This is a 400 error, which is displayed as follows:

Code: [SyncMaximumNumberOfPagesExceeded], Message: Maximum Number Of Pages Exceeded.

If you encounter this error, use asysnchronous APIs. Asynchronous APIs are always recommended for production use cases. Synchronous APIs are only recommended in the following situations:

• When you are certain that you will never have more than five pages.
• If you only have single-page images, no PDFs or TIFFs.
• When preparing proof of concepts or demos.

Digitization failed due to server error

If digitization failed due to a server error, this can generate an error message. This is a 500 error, which is displayed as follows: Code: [DigitizationFailedServerError], Message: Internal Server Error.

If you encounter this error, we recommend contacting the UiPath® Support team.

Digitization job or DocumentId not found

If the digitization job or DocumentId is not found, an error message is generated:

• If you use either the Classification or Extraction synchronous APIs, a 404, Code:[DocumentIdNotFound] error is displayed.
• Is you use the asynchronous APIs for either Classification or Extraction, the get result will return a 200, Code:[DocumentIdNotFound] error.

In both situation, the following error message is generated:

Cannot perform the operation for the given documentId: Ensure it is correct, the digitization is successful (retrieving the digitization result), and not more than 7 days since the digitization call passed (case in which, it expired).

Resolution steps

1. Check the result of the digitization calling /digitization/result/{documentId}.
2. Retry the digitization process.
3. Once the new document ID has been generated from the retry, use it to classify and extract data.

Classification

Classification error codes

In some cases, a failure related to classification may result in an error message. This is an error, which is displayed as follows: Code: [ClassifierErrorCode], Message: "ClassifierErrorMessage". The error code and the corresponding message can have one of the following values:

ClassifierRequestTooLargeError - Maximum number of pages per document exceeded for the given classifier. Please use a custom trained model to process the input document.
ClassifierSyncMaximumNumberOfPagesExceeded - Maximum number of pages (5) for synchronous classification exceeded for this document. Please use the asynchronous APIs.
InvalidClassifierIdInput - The provided classifier id from the URL is different from the one in the classification result.
ClassifierIdNotFound - Classifier Id not found in the given project.
DeploymentClassifierNotFoundError - The requested project version deployment does not contain a classifier for the requested tag.
PrivateSkillClassifierUnavailable - The private skill classifier is not available at the moment. Check classifier status and try again.

Generative classifier error codes

In some cases, a failure related to generative classifier may result in an error message. This is an error, which is displayed as follows: Code: [GptClassifierErrorCode], Message: "GptClassifierErrorMessage". The error code and the corresponding message can have one of the following values:

GptClassifierPromptDescriptionLengthError - Prompts required for generative classification are invalid. Please provide description of maximum 2000 characters.
GptClassifierPromptNameLengthError - Prompts required for generative classification are invalid. Please provide names of minimum 1 characters and maximum 50 characters.
GptClassifierMissingPromptsError - Prompts are required for generative classification. Please provide a set of prompts composed of name and description.
GptClassifierTooManyPromptsError - The list of prompts provided is too large. Please provide a set of maximum 50 prompts composed of name and description.
GptClassifierDuplicatePromptNamesError - Prompts required for generative classification must be unique. Please provide a set of unique prompts composed of name and description.
GptClassifierUnavailableError - The generative classifier is not available at the moment. Check the status and try again.
GptClassifierForbiddenError - Generative classification capabilities are disabled.
GptClassifierContentFilteredError - The generative classifier identified filtered content. Check the document and try again.
GptClassifierInvalidInputError - The generative classifier requires valid prompts. Check the prompts and try again.
InvalidGptClassifierMissingPromptInput - Prompts are required for validation of generative classification. Please provide a set of prompts composed of name and description.
InvalidGptClassifierTooManyPromptInput - The list of prompts provided for validation of generative classification is too large. Please provide a set of maximum 50 prompts composed of name and description.
InvalidGptClassifierDuplicatePromptNameInput - Prompts required for validation of generative classification must be unique. Please provide a set of unique prompts composed of name and description.
InvalidGptClassifierPromptNameLengthError - Prompts required for validation of generative classification are invalid. Please provide names of minimum 1 characters and maximum 50 characters.

Extraction

Extraction error codes

In some cases, a failure related to extraction may result in an error message. This is an error, which is displayed as follows: Code: [ExtractorErrorCode], Message: "ExtractorErrorMessage". The error code and the corresponding message can have one of the following values:

ExtractorRequestTooLargeError - Maximum number of pages per document exceeded for the given extractor. Please use a custom trained model to process the input document.
ExtractorSyncMaximumNumberOfPagesExceeded - Maximum number of pages (5) for synchronous extraction exceeded for this document. Please use the asynchronous APIs.
ExtractionAutoValidationConfidenceInvalidError - The parameter value for autovalidation Confidence is invalid. Please provide a confidence value in the 0-100 range.
ExtractionAutoValidationConfidenceInvalidGenerativeError - The autovalidation extraction confidence option is invalid for the generative extractor. Please select a different extractor and try again.
ExtractorIdNotFound - Extractor Id not found in the given project.
PrivateSkillExtractorUnavailable - The private skill extractor is not available at the moment. Check extractor status and try again.
PrivateSkillExtractorGetInfoModelCallFailedError - The private skill extractor is not available at the moment. Check extractor status and try again.
PrivateMLSkillUnavailable - The private ML Skill is not available at the moment. Check the skill configuration. Possible configuration issues might include: insufficient replica count, insufficient memory or cpu, using cpu when a gpu is more appropriate.
PublicSkillUnavailable - The public skill is not available at the moment. Check the status and try again.

Generative extractor error codes

In some cases, a failure related to generative extractor may result in an error message. This is an error, which is displayed as follows: Code: [GptExtractorErrorCode], Message: "GptExtractorErrorMessage". The error code and the corresponding message can have one of the following values:

GptExtractorMissingPromptsError - Prompts are required for generative extraction. Please provide a set of prompts composed of id and question.
GptExtractorTooManyPromptsError - The list of prompts provided is too large. Please provide a set of maximum 150 prompts composed of id and question.
GptExtractorPromptQuestionLengthError - Prompts required for generative extraction are invalid. Please provide questions of minimum 3 characters and maximum 1000 characters.
GptExtractorMissingIdPromptError - Prompts required for generative extraction must have an id. Please provide a set of prompts composed of id and question.
GptExtractorDuplicatePromptIdsError - Prompts required for generative extraction must have a unique id. Please provide a set of unique prompts composed of id and question.
GptExtractorDuplicatePromptQuestionsError - Prompts required for generative extraction must have a unique question. Please provide a set of unique prompts composed of id and question.
GptExtractorUnavailableError - The generative extractor is not available at the moment. Try again later.
GptExtractorForbiddenError - Generative extraction capabilities are disabled.
GptExtractorContentFilteredError - The generative extractor identified filtered content. Check the document and try again.
GptExtractorInvalidInputError - The generative extractor requires valid prompts. Check the prompts and try again.
GptNotFoundError - LLMGateway not found in the organization or tenant. Check if generative features are disabled by AI Trust Layer or if they are available in your region.
GptGenAIDisabledError - Generative AI is disabled.
GptForbiddenError - User is forbidden by governance policy to use documentunderstanding.
GptExtractorDocumentTextLengthTooShortError - Document text should have minimum 10 characters. Check the document and try again.
InvalidGptExtractorMissingPromptsInput - Prompts are required for validation of generative extraction. Please provide a set of prompts composed of id and question.
InvalidGptExtractorTooManyPromptsInput - The list of prompts provided for validation of generative extraction is too large. Please provide a set of maximum 150  prompts composed of id and question.
InvalidGptExtractorMissingPromptIdInput - Prompts required for validation of generative extraction must have an id. Please provide a set of prompts composed of id and question.
InvalidGptExtractorDuplicatePromptIdsInput - Prompts required for validation of generative extraction must have a unique id. Please provide a set of unique prompts composed of id and question.
InvalidGptExtractorPromptsNotExtractedInput - Prompts required for validation of generative extraction must be the prompts used for generative extraction. Please provide a set of prompts composed of id and question that were used for generative extraction.
CannotSetAllowDocumentTypeChangeForGenerative - Cannot set allowChangeOfDocumentType defaulted to true when using the generative extractor, as there are no Document Types to be referenced.

Validation

Orchestrator error codes

In some cases, a failure related to Orchestrator may result in an error message. The error code and the corresponding message can have one of the following values:

OrchestratorDisabledError - The orchestrator service is not enabled for this tenant.
OrchestratorTaskCatalogNotFound - Orchestrator catalog does not exist.
OrchestratorFolderNotFoundError - Orchestrator folder does not exist.
OrchestratorBucketNotFoundError - Orchestrator bucket does not exist.
OrchestratorAssetNotFoundError - Orchestrator asset does not exist.
OrchestratorAlreadyExistsError - Orchestrator resource already exists.

Validation error codes

In some cases, a failure related to validation may result in an error message. This is an error, which is displayed as follows: Code: [ValidationErrorCode], Message: "ValidationErrorMessage". The error code and the corresponding message can have one of the following values:

ValidationTaskNotFoundError - Validation task not found in Action Center.
ValidationInvalidActionTitle - Required input ActionTitle is missing or invalid.
ValidationMaxLengthActionCatalog - Validation action catalog maximum length exceeded: 50 characters.
ValidationMaxLengthActionFolder - Validation action folder maximum length exceeded: 200 characters.
ValidationActionCatalogNotFoundError - Action Catalog not found.
InvalidOrMalformedClassificationResultInput - The provided classification result is invalid or malformed.
InvalidFieldsValidationConfidenceInput - The value set for Extracted Fields Validation Confidence is not valid. It should be an integer value between 0 and 100.
InvalidOrMalformedExtractionResultInput - The provided extraction result is invalid or malformed.
InvalidExtractorIdMismatchInput - The provided extractor id from the URL is different from the one in the extraction result.