NewYour coding agent can read the release notes before it upgrades.Set up the MCP server →
PyPI · #1109 most downloaded on PyPI
Microsoft Azure AI Document Intelligence Client Library for Python
Last release 2 years ago
no release in 18 months
Ships fairly regularly
a new release about every 3 months
Nearly every release is documented
notes for 3 of 3 stable releases
Nothing withdrawn
no release was ever pulled
3 years old
7 releases · first in 2023
Fix polling for begin_copy_classifier_to() to stop on success response from the "Operation-Location" endpoint and correctly parse the result.
begin_copy_classifier_to() to stop on success response from the "Operation-Location" endpoint and correctly parse the result.Fix polling for begin_copy_model_to() to stop on success response from the "Operation-Location" endpoint and correctly parse the result.
begin_copy_model_to() to stop on success response from the "Operation-Location" endpoint and correctly parse the result.One column per month.
Added support for the Analyze Batch Documents API:
delete_analyze_batch_result(), get_analyze_batch_result() and list_analyze_batch_results() to DocumentIntelligenceClient.delete_analyze_result() to DocumentIntelligenceClient.body.get_resource_info() to get_resource_details().ContentFormat to DocumentContentFormat.AnalyzeBatchResultOperation to AnalyzeBatchOperation.CopyAuthorization to ModelCopyAuthorization.Document to AnalyzedDocument.Error to DocumentIntelligenceError.ErrorResponse to DocumentIntelligenceErrorResponse.InnerError to DocumentIntelligenceInnerError.OperationDetails to DocumentIntelligenceOperationDetails.OperationStatus to DocumentIntelligenceOperationStatus.ResourceDetails to DocumentIntelligenceResourceDetails.Warning to DocumentIntelligenceWarning.items_property in model DocumentFieldSchema to items_schema.FontStyle to DocumentFontStyle.FontWeight to DocumentFontWeight.AnalyzeResultOperation.GENERATIVE in enum DocumentBuildMode.2024-11-30.content-type when analyze_request is a stream in begin_analyze_document() and begin_classify_document().Added support for the Analyze Batch Documents API:
begin_analyze_batch_documents() to DocumentIntelligenceClient.AnalyzeBatchDocumentsRequest, AnalyzeBatchResult and AnalyzeBatchOperationDetail.get_analyze_result_figure() and get_analyze_result_pdf() to DocumentIntelligenceClient.output to LRO operation begin_analyze_document() overloads in DocumentIntelligenceClient.AnalyzeOutputOption to specify output kind, either pdf or figures.id to model DocumentFigure.authorize_classifier_copy() and begin_copy_classifier_to() to DocumentIntelligenceAdministrationClient.AuthorizeClassifierCopyRequest and ClassifierCopyAuthorization.pages to LRO operation begin_classify_document() overloads in DocumentIntelligenceClient.GENERATIVE to enum DocumentBuildMode.warnings to model AnalyzeResult.classifier_id, split, and training_hours to model DocumentModelDetails.model_id, confidence_threshold, features, query_fields and max_documents_to_analyze to model DocumentTypeDetails.allow_overwrite to model BuildDocumentClassifierRequest.allow_overwrite and max_training_hours to model BuildDocumentModelRequest.classifier_id, split and doc_types to model ComposeDocumentModelRequest.operation_id via details property in the new return types AnalyzeDocumentLROPoller and AsyncAnalyzeDocumentLROPoller in operation begin_analyze_document().DocumentList and DocumentListItem.lists from model AnalyzeResult.ComponentDocumentModelDetails.component_models from model ComposeDocumentModelRequest.ComposeDocumentModelRequest now requires a dictionary of DocumentTypeDetails instances and a classifier ID to be constructed.QuotaDetails.custom_neural_document_model_builds from model ResourceDetails.field_schema from DocumentTypeDetails to be optional.2024-07-31-preview.1.5X faster when deserializing JSON to an AnalyzeResult object compared to last version 1.0.0b3.Changed the default polling interval from 5s to 1s.
Added models AnalyzeResultOperation and Warning.
AnalyzeResultOperation and Warning.base_classifier_id to model BuildDocumentClassifierRequest.base_classifier_id and warnings to model DocumentClassifierDetails.warnings to model DocumentModelDetails.value_selection_group to model DocumentField.selectionGroup to enum DocumentFieldType.completed to enum OperationStatus.items to items_property in model DocumentFieldSchema and DocumentList.base64_source to bytes_source in model AnalyzeDocumentRequest and ClassifyDocumentRequest.2024-02-29-preview.azure-core to >=1.30.0.typing-extensions to >=4.6.0.This is the first preview of the azure-ai-documentintelligence package, targeting API version 2023-10-31-preview of the Document Intelligence service(
This is the first preview of the azure-ai-documentintelligence package, targeting API version 2023-10-31-preview of the Document Intelligence service(formerly known as Form Recognizer).
Note: Form Recognizer has been rebranded to Document Intelligence.
azure-ai-formrecognizer to DocumentIntelligenceClient and DocumentIntelligenceAdministrationClient in API version 2023-10-31-preview in azure-ai-documentintelligence.{endpoint}/formrecognizer to {endpoint}/documentintelligence.prebuilt-receipt model.prebuilt-businessCard and prebuilt-document. prebuilt-document model is essentially prebuilt-layout with features="keyValuePairs" specified. (This is only supported as an optional feature for "prebuilt-layout" and "prebuilt-invoice".)If you were using the old azure-ai-formrecognizer package, please refer MIGRATION_GUIDE.MD for more details.
Your coding agent can read these notes before it upgrades. Set up the MCP server →