Azure/azure-sdk-for-python azure-ai-documentintelligence_1.0.0b4

on GitHub

1.0.0b4 (2024-09-05)

Features Added

• Added support for the Analyze Batch Documents API:
  • Added LRO operation begin_analyze_batch_documents() to DocumentIntelligenceClient.
  • Added models AnalyzeBatchDocumentsRequest, AnalyzeBatchResult and AnalyzeBatchOperationDetail.
• Added support for different kinds of output in the Analyze Document API:
  • Added operations get_analyze_result_figure() and get_analyze_result_pdf() to DocumentIntelligenceClient.
  • Added optional kwarg output to LRO operation begin_analyze_document() overloads in DocumentIntelligenceClient.
  • Added enum AnalyzeOutputOption to specify output kind, either pdf or figures.
  • Added property id to model DocumentFigure.
• Added support for the Copy Classifier API:
  • Added operations authorize_classifier_copy() and begin_copy_classifier_to() to DocumentIntelligenceAdministrationClient.
  • Added models AuthorizeClassifierCopyRequest and ClassifierCopyAuthorization.
• Added optional kwarg pages to LRO operation begin_classify_document() overloads in DocumentIntelligenceClient.
• Added new kind GENERATIVE to enum DocumentBuildMode.
• Added property warnings to model AnalyzeResult.
• Added properties classifier_id, split, and training_hours to model DocumentModelDetails.
• Added properties model_id, confidence_threshold, features, query_fields and max_documents_to_analyze to model DocumentTypeDetails.
• Added property allow_overwrite to model BuildDocumentClassifierRequest.
• Added properties allow_overwrite and max_training_hours to model BuildDocumentModelRequest.
• Added properties classifier_id, split and doc_types to model ComposeDocumentModelRequest.
• Added support for getting operation_id via details property in the new return types AnalyzeDocumentLROPoller and AsyncAnalyzeDocumentLROPoller in operation begin_analyze_document().

Breaking Changes

• Removed support for extracting lists from analyzed documents:
  • Removed models DocumentList and DocumentListItem.
  • Removed property lists from model AnalyzeResult.
• Changes to the Compose Document API:
  • Removed model ComponentDocumentModelDetails.
  • Removed property component_models from model ComposeDocumentModelRequest.
  • ComposeDocumentModelRequest now requires a dictionary of DocumentTypeDetails instances and a classifier ID to be constructed.
• Removed model QuotaDetails.
• Removed property custom_neural_document_model_builds from model ResourceDetails.
• Changed the required property field_schema from DocumentTypeDetails to be optional.

Other Changes

• Changed the default service API version to 2024-07-31-preview.
• Improved performance by about 1.5X faster when deserializing JSON to an AnalyzeResult object compared to last version 1.0.0b3.