AnalyzeResult Type Error in `AzureOCRDocumentConverter

[vblagoje]

Description:

The static method _convert_azure_result_to_document in the AzureOCRDocumentConverter class has a type hint for AnalyzeResult in its method signature. The issue is that AnalyzeResult comes from the azure.ai.formrecognizer package, which is an optional dependency for our project. This leads to a NameError when the package isn't installed.

Stack Trace:

NameError                                 Traceback (most recent call last)
<ipython-input-4-808600d3e9ab> in <cell line: 6>()
      4 from haystack.preview.components.builders.prompt_builder import PromptBuilder
      5 from haystack.preview.components.fetchers import LinkContentFetcher
----> 6 from haystack.preview.components.file_converters import HTMLToDocument
      7 from haystack.preview.components.generators.openai.gpt import GPTGenerator, default_streaming_callback
      8 from haystack.preview.components.routers.file_type_router import FileTypeRouter

2 frames
/usr/local/lib/python3.10/dist-packages/haystack/preview/components/file_converters/azure.py in AzureOCRDocumentConverter()
     93 
     94     @staticmethod
---> 95     def _convert_azure_result_to_document(result: AnalyzeResult, id_hash_keys: List[str], file_suffix: str) -> Document:
     96         """
     97         Convert the result of Azure OCR to a Haystack text Document.

NameError: name 'AnalyzeResult' is not defined

[vblagoje]

Fixed with #6030