[form recognizer] actionable issues for arch board (#11334)

* updated docstring for stream parameter to be 'bytes or IO[bytes'

* directly pass stream to sync samples

* got rid of unnecessary imports

* looping through words and their confidences

Author: iscai-msft

sdk/formrecognizer/azure-ai-formrecognizer/azure/ai/formrecognizer/_form_recognizer_client.py

@@ -78,7 +78,7 @@ def begin_recognize_receipts(self, stream, **kwargs):
 
         :param stream: .pdf, .jpg, .png or .tiff type file stream.
              Currently only supports US sales receipts.
-        :type stream: stream
+        :type stream: bytes or IO[bytes]
         :keyword bool include_text_content:
             Whether or not to include text elements such as lines and words in addition to form fields.
         :keyword str content_type: Media type of the body sent to the API. Content-type is
@@ -172,7 +172,7 @@ def begin_recognize_content(self, stream, **kwargs):
         'image/jpeg', 'image/png' or 'image/tiff'.
 
         :param stream: .pdf, .jpg, .png or .tiff type file stream.
-        :type stream: stream
+        :type stream: bytes or IO[bytes]
         :keyword str content_type: Media type of the body sent to the API. Content-type is
             auto-detected, but can be overridden by passing this keyword argument. For options,
             see :class:`~azure.ai.formrecognizer.FormContentType`.
@@ -246,7 +246,7 @@ def begin_recognize_custom_forms(self, model_id, stream, **kwargs):
 
         :param str model_id: Custom model identifier.
         :param stream: .pdf, .jpg, .png or .tiff type file stream.
-        :type stream: stream
+        :type stream: bytes or IO[bytes]
         :keyword bool include_text_content:
             Whether or not to include text elements such as lines and words in addition to form fields.
         :keyword str content_type: Media type of the body sent to the API. Content-type is

sdk/formrecognizer/azure-ai-formrecognizer/azure/ai/formrecognizer/aio/_form_recognizer_client_async.py

@@ -90,7 +90,7 @@ async def recognize_receipts(
 
         :param stream: .pdf, .jpg, .png or .tiff type file stream.
             Currently only supports US sales receipts.
-        :type stream: stream
+        :type stream: bytes or IO[bytes]
         :keyword bool include_text_content:
             Whether or not to include text elements such as lines and words in addition to form fields.
         :keyword str content_type: Media type of the body sent to the API. Content-type is
@@ -184,7 +184,7 @@ async def recognize_content(self, stream: Union[bytes, IO[bytes]], **kwargs: Any
         'image/jpeg', 'image/png' or 'image/tiff'.
 
         :param stream: .pdf, .jpg, .png or .tiff type file stream.
-        :type stream: stream
+        :type stream: bytes or IO[bytes]
         :keyword str content_type: Media type of the body sent to the API. Content-type is
             auto-detected, but can be overridden by passing this keyword argument. For options,
             see :class:`~azure.ai.formrecognizer.FormContentType`.
@@ -258,7 +258,7 @@ async def recognize_custom_forms(
 
         :param str model_id: Custom model identifier.
         :param stream: .pdf, .jpg, .png or .tiff type file stream.
-        :type stream: stream
+        :type stream: bytes or IO[bytes]
         :keyword bool include_text_content:
             Whether or not to include text elements such as lines and words in addition to form fields.
         :keyword str content_type: Media type of the body sent to the API. Content-type is

sdk/formrecognizer/azure-ai-formrecognizer/samples/async_samples/sample_recognize_content_async.py

@@ -39,7 +39,6 @@ class RecognizeContentSampleAsync(object):
     async def recognize_content(self):
         # the sample forms are located in this file's parent's parent's files.
         path_to_sample_forms = Path(__file__).parent.parent.absolute() / Path("sample_forms/forms/Invoice_1.pdf")
-        from azure.ai.formrecognizer import FormWord, FormLine
         # [START recognize_content_async]
         from azure.core.credentials import AzureKeyCredential
         from azure.ai.formrecognizer.aio import FormRecognizerClient
@@ -74,6 +73,8 @@ async def recognize_content(self):
                         line.text,
                         format_bounding_box(line.bounding_box)
                     ))
+                    for word in line.words:
+                        print("...Word '{}' has a confidence of {}".format(word.text, word.confidence))
                 print("----------------------------------------")
 
 

sdk/formrecognizer/azure-ai-formrecognizer/samples/sample_get_bounding_boxes.py

@@ -48,7 +48,7 @@ def get_bounding_boxes(self):
         # Make sure your form's type is included in the list of form types the custom model can recognize
         with open("sample_forms/forms/Form_1.jpg", "rb") as f:
             poller = form_recognizer_client.begin_recognize_custom_forms(
-                model_id=self.model_id, stream=f.read(), include_text_content=True
+                model_id=self.model_id, stream=f, include_text_content=True
             )
         forms = poller.result()
 

sdk/formrecognizer/azure-ai-formrecognizer/samples/sample_recognize_content.py

@@ -33,13 +33,12 @@ class RecognizeContentSample(object):
     key = os.environ["AZURE_FORM_RECOGNIZER_KEY"]
 
     def recognize_content(self):
-        from azure.ai.formrecognizer import FormWord, FormLine
         # [START recognize_content]
         from azure.core.credentials import AzureKeyCredential
         from azure.ai.formrecognizer import FormRecognizerClient
         form_recognizer_client = FormRecognizerClient(endpoint=self.endpoint, credential=AzureKeyCredential(self.key))
         with open("sample_forms/forms/Invoice_1.pdf", "rb") as f:
-            poller = form_recognizer_client.begin_recognize_content(stream=f.read())
+            poller = form_recognizer_client.begin_recognize_content(stream=f)
         contents = poller.result()
 
         for idx, content in enumerate(contents):
@@ -66,6 +65,8 @@ def recognize_content(self):
                     line.text,
                     format_bounding_box(line.bounding_box)
                 ))
+                for word in line.words:
+                    print("...Word '{}' has a confidence of {}".format(word.text, word.confidence))
             print("----------------------------------------")
 
 

sdk/formrecognizer/azure-ai-formrecognizer/samples/sample_recognize_custom_forms.py

@@ -43,7 +43,7 @@ def recognize_custom_forms(self):
         # Make sure your form's type is included in the list of form types the custom model can recognize
         with open("sample_forms/forms/Form_1.jpg", "rb") as f:
             poller = form_recognizer_client.begin_recognize_custom_forms(
-                model_id=self.model_id, stream=f.read()
+                model_id=self.model_id, stream=f
             )
         forms = poller.result()
 

sdk/formrecognizer/azure-ai-formrecognizer/samples/sample_recognize_receipts.py

@@ -36,7 +36,7 @@ def recognize_receipts(self):
             endpoint=self.endpoint, credential=AzureKeyCredential(self.key)
         )
         with open("sample_forms/receipt/contoso-allinone.jpg", "rb") as f:
-            poller = form_recognizer_client.begin_recognize_receipts(stream=f.read())
+            poller = form_recognizer_client.begin_recognize_receipts(stream=f)
         receipts = poller.result()
 
         for idx, receipt in enumerate(receipts):