I get this error- Object of type FileStorage is not JSON serializable while enabling logging and user upload

[libis22]

Hi all, when I enabled file upload with user login features, I got this error: Object of type FileStorage is not JSON serializable

here is the call stack: Traceback (most recent call last): File "/tmp/8de1fc16293c5fa/antenv/lib/python3.11/site-packages/azure/ai/documentintelligence/_model_base.py", line 146, in default return super(SdkJSONEncoder, self).default(o) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/opt/python/3.11.14/lib/python3.11/json/encoder.py", line 180, in default raise TypeError(f'Object of type {o.class.name} ' TypeError: Object of type FileStorage is not JSON serializable During handling of the above exception, another exception occurred: Traceback (most recent call last): File "/tmp/8de1fc16293c5fa/antenv/lib/python3.11/site-packages/opentelemetry/trace/init.py", line 589, in use_span yield span File "/tmp/8de1fc16293c5fa/antenv/lib/python3.11/site-packages/azure/core/tracing/decorator_async.py", line 138, in wrapper_use_tracer return await func(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/8de1fc16293c5fa/antenv/lib/python3.11/site-packages/azure/ai/documentintelligence/aio/_operations/_patch.py", line 529, in begin_analyze_document raw_result = await self._analyze_document_initial( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/8de1fc16293c5fa/antenv/lib/python3.11/site-packages/azure/ai/documentintelligence/aio/_operations/_operations.py", line 120, in _analyze_document_initial _content = json.dumps(analyze_request, cls=SdkJSONEncoder, exclude_readonly=True) # type: ignore ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/opt/python/3.11.14/lib/python3.11/json/init.py", line 238, in dumps **kw).encode(obj) ^^^^^^^^^^^ File "/opt/python/3.11.14/lib/python3.11/json/encoder.py", line 200, in encode chunks = self.iterencode(o, _one_shot=True) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/opt/python/3.11.14/lib/python3.11/json/encoder.py", line 258, in iterencode return _iterencode(o, 0) ^^^^^^^^^^^^^^^^^ File "/tmp/8de1fc16293c5fa/antenv/lib/python3.11/site-packages/azure/ai/documentintelligence/_model_base.py", line 165, in default return super(SdkJSONEncoder, self).default(o) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/opt/python/3.11.14/lib/python3.11/json/encoder.py", line 180, in default raise TypeError(f'Object of type {o.class.name} ' TypeError: Object of type FileStorage is not JSON serializable

Fix was this change at /workspaces/azure-search-openai-demo/app/backend/prepdocslib/pdfparser.py

if media_describer is not None:
content_bytes = content.read()
try:
#change for Object of type FileStorage is not JSON serializable Error
# poller = await document_intelligence_client.begin_analyze_document(
# model_id="prebuilt-layout",
# analyze_request=AnalyzeDocumentRequest(bytes_source=content_bytes),
# output=["figures"],
# features=["ocrHighResolution"],
# output_content_format="markdown",
# )
poller = await document_intelligence_client.begin_analyze_document(
model_id="prebuilt-layout",
body=content_bytes,
content_type="application/octet-stream",
output=["figures"],
features=["ocrHighResolution"],
output_content_format="markdown",
)
doc_for_pymupdf = pymupdf.open(stream=io.BytesIO(content_bytes))
file_analyzed = True
except HttpResponseError as e:
content.seek(0)
if e.error and e.error.code == "InvalidArgument":
logger.error(
"This document type does not support media description. Proceeding with standard analysis."
)
else:
logger.error(
"Unexpected error analyzing document for media description: %s. Proceeding with standard analysis.",
e,
)
#change for Object of type FileStorage is not JSON serializable Error
# if file_analyzed is False:
# poller = await document_intelligence_client.begin_analyze_document(
# model_id=self.model_id, analyze_request=content, content_type="application/octet-stream"
# )
if file_analyzed is False:
# ensure stream at start if it was read earlier
try:
content.seek(0)
except Exception:
pass
poller = await document_intelligence_client.begin_analyze_document(
model_id=self.model_id,
body=content, # ✅ stream/bytes go in body
content_type="application/octet-stream",
)
analyze_result: AnalyzeResult = await poller.result()

        offset = 0

this worked for me! but not sure if this is the best solution...

[Bennzn]

I was just experiencing the same problem! (Users loved this feature, but within the last versions it seems to be broken)
Many thanks for sharing your solution.
I'm running the bot in Azure and using Container apps.
After deploying your workaround, unfortunately i receive an "ERROR - Error uploading file: ClientSession._request() got an unexpected keyword argument 'body'"

Any Idea how i can dig deeper to find more out of it?

Many Thanks

[pamelafox]

I'm working on replicating this issue now.

[pamelafox]

Replicated! Working on a fix.

[pamelafox]

See PR for a fix:
#2830

I also updated SDK in that fix, but if you dont want to upgrade SDK, you should be able to apply the other change (turning the content into bytes before passing it to the SDK).

I'm still not sure why this regression popped up this week, as I don't think these particular components changed very recently. I'm glad you filed the issue though, thanks!

[libis22]

Thank you so much for working on this!! 💯