Enable base64 stripping from sqs

[deepanshululla]

Title

Enables Strips base64 files from SQS and adds documentation for unexposed SQS configs

[92m01:01:54 - LiteLLM:ERROR�[0m: sqs.py:339 - Error sending to SQS: Client error '413 Request Entity Too Large' for url 'https://sqs.us-west-2.amazonaws.com/....'

Relevant issues

Large files can go in base64

Pre-Submission checklist

Please complete all items before asking a LiteLLM maintainer to review your PR

• I have Added testing in the tests/litellm/ directory, Adding at least 1 test is a hard requirement - see details
• I have added a screenshot of my new test passing locally
• My PR passes all unit tests on make test-unit
• My PR's scope is as isolated as possible, it only solves 1 specific problem

((litellm-py3.12) (base) ➜  litellm git:(feature/strip_base64_sqs2) poetry run pytest tests/logging_callback_tests/test_sqs_logger.py --tb=short -vv --maxfail=10 -n 4 
=========================================================================================== test session starts ===========================================================================================
platform darwin -- Python 3.12.2, pytest-7.4.4, pluggy-1.5.0 -- /Users/deepanshu.deepanshu/Library/Caches/pypoetry/virtualenvs/litellm-ohMn1BPm-py3.12/bin/python
cachedir: .pytest_cache
rootdir: /Users/deepanshu.deepanshu/code/code/litellm/deepanshu/litellm
configfile: pyproject.toml
plugins: respx-0.22.0, xdist-3.8.0, anyio-4.5.2, mock-3.14.1, asyncio-0.21.2, retry-1.6.3, requests-mock-1.12.1
asyncio: mode=Mode.AUTO
4 workers [15 items]    
scheduling tests via LoadScheduling

tests/logging_callback_tests/test_sqs_logger.py::test_async_sqs_logger_flush 
tests/logging_callback_tests/test_sqs_logger.py::test_appcrypto_encrypt_decrypt_roundtrip 
tests/logging_callback_tests/test_sqs_logger.py::test_async_log_failure_event_adds_to_queue 
tests/logging_callback_tests/test_sqs_logger.py::test_async_send_batch_triggers_tasks 
[gw1] [  6%] PASSED tests/logging_callback_tests/test_sqs_logger.py::test_async_log_failure_event_adds_to_queue 
tests/logging_callback_tests/test_sqs_logger.py::test_async_log_success_event_adds_to_queue 
[gw0] [ 13%] PASSED tests/logging_callback_tests/test_sqs_logger.py::test_appcrypto_encrypt_decrypt_roundtrip 
tests/logging_callback_tests/test_sqs_logger.py::test_appcrypto_invalid_key_length 
[gw2] [ 20%] PASSED tests/logging_callback_tests/test_sqs_logger.py::test_async_send_batch_triggers_tasks 
tests/logging_callback_tests/test_sqs_logger.py::test_async_sqs_logger_error_flush 
[gw0] [ 26%] PASSED tests/logging_callback_tests/test_sqs_logger.py::test_appcrypto_invalid_key_length 
tests/logging_callback_tests/test_sqs_logger.py::test_sqs_logger_init_without_encryption 
[gw1] [ 33%] PASSED tests/logging_callback_tests/test_sqs_logger.py::test_async_log_success_event_adds_to_queue 
tests/logging_callback_tests/test_sqs_logger.py::test_sqs_logger_init_with_encryption_missing_key 
[gw0] [ 40%] PASSED tests/logging_callback_tests/test_sqs_logger.py::test_sqs_logger_init_without_encryption 
[gw1] [ 46%] PASSED tests/logging_callback_tests/test_sqs_logger.py::test_sqs_logger_init_with_encryption_missing_key 
tests/logging_callback_tests/test_sqs_logger.py::test_strip_base64_keeps_non_file_content 
tests/logging_callback_tests/test_sqs_logger.py::test_strip_base64_mixed_nested_objects 
[gw0] [ 53%] PASSED tests/logging_callback_tests/test_sqs_logger.py::test_strip_base64_keeps_non_file_content 
[gw1] [ 60%] PASSED tests/logging_callback_tests/test_sqs_logger.py::test_strip_base64_mixed_nested_objects 
tests/logging_callback_tests/test_sqs_logger.py::test_strip_base64_removes_file_and_nontext_entries 
tests/logging_callback_tests/test_sqs_logger.py::test_strip_base64_recursive_redaction 
[gw0] [ 66%] PASSED tests/logging_callback_tests/test_sqs_logger.py::test_strip_base64_recursive_redaction 
[gw1] [ 73%] PASSED tests/logging_callback_tests/test_sqs_logger.py::test_strip_base64_removes_file_and_nontext_entries 
[gw3] [ 80%] PASSED tests/logging_callback_tests/test_sqs_logger.py::test_async_sqs_logger_flush 
tests/logging_callback_tests/test_sqs_logger.py::test_sqs_logger_init_with_encryption 
[gw2] [ 86%] PASSED tests/logging_callback_tests/test_sqs_logger.py::test_async_sqs_logger_error_flush 
tests/logging_callback_tests/test_sqs_logger.py::test_strip_base64_handles_empty_or_missing_messages 
[gw3] [ 93%] PASSED tests/logging_callback_tests/test_sqs_logger.py::test_sqs_logger_init_with_encryption 
[gw2] [100%] PASSED tests/logging_callback_tests/test_sqs_logger.py::test_strip_base64_handles_empty_or_missing_messages Task was destroyed but it is pending!
task: <Task pending name='Task-9' coro=<CustomBatchLogger.periodic_flush() running at /Users/deepanshu.deepanshu/code/code/litellm/deepanshu/litellm/litellm/integrations/custom_batch_logger.py:38> wait_for=<Future pending cb=[Task.task_wakeup()]>>
Task was destroyed but it is pending!
task: <Task pending name='Task-18' coro=<CustomBatchLogger.periodic_flush() done, defined at /Users/deepanshu.deepanshu/code/code/litellm/deepanshu/litellm/litellm/integrations/custom_batch_logger.py:36> wait_for=<Future pending cb=[Task.task_wakeup()]>>
Task was destroyed but it is pending!
task: <Task pending name='Task-21' coro=<CustomBatchLogger.periodic_flush() done, defined at /Users/deepanshu.deepanshu/code/code/litellm/deepanshu/litellm/litellm/integrations/custom_batch_logger.py:36> wait_for=<Future pending cb=[Task.task_wakeup()]>>
Task was destroyed but it is pending!
task: <Task pending name='Task-16' coro=<CustomBatchLogger.periodic_flush() running at /Users/deepanshu.deepanshu/code/code/litellm/deepanshu/litellm/litellm/integrations/custom_batch_logger.py:38> wait_for=<Future pending cb=[Task.task_wakeup()]>>
Task was destroyed but it is pending!
task: <Task pending name='Task-19' coro=<CustomBatchLogger.periodic_flush() done, defined at /Users/deepanshu.deepanshu/code/code/litellm/deepanshu/litellm/litellm/integrations/custom_batch_logger.py:36> wait_for=<Future pending cb=[Task.task_wakeup()]>>
Task was destroyed but it is pending!
task: <Task pending name='Task-14' coro=<CustomBatchLogger.periodic_flush() running at /Users/deepanshu.deepanshu/code/code/litellm/deepanshu/litellm/litellm/integrations/custom_batch_logger.py:38> wait_for=<Future pending cb=[Task.task_wakeup()]>>
Task was destroyed but it is pending!
task: <Task pending name='Task-5' coro=<CustomBatchLogger.periodic_flush() running at /Users/deepanshu.deepanshu/code/code/litellm/deepanshu/litellm/litellm/integrations/custom_batch_logger.py:38> wait_for=<Future pending cb=[Task.task_wakeup()]>>
sys:1: RuntimeWarning: coroutine 'CustomBatchLogger.periodic_flush' was never awaited
RuntimeWarning: Enable tracemalloc to get the object allocation traceback


============================================================================================ warnings summary =============================================================================================
litellm/litellm_core_utils/get_model_cost_map.py:41: 5 warnings
tests/logging_callback_tests/test_sqs_logger.py: 15 warnings
  /Users/deepanshu.deepanshu/code/code/litellm/deepanshu/litellm/litellm/litellm_core_utils/get_model_cost_map.py:41: DeprecationWarning: open_text is deprecated. Use files() instead. Refer to https://importlib-resources.readthedocs.io/en/latest/using.html#migrating-from-legacy for migration advice.
    with importlib.resources.open_text(

tests/logging_callback_tests/test_sqs_logger.py::test_async_log_success_event_adds_to_queue
  /opt/anaconda3/lib/python3.12/json/decoder.py:353: RuntimeWarning: coroutine 'CustomBatchLogger.periodic_flush' was never awaited
    obj, end = self.scan_once(s, idx)
  Enable tracemalloc to get traceback where the object was allocated.
  See https://docs.pytest.org/en/stable/how-to/capture-warnings.html#resource-warnings for more info.

tests/logging_callback_tests/test_sqs_logger.py::test_sqs_logger_init_with_encryption_missing_key
  /opt/anaconda3/lib/python3.12/importlib/__init__.py:131: RuntimeWarning: coroutine 'CustomBatchLogger.periodic_flush' was never awaited
    _bootstrap._exec(spec, module)
  Enable tracemalloc to get traceback where the object was allocated.
  See https://docs.pytest.org/en/stable/how-to/capture-warnings.html#resource-warnings for more info.

tests/logging_callback_tests/test_sqs_logger.py::test_strip_base64_keeps_non_file_content
  /Users/deepanshu.deepanshu/Library/Caches/pypoetry/virtualenvs/litellm-ohMn1BPm-py3.12/lib/python3.12/site-packages/httpx/_client.py:1424: RuntimeWarning: coroutine 'CustomBatchLogger.periodic_flush' was never awaited
    for key, proxy in proxy_map.items()
  Enable tracemalloc to get traceback where the object was allocated.
  See https://docs.pytest.org/en/stable/how-to/capture-warnings.html#resource-warnings for more info.

tests/logging_callback_tests/test_sqs_logger.py::test_async_sqs_logger_flush
  /Users/deepanshu.deepanshu/Library/Caches/pypoetry/virtualenvs/litellm-ohMn1BPm-py3.12/lib/python3.12/site-packages/pydantic/main.py:426: UserWarning: Pydantic serializer warnings:
    PydanticSerializationUnexpectedValue: Expected 10 fields but got 5 for type `Message` with value `Message(content='hi', role='assistant', tool_calls...unction_call=None, provider_specific_fields=None)` - serialized value may not be as expected.
    PydanticSerializationUnexpectedValue: Expected `StreamingChoices` but got `Choices` with value `Choices(finish_reason='st...r_specific_fields=None))` - serialized value may not be as expected
    return self.__pydantic_serializer__.to_python(

tests/logging_callback_tests/test_sqs_logger.py::test_async_sqs_logger_flush
tests/logging_callback_tests/test_sqs_logger.py::test_async_sqs_logger_error_flush
  /Users/deepanshu.deepanshu/Library/Caches/pypoetry/virtualenvs/litellm-ohMn1BPm-py3.12/lib/python3.12/site-packages/botocore/auth.py:425: DeprecationWarning: datetime.datetime.utcnow() is deprecated and scheduled for removal in a future version. Use timezone-aware objects to represent datetimes in UTC: datetime.datetime.now(datetime.UTC).
    datetime_now = datetime.datetime.utcnow()

tests/logging_callback_tests/test_sqs_logger.py::test_async_sqs_logger_error_flush
  /Users/deepanshu.deepanshu/Library/Caches/pypoetry/virtualenvs/litellm-ohMn1BPm-py3.12/lib/python3.12/site-packages/httpx/_urlparse.py:186: RuntimeWarning: coroutine 'CustomBatchLogger.periodic_flush' was never awaited
    def copy_with(self, **kwargs: str | None) -> ParseResult:
  Enable tracemalloc to get traceback where the object was allocated.
  See https://docs.pytest.org/en/stable/how-to/capture-warnings.html#resource-warnings for more info.

tests/logging_callback_tests/test_sqs_logger.py::test_async_sqs_logger_error_flush
  /Users/deepanshu.deepanshu/Library/Caches/pypoetry/virtualenvs/litellm-ohMn1BPm-py3.12/lib/python3.12/site-packages/httpx/_urlparse.py:186: RuntimeWarning: coroutine 'AsyncMockMixin._execute_mock_call' was never awaited
    def copy_with(self, **kwargs: str | None) -> ParseResult:
  Enable tracemalloc to get traceback where the object was allocated.
  See https://docs.pytest.org/en/stable/how-to/capture-warnings.html#resource-warnings for more info.

tests/logging_callback_tests/test_sqs_logger.py::test_async_sqs_logger_error_flush
  /Users/deepanshu.deepanshu/Library/Caches/pypoetry/virtualenvs/litellm-ohMn1BPm-py3.12/lib/python3.12/site-packages/pydantic/main.py:426: UserWarning: Pydantic serializer warnings:
    PydanticSerializationUnexpectedValue: Expected 10 fields but got 5 for type `Message` with value `Message(content='Error occurred', role='assistant'...unction_call=None, provider_specific_fields=None)` - serialized value may not be as expected.
    PydanticSerializationUnexpectedValue: Expected `StreamingChoices` but got `Choices` with value `Choices(finish_reason='st...r_specific_fields=None))` - serialized value may not be as expected
    return self.__pydantic_serializer__.to_python(

-- Docs: https://docs.pytest.org/en/stable/how-to/capture-warnings.html
===================================================================================== 15 passed, 29 warnings in 4.39s =====================================================================================
(litellm-py3.12) (base) ➜  litellm git:(feature/strip_base64_sqs2) 

Type

Bug fix

Changes

Adds an option to strip

[vercel]

Someone is attempting to deploy a commit to the CLERKIEAI Team on Vercel.

A member of the Team first needs to authorize it.

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.

---

deepanshu seems not to be a GitHub user. You need a GitHub account to be able to sign the CLA. If you have already a GitHub account, please add the email address used for this commit to your account.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

[krrishdholakia]

hi @deepanshululla can you please fix the linting errors?

[deepanshululla]

@krrishdholakia fixed. thanks

[ishaan-jaff]

reviewed

[ishaan-jaff]

there's a lot more content here, can u just add the new field

[ishaan-jaff]

please add a max depth, we don't want infinite recursion