Fix optional inputs with None/empty string values in job submissions

[Copilot]

Optional string inputs with None or empty string values fail with "The Value field is required" error when submitting AzureML jobs, despite being marked as optional=True in component definitions.

Root Cause

to_rest_dataset_literal_inputs() was creating LiteralJobInput objects for None and empty string values, violating REST API constraints (value required, min_length=1).

Changes

azure/ai/ml/entities/_job/_input_output_helpers.py

• Skip adding inputs with None values to REST request
• Skip adding inputs with empty string values to REST request
• Handle dict inputs: skip {"value": None} and {"value": ""}
• Applies to: CommandJob, PipelineJob, SparkJob, SweepJob, ImportJob

tests/command_job/unittests/test_command_job_entity.py

• Add test covering None, empty string, and dict input scenarios

CHANGELOG.md

• Document bug fix in v1.32.0

Example

# Previously failed, now works - backend uses default value
job = ml_client.jobs.create_or_update(
    command_component(
        run_date="2023-01-31",
        tags=""  # or tags=None, or omit entirely
    )
)

Backward compatible - only affects previously failing cases.

Warning

Firewall rules blocked me from connecting to one or more addresses (expand for details)

I tried to connect to the following addresses, but was blocked by firewall rules:

• pypi.org
  • Triggering command: /home/REDACTED/work/azure-sdk-for-python/azure-sdk-for-python/.venv/bin/python /home/REDACTED/work/azure-sdk-for-python/azure-sdk-for-python/.venv/bin/python /home/REDACTED/work/azure-sdk-for-python/azure-sdk-for-python/.venv/lib/python3.9/site-packages/pip/__pip-REDACTED__.py install --ignore-installed --no-user --prefix /tmp/pip-build-env-gea9m53z/overlay --no-warn-script-location --no-binary :none: --only-binary :none: -i REDACTED -- setuptools>=40.8.0 /aio/operations/bash /aio/operations/--norc /aio/operations/--noprofile /aio/operations/_featurestore_entity_versions_operations.py (dns block)
• scanning-api.github.com
  • Triggering command: /home/REDACTED/work/_temp/ghcca-node/node/bin/node /home/REDACTED/work/_temp/ghcca-node/node/bin/node --enable-source-maps /home/REDACTED/work/_temp/copilot-developer-action-main/dist/index.js (dns block)

If you need me to access, download, or install something from one of these locations, you can either:

• Configure Actions setup steps to set up my environment, which run before the firewall is enabled
• Add the appropriate URLs or hosts to the custom allowlist in this repository's Copilot coding agent settings (admins only)

Original prompt

---

This section details on the original issue you should resolve

<issue_title>Bug: Optional command Component Input Not Working—Fails When Argument Not Passed or Empty (AzureML SDK v2)</issue_title>
<issue_description>- Package Name: azure-ai-ml

• Package Version: 1.27.1
• Operating System: Ubuntu 22.04.5 LTS
• Python Version: Python 3.12.11

Describe the bug
When defining a command component in AzureML SDK v2, a string input marked as optional=True with a default value (or not) does not behave as expected.
If I either omit the input or pass an empty string (""), I receive a Request is invalid and/or missing fields. The Value field is required. error, even though the documentation states this is supported.

To Reproduce
Steps to reproduce the behavior:

1. Define a component input as optional:

from azure.ai.ml.entities import Input

inputs = {
    "run_date" : Input(type=STRING),
    "tags" : Input(type=STRING, optional=True, default="yolo")
}

2. Use the following command in the component:

arguments = [
    "--run_date ${{inputs.run_date}}",
    "$[[--tags ${{inputs.tags}}]]"
]

3. Submit a job without specifying "tags" in inputs_run, or set "tags": "".

python_step_component.compute = COMPUTE_TARGET_RUN
inputs_run = {
    "run_date": "2023-01-31",
    }
# Submit the job
job = ml_client.jobs.create_or_update(
    python_step_component(
        **inputs_run
    ),
    experiment_name="test-optional-args",
    
)

4. Error occurs:

HttpResponseError: (UserError) Request is invalid and/or missing fields.  
(RequestInvalid) The Value field is required.

Click for full error

HttpResponseError                         Traceback (most recent call last)
Cell In[6], line 13
      5 inputs_run = {
      6     "run_date": "***redacted****",
      7     "use_case": "***redacted****",
      8     "kedro_pipeline_to_run": "***redacted****",
      9     # "tags": "aa"  # Optional, default is ''
     10 }
     12 # 4. Submit the component as a standalone job
---> 13 job = ml_client.jobs.create_or_update(
     14     python_step_component(**inputs_run),
     15     experiment_name="nptb_bpyn_mt"
     16 )
     18 print(f"Submitted job name: {job.name}")

File /anaconda/envs/nptb_bpyn/lib/python3.12/site-packages/azure/core/tracing/decorator.py:138, in distributed_trace.<locals>.decorator.<locals>.wrapper_use_tracer(*args, **kwargs)
    136             for key, value in span_attributes.items():
    137                 span.add_attribute(key, value)  # type: ignore
--> 138             return func(*args, **kwargs)
    139 else:
    140     # Native path
    141     config = {}

File /anaconda/envs/nptb_bpyn/lib/python3.12/site-packages/azure/ai/ml/_telemetry/activity.py:371, in monitor_with_telemetry_mixin.<locals>.monitor.<locals>.wrapper(*args, **kwargs)
    369 dimensions = {**parameter_dimensions, **(custom_dimensions or {})}
    370 with log_activity(logger, activity_name or f.__name__, activity_type, dimensions) as activityLogger:
--> 371     return_value = f(*args, **kwargs)
    372     if not parameter_dimensions:
    373         # collect from return if no dimensions from parameter
    374         activityLogger.activity_info.update(_collect_from_return_value(return_value))

File /anaconda/envs/nptb_bpyn/lib/python3.12/site-packages/azure/ai/ml/operations/_job_operations.py:713, in JobOperations.create_or_update(self, job, description, compute, tags, experiment_name, skip_validation, **kwargs)
    707 if (rest_job_resource.properties.job_type == RestJobType.PIPELINE) or (
    708     hasattr(rest_job_resource.properties, "identity")
    709     and (isinstance(rest_job_resource.properties.identity, UserIdentity))
    710 ):
    711     self._set_headers_with_user_aml_token(kwargs)
--> 713 result = self._create_or_update_with_different_version_api(rest_job_resource=rest_job_resource, **kwargs)
    715 if is_local_run(result):
    716     ws_base_url = self._all_operations.all_operations[
    717         AzureMLResourceType.WORKSPACE
    718     ]._operation._client._base_url

File /anaconda/envs/nptb_bpyn/lib/python3.12/site-packages/azure/ai/ml/operations/_job_operations.py:761, in JobOperations._create_or_update_with_different_version_api(self, rest_job_resource, **kwargs)
    758 if rest_job_resource.properties.job_type == RestJobType.COMMAND:
    759     service_client_operation = self.service_client_01_2025_preview.jobs
--> 761 result = service_client_operation.create_or_update(
    762     id=rest_job_resource.name,
    763     resource_group_name=self._operation_scope.resource_group_name,
    764     workspace_name=self._workspace_name,
    765     bod...

</details>



<!-- START COPILOT CODING AGENT SUFFIX -->

- Fixes Azure/azure-sdk-for-python#41834

<!-- START COPILOT CODING AGENT TIPS -->
---

✨ Let Copilot coding agent [set things up for you](https://github.com/Azure/azure-sdk-for-python/issues/new?title=✨+Set+up+Copilot+instructions&body=Configure%20instructions%20for%20this%20repository%20as%20documented%20in%20%5BBest%20practices%20for%20Copilot%20coding%20agent%20in%20your%20repository%5D%28https://gh.io/copilot-coding-agent-tips%29%2E%0A%0A%3COnboard%20this%20repo%3E&assignees=copilot) — coding agent works faster and does higher quality work when set up for your repo.