[https://nvbugs/5944411][fix] Handle anyOf parameter schemas in Qwen3Coder tool parser by tijyojwad · Pull Request #12173 · NVIDIA/TensorRT-LLM

tijyojwad · 2026-03-12T23:40:51Z

Parameters using anyOf (e.g. Optional[int]) have no top-level "type" key, causing _convert_param_value to always default to "string". Treat anyOf schemas as "object" to trigger json.loads, which correctly parses integers, floats, booleans, objects, and arrays.

Made-with: Cursor

Fix cherry picked from vllm-project/vllm#36032

Summary by CodeRabbit

Release Notes

New Features
- Added support for OpenAPI-style anyOf schemas in parameter definitions, enabling proper type conversion and handling of union types.
Tests
- Added comprehensive unit tests for anyOf schema parameter parsing, validating type conversion and null value handling.

Description

Test Coverage

PR Checklist

Please review the following before submitting your PR:

PR description clearly explains what and why. If using CodeRabbit's summary, please make sure it makes sense.
PR Follows TRT-LLM CODING GUIDELINES to the best of your knowledge.
Test cases are provided for new code paths (see test instructions)
Any new dependencies have been scanned for license and vulnerabilities
CODEOWNERS updated if ownership changes
Documentation updated as needed
Update tava architecture diagram if there is a significant design change in PR.
The reviewers assigned automatically/manually are appropriate for the PR.
Please check this after reviewing the above items as appropriate for this PR.

GitHub Bot Help

To see a list of available CI bot commands, please comment /bot help.

Parameters using anyOf (e.g. Optional[int]) have no top-level "type" key, causing _convert_param_value to always default to "string". Treat anyOf schemas as "object" to trigger json.loads, which correctly parses integers, floats, booleans, objects, and arrays. Signed-off-by: Joyjit Daw <1127155+tijyojwad@users.noreply.github.com> Made-with: Cursor

coderabbitai · 2026-03-12T23:45:02Z

📝 Walkthrough

Walkthrough

Adds support for OpenAPI-style anyOf parameter schemas in the Qwen3 coder parser by treating anyOf entries as object types instead of defaulting to strings. When parameter config contains "anyOf", the parser now sets the parameter type to "object" to enable proper JSON parsing. Includes comprehensive unit tests validating type coercion and null value handling across multiple anyOf-based parameter types.

Changes

Cohort / File(s)	Summary
anyOf Schema Support `tensorrt_llm/serve/tool_parser/qwen3_coder_parser.py`	Added conditional logic to detect "anyOf" in parameter configs and set param_type to "object" for proper JSON deserialization of union-type parameters.
anyOf Parameter Parsing Tests `tests/unittest/llmapi/apps/test_tool_parsers.py`	Added two new test cases verifying anyOf schema parameter parsing: one testing type coercion (int, float, bool, dict, list, string) and another validating null value handling for anyOf integer/null parameters.

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~20 minutes

🚥 Pre-merge checks | ✅ 2 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Description check	⚠️ Warning	PR description lacks required sections: title format incomplete, Description and Test Coverage sections empty, checklist items not individually reviewed.	Add proper PR title with ticket format [https://nvbugs/5944411][fix] at top, fill Description and Test Coverage sections with details, and individually address each checklist item with clear explanations rather than just checking the final box.

✅ Passed checks (2 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title accurately and specifically describes the main change: handling anyOf parameter schemas in the Qwen3Coder tool parser, with proper formatting following the template.
Docstring Coverage	✅ Passed	Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

🧪 Generate unit tests (beta)

Create PR with unit tests
Post copyable unit tests in a comment

📝 Coding Plan

Generate coding plan for human review comments

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

Tip

You can disable sequence diagrams in the walkthrough.

Disable the reviews.sequence_diagrams setting to disable sequence diagrams in the walkthrough.

coderabbitai

Actionable comments posted: 1

🧹 Nitpick comments (1)

tests/unittest/llmapi/apps/test_tool_parsers.py (1)

1010-1130: Add one literal-like string case for anyOf: [string, null]

Great coverage overall. Consider adding a case where label is true (unquoted) and assert it remains "true" as a string. That will guard against unintended coercion in union-with-null string schemas.

✅ Suggested test addition

     def test_parse_anyof_parameter_type_conversion(self, parser):
@@
         assert params["tags"] == [1, 2, 3]
         assert params["label"] == "hello"
+
+        text_string_literal = ("<tool_call>\n"
+                               "<function=create_record>\n"
+                               "<parameter=name>test</parameter>\n"
+                               "<parameter=label>true</parameter>\n"
+                               "</function>\n"
+                               "</tool_call>")
+        result_string_literal = parser.detect_and_parse(text_string_literal, [tool_def])
+        params_string_literal = json.loads(result_string_literal.calls[0].parameters)
+        assert params_string_literal["label"] == "true"

🤖 Prompt for AI Agents

Verify each finding against the current code and only fix it if needed.

In `@tests/unittest/llmapi/apps/test_tool_parsers.py` around lines 1010 - 1130,
Update the anyOf string/null coverage by adding a literal-like case for the
"label" field in test_parse_anyof_parameter_type_conversion (or as a new test)
where the input uses an unquoted true token (<parameter=label>true</parameter>);
run parser.detect_and_parse and assert that result.calls[0].parameters parses
"label" to the string "true" (and isinstance(params["label"], str)) to ensure no
coercion to boolean occurs for anyOf: [{type: "string"}, {type: "null"}].

🤖 Prompt for all review comments with AI agents

Verify each finding against the current code and only fix it if needed.

Inline comments:
In `@tensorrt_llm/serve/tool_parser/qwen3_coder_parser.py`:
- Around line 377-379: The code currently forces param_type = "object" whenever
param_config[param_name] has "anyOf"; instead, inspect
param_config[param_name]["anyOf"] to find concrete non-null subtype(s) and set
param_type to that subtype when unambiguous (e.g., pick "string" if anyOf
contains {"type":"string"} and {"type":"null"}), otherwise fall back to
"object"; update the logic around param_config[param_name] and param_type so
values like true/false are parsed as boolean when anyOf includes boolean, and
ensure null remains allowed when anyOf includes null.

---

Nitpick comments:
In `@tests/unittest/llmapi/apps/test_tool_parsers.py`:
- Around line 1010-1130: Update the anyOf string/null coverage by adding a
literal-like case for the "label" field in
test_parse_anyof_parameter_type_conversion (or as a new test) where the input
uses an unquoted true token (<parameter=label>true</parameter>); run
parser.detect_and_parse and assert that result.calls[0].parameters parses
"label" to the string "true" (and isinstance(params["label"], str)) to ensure no
coercion to boolean occurs for anyOf: [{type: "string"}, {type: "null"}].

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: a9a72c29-b24c-40c1-afad-bb3d1a0701b6

📥 Commits

Reviewing files that changed from the base of the PR and between 5cc0ccd and a9353bc.

📒 Files selected for processing (2)

tensorrt_llm/serve/tool_parser/qwen3_coder_parser.py
tests/unittest/llmapi/apps/test_tool_parsers.py

tensorrt_llm/serve/tool_parser/qwen3_coder_parser.py

tijyojwad · 2026-03-12T23:55:06Z

/bot run

tensorrt-cicd · 2026-03-13T00:02:14Z

PR_Github #38792 [ run ] triggered by Bot. Commit: a9353bc Link to invocation

QiJune

LGTM

tensorrt-cicd · 2026-03-13T04:52:53Z

PR_Github #38792 [ run ] completed with state SUCCESS. Commit: a9353bc
/LLM/main/L0_MergeRequest_PR pipeline #30106 completed with status: 'FAILURE'

CI Report

⚠️ Action Required:

Please check the failed tests and fix your PR
If you cannot view the failures, ask the CI triggerer to share details
Once fixed, request an NVIDIA team member to trigger CI again

Link to invocation

tijyojwad · 2026-03-13T16:16:39Z

/bot run --only-multi-gpu-test --disable-fail-fast

tensorrt-cicd · 2026-03-13T16:22:59Z

PR_Github #38883 [ run ] triggered by Bot. Commit: a9353bc Link to invocation

tensorrt-cicd · 2026-03-13T19:37:48Z

PR_Github #38883 [ run ] completed with state SUCCESS. Commit: a9353bc
/LLM/main/L0_MergeRequest_PR pipeline #30191 (Partly Tested) completed with status: 'SUCCESS'
Pipeline passed with automatic retried tests. Check the rerun report for details.

CI Report

Link to invocation

2ez4bz · 2026-03-13T19:53:00Z

/bot help

github-actions · 2026-03-13T19:53:10Z

GitHub Bot Help

/bot [-h] ['run', 'kill', 'skip', 'reuse-pipeline'] ...

Provide a user friendly way for developers to interact with a Jenkins server.

Run /bot [-h|--help] to print this help message.

See details below for each supported subcommand.

Details

run [--reuse-test (optional)pipeline-id --disable-fail-fast --skip-test --stage-list "A10-PyTorch-1, xxx" --gpu-type "A30, H100_PCIe" --test-backend "pytorch, cpp" --add-multi-gpu-test --only-multi-gpu-test --disable-multi-gpu-test --post-merge --extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx" --detailed-log --debug(experimental) --high-priority]

Launch build/test pipelines. All previously running jobs will be killed.

--reuse-test (optional)pipeline-id (OPTIONAL) : Allow the new pipeline to reuse build artifacts and skip successful test stages from a specified pipeline or the last pipeline if no pipeline-id is indicated. If the Git commit ID has changed, this option will be always ignored. The DEFAULT behavior of the bot is to reuse build artifacts and successful test results from the last pipeline.

--disable-reuse-test (OPTIONAL) : Explicitly prevent the pipeline from reusing build artifacts and skipping successful test stages from a previous pipeline. Ensure that all builds and tests are run regardless of previous successes.

--disable-fail-fast (OPTIONAL) : Disable fail fast on build/tests/infra failures.

--skip-test (OPTIONAL) : Skip all test stages, but still run build stages, package stages and sanity check stages. Note: Does NOT update GitHub check status.

--stage-list "A10-PyTorch-1, xxx" (OPTIONAL) : Only run the specified test stages. Examples: "A10-PyTorch-1, xxx". Note: Does NOT update GitHub check status.

--gpu-type "A30, H100_PCIe" (OPTIONAL) : Only run the test stages on the specified GPU types. Examples: "A30, H100_PCIe". Note: Does NOT update GitHub check status.

--test-backend "pytorch, cpp" (OPTIONAL) : Skip test stages which don't match the specified backends. Only support [pytorch, cpp, tensorrt, triton]. Examples: "pytorch, cpp" (does not run test stages with tensorrt or triton backend). Note: Does NOT update GitHub pipeline status.

--only-multi-gpu-test (OPTIONAL) : Only run the multi-GPU tests. Note: Does NOT update GitHub check status.

--disable-multi-gpu-test (OPTIONAL) : Disable the multi-GPU tests. Note: Does NOT update GitHub check status.

--add-multi-gpu-test (OPTIONAL) : Force run the multi-GPU tests in addition to running L0 pre-merge pipeline.

--post-merge (OPTIONAL) : Run the L0 post-merge pipeline instead of the ordinary L0 pre-merge pipeline.

--extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx" (OPTIONAL) : Run the ordinary L0 pre-merge pipeline and specified test stages. Examples: --extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx".

--detailed-log (OPTIONAL) : Enable flushing out all logs to the Jenkins console. This will significantly increase the log volume and may slow down the job.

--debug (OPTIONAL) : Experimental feature. Enable access to the CI container for debugging purpose. Note: Specify exactly one stage in the stage-list parameter to access the appropriate container environment. Note: Does NOT update GitHub check status.

--high-priority (OPTIONAL) : Run the pipeline with high priority. This option is restricted to authorized users only and will route the job to a high-priority queue.

kill

kill

Kill all running builds associated with pull request.

skip

skip --comment COMMENT

Skip testing for latest commit on pull request. --comment "Reason for skipping build/test" is required. IMPORTANT NOTE: This is dangerous since lack of user care and validation can cause top of tree to break.

reuse-pipeline

reuse-pipeline

Reuse a previous pipeline to validate current commit. This action will also kill all currently running builds associated with the pull request. IMPORTANT NOTE: This is dangerous since lack of user care and validation can cause top of tree to break.

2ez4bz · 2026-03-13T19:54:09Z

/bot --reuse-test 30191

github-actions · 2026-03-13T19:54:19Z