fix: nullable types with gen ai bump

[MarlzRana]

Link to Issue or Description of Change

1. Link to an existing issue (if applicable):

• Closes: Nullable Fields Broken with Google GenAI Bump

Problem:
In 1.45.0 of the google-genai Python SDK release, changes were made to handle null types out of the box. This conflicts with the code in the google-adk to handle this, and causes the below exception:

Error in event_generator: 400 INVALID_ARGUMENT. {'error': {'code': 400, 'message': 'Unable to submit request because `create_payee` functionDeclaration `parameters.person.age` schema specified other fields alongside any_of. When using any_of, it must be the only field set. Learn more: https://cloud.google.com/vertex-ai/generative-ai/docs/multimodal/function-calling', 'status': 'INVALID_ARGUMENT'}}

This is what the Gemini schema looks like when they collide:

"age": {
          "additional_properties": null,
          "defs": null,
          "ref": null,
          "any_of": [
            {
              "additional_properties": null,
              "defs": null,
              "ref": null,
              "any_of": null,
              "default": null,
              "description": null,
              "enum": null,
              "example": null,
              "format": null,
              "items": null,
              "max_items": null,
              "max_length": null,
              "max_properties": null,
              "maximum": null,
              "min_items": null,
              "min_length": null,
              "min_properties": null,
              "minimum": null,
              "nullable": null,
              "pattern": null,
              "properties": null,
              "property_ordering": null,
              "required": null,
              "title": null,
              "type": "STRING"
            },
            {
              "additional_properties": null,
              "defs": null,
              "ref": null,
              "any_of": null,
              "default": null,
              "description": null,
              "enum": null,
              "example": null,
              "format": null,
              "items": null,
              "max_items": null,
              "max_length": null,
              "max_properties": null,
              "maximum": null,
              "min_items": null,
              "min_length": null,
              "min_properties": null,
              "minimum": null,
              "nullable": true,
              "pattern": null,
              "properties": null,
              "property_ordering": null,
              "required": null,
              "title": null,
              "type": "OBJECT"
            }
          ],
          "default": null,
          "description": "Person's age",
          "enum": null,
          "example": null,
          "format": null,
          "items": null,
          "max_items": null,
          "max_length": null,
          "max_properties": null,
          "maximum": null,
          "min_items": null,
          "min_length": null,
          "min_properties": null,
          "minimum": null,
          "nullable": null,
          "pattern": null,
          "properties": null,
          "property_ordering": null,
          "required": null,
          "title": null,
          "type": "OBJECT"
        },

As one can observe, this schema does have fields specified alongside any_of and it is also incorrect logically:

• the overall schema is not marked as nullable
• we should not need to use any_of, to define an extra schema for null

Instead, we should be seeing the absence of any_of where we have a nullable type. Instead we should see the type specified for the type, and nullable set to true.

The line in the code that cause this is:

adk-python/src/google/adk/tools/_gemini_schema_util.py

Lines 96 to 97 in 0dbfb43

	elif schema.get("type") == "null":
	schema["type"] = ["object", "null"]

It doesn't correctly handle the situation were we define a nullable using anyOf (where one type is the non-nullable type, and the other is the null type), as it's recursion operates on the individual type property level - it does not consider a type value in the context of another, as one has with anyOf.

Solution:

Let's defer to the changes to handle null types properly (one of the mechanisms being flattening an any_of type pair with one element being null).

This results in the correct Gemini schema being produced:

"age": {
          "additional_properties": null,
          "defs": null,
          "ref": null,
          "any_of": null,
          "default": null,
          "description": "Person's age",
          "enum": null,
          "example": null,
          "format": null,
          "items": null,
          "max_items": null,
          "max_length": null,
          "max_properties": null,
          "maximum": null,
          "min_items": null,
          "min_length": null,
          "min_properties": null,
          "minimum": null,
          "nullable": true,
          "pattern": null,
          "properties": null,
          "property_ordering": null,
          "required": null,
          "title": null,
          "type": "STRING"
        },

Testing Plan

Unit Tests:

• I have added or updated unit tests for my change.
• All unit tests pass locally.

Manual End-to-End (E2E) Tests:

I utilized the ADK web, and no longer observed the error.

Checklist

• I have read the CONTRIBUTING.md document.
• I have performed a self-review of my own code.
• I have commented my code, particularly in hard-to-understand areas.
• I have added tests that prove my fix is effective or that my feature works.
• New and existing unit tests pass locally with my changes.
• I have manually tested my changes end-to-end.
• Any dependent changes have been merged and published in downstream modules.

Additional context

This was quite a tricky issue to debug, but here are the notes I took:

I hope this helps and I look forward to hearing back from you soon! 😄

[google-cla]

Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

View this failed invocation of the CLA check for more information.

For the most up to date status, view the checks section at the bottom of the pull request.

[gemini-code-assist]

Summary of Changes

Hello @MarlzRana, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request refines the schema sanitization process for Gemini, primarily focusing on how nullable types are interpreted and converted. It includes an update to the google-genai SDK and a significant refactor of the schema utility, removing a custom sanitization function in favor of a more robust approach that correctly handles complex type definitions, especially those involving multiple types or anyOf constructs with nullability. The changes aim to improve the accuracy and consistency of schema representation.

Highlights

• Dependency Update: The google-genai SDK dependency has been updated from version >=1.41.0 to >=1.44.0 in pyproject.toml.
• Schema Sanitization Refactor: The _sanitize_schema_type helper function, which previously handled schema type normalization, has been removed. The _sanitize_schema_formats_for_gemini function was updated to reflect this change and now defaults an empty schema to type 'object'.
• Improved Nullable Type Handling: The logic for interpreting nullable types has been enhanced. Schemas defining multiple types, including 'null' (e.g., ['string', 'null', 'integer']), are now correctly represented using any_of for the non-null types and setting the nullable property to true.
• Enhanced anyOf Support: New unit tests have been added to specifically validate the correct processing of anyOf schemas that include 'null' as one of the possible types, ensuring accurate conversion to Gemini's schema representation.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[adk-bot]

Response from ADK Triaging Agent

Hello @MarlzRana, thank you for your contribution!

Before we can review this PR, please address the following items from our contribution guidelines:

• Sign the Contributor License Agreement (CLA): It looks like the CLA check has failed. You'll need to sign the Google CLA before we can accept your contribution.
• Link to an Issue: This PR contains a bug fix. Could you please create a GitHub issue that this PR addresses and link it in the PR description? If an issue already exists, please link it.
• Provide a Testing Plan: Could you please fill out the "Testing Plan" section in the PR description to explain how you've tested these changes?

You can find more details in our contribution guide.

Thank you!

[gemini-code-assist]

Code Review

This pull request updates the google-genai dependency and refactors the schema sanitization logic to leverage improvements in the new version. The removal of the custom _sanitize_schema_type function simplifies the codebase, and the updated logic now correctly handles nullable and multi-typed fields by using anyOf, which is a significant improvement over the previous implementation. The test suite has been updated accordingly with new tests for nullable anyOf cases and adjustments to existing tests to reflect the more accurate schema conversion. The changes are well-implemented and improve both correctness and maintainability.

[speedstorm1]

@speedstorm1 will take a look at this

[xuanyang15]

Thank you @speedstorm1, this PR is merged now.