fix(go): nested render and generate spans under the dotprompt span

[apascal07]

Fixes #4063.

Checklist (if applicable):

• PR title is following https://www.conventionalcommits.org/en/v1.0.0/
• Tested (manually, unit tested, etc.)
• Docs updated (updated docs or a docs bug required)

[gemini-code-assist]

Summary of Changes

Hello @apascal07, 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 primarily focuses on improving the tracing capabilities within the Go codebase, particularly for AI-related operations. It refactors how tracing spans are created and nested, ensuring a clearer hierarchy for render and generate operations under a dotprompt execution. Additionally, it introduces a new method for executing actions without automatic tracing, and a utility function for comprehensive nil checks, enhancing both observability and code robustness.

Highlights

• Tracing Span Refactoring: Simplified the creation of SpanMetadata objects across evaluator, generate, and flow components by introducing a new tracing.Span helper function, reducing boilerplate.
• Nested Prompt Spans: Modified Prompt.Execute to ensure render and generate operations are properly nested as child spans under a main dotprompt execution span, providing a clearer hierarchy in tracing.
• Raw Action Execution: Introduced ActionDef.RunRaw to allow direct execution of an action's function without automatic trace span creation, providing more control for callers managing their own tracing contexts.
• Nil Stream Value Handling: Added logic to GenerateDataStream and DataPrompt.ExecuteStream to skip yielding if a stream value is nil, preventing potential issues with incomplete JSON during streaming.
• IsNil Utility Function: Implemented a generic IsNil helper function in go/internal/base/misc.go for robust nil checking across various Go types, including pointers, interfaces, maps, slices, channels, and functions.

🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

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. ↩

[gemini-code-assist]

Code Review

This pull request refactors tracing to correctly nest render and generate spans within a dotprompt span, which is a great improvement for trace visibility. The changes are mostly clean and use new helpers like tracing.Span and base.IsNil to simplify the code. However, I've found an issue with the new base.IsNil check for skipping incomplete stream chunks: it doesn't work correctly for non-pointer struct types and can cause the stream to terminate with an error, which I've detailed in my comments.

[gemini-code-assist]

This check only works correctly when Out is a pointer, slice, map, or interface type. If Out is a struct type, this change introduces a bug where the stream will error out on an incomplete JSON chunk instead of skipping it.

This happens because chunk.Output() will try to unmarshal null into a struct value when a partial chunk is parsed, which results in a json.UnmarshalTypeError. This error is then yielded, terminating the stream, instead of skipping the chunk as intended.

To properly fix this, chunk.Output should be modified to handle this case without erroring, or this logic needs to handle that specific error. Since chunk.Output is not in this diff, a more robust solution should be considered to make this feature work consistently for all types.

[gemini-code-assist]

This check only works correctly when Out is a pointer, slice, map, or interface type. If Out is a struct type, this change introduces a bug where the stream will error out on an incomplete JSON chunk instead of skipping it.

This happens because chunk.Output() will try to unmarshal null into a struct value when a partial chunk is parsed, which results in a json.UnmarshalTypeError. This error is then yielded, terminating the stream, instead of skipping the chunk as intended.

To properly fix this, chunk.Output should be modified to handle this case without erroring, or this logic needs to handle that specific error. Since chunk.Output is not in this diff, a more robust solution should be considered to make this feature work consistently for all types.