[Bugfix] input prompt was not logged

[alphabetc1]

Motivation

Input prompt was not logged when --log-requests was enabled and --log-requests-level was set to 2 or 3.

log was enabled:

while text is None:

after fix:

Modifications

Accuracy Tests

Benchmarking and Profiling

Checklist

• Format your code according to the Format code with pre-commit.
• Add unit tests according to the Run and add unit tests.
• Update documentation according to Write documentations.
• Provide accuracy and speed benchmark results according to Test the accuracy and Benchmark the speed.
• Follow the SGLang code style guidance.
• Work with maintainers to merge your PR. See the PR Merge Process

[gemini-code-assist]

Summary of Changes

Hello @alphabetc1, 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 addresses a logging oversight where the actual text of input prompts was not being captured in logs under certain configurations, particularly when prompts were processed as token IDs. The changes ensure comprehensive logging of input prompts, improving debuggability and traceability for requests, and also refine the internal handling of prompt inputs to correctly determine batch sizes.

Highlights

• Fix Input Prompt Logging: Resolved a bug where the input prompt text was not logged when --log-requests was enabled and --log-requests-level was 2 or 3, specifically when prompts were provided as input_ids.
• Prioritize Input IDs for Logging: Modified the _convert_to_internal_request function to explicitly decode and log the prompt text from input_ids under the specified logging conditions, ensuring the full prompt content is captured.
• Reordered Batch Size Determination Logic: Adjusted the _determine_batch_size logic in io_struct.py to prioritize input_ids over text when inferring batch size, aligning with the primary input method.

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 fixes a bug where the input prompt was not logged when using input_ids. The changes correctly add logic to decode and log the prompt text when necessary. Additionally, the change in io_struct.py to prioritize input_ids over text is a good improvement for consistency. I have one suggestion to make the logging check logic more concise.

[gemini-code-assist]

This logic for checking whether to log the prompt text can be made more concise. You can combine the getattr calls into a single if statement to improve readability and reduce the number of local variables.

                server_args = self.tokenizer_manager.server_args
                if (
                    getattr(server_args, "log_requests", False)
                    and getattr(server_args, "log_requests_level", 0) >= 2
                ):
                    prompt_kwargs["text"] = (
                        processed_messages.prompt
                        or self.tokenizer_manager.tokenizer.decode(
                            processed_messages.prompt_ids
                        )
                    )

[alphabetc1]

D

[hnyls2002]

Do not do de-tokenization here.

[alphabetc1]

why? or can i de-tokeniozation here

[hnyls2002]

Add a #FIXME comment here, saying this can be moved elsewhere.

[alphabetc1]

D

[hnyls2002]

/tag-and-rerun-ci