Fix message content field handling for tool calls and multimodal input #4029
+410
−5
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This commit fixes two issues with message content processing: 1. **Missing content field in tool call responses**: When assistant messages contain tool_calls, the content field may be missing or None, causing KeyError when checking isinstance(msg['content'], list). 2. **Incomplete multimodal text merging**: Previous implementation only extracted the first text block. Now correctly merges all text blocks with newline separator. Changes: - Add _merge_message_content() module-level function in async_engine.py - Handle 4 cases: missing content, None content, string content, list content - Convert None/missing content to empty string '' (prevents Jinja2 errors) - Use '\n'.join() to merge text blocks (matches vLLM behavior) - Preserve all message fields (e.g., tool_calls, name, etc.) - Add comprehensive unit tests (19 test cases in test_content_merge.py) Implementation based on vLLM's content normalization logic: https://github.com/vllm-project/vllm/blob/main/vllm/entrypoints/chat_utils.py 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <[email protected]>
This was referenced Oct 11, 2025
Open
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This commit fixes two issues with message content processing:
Missing content field in tool call responses: When assistant messages contain tool_calls, the content field may be missing or None, causing KeyError when checking isinstance(msg['content'], list).
Incomplete multimodal text merging: Previous implementation only extracted the first text block. Now correctly merges all text blocks with newline separator.
Changes:
Implementation based on vLLM's content normalization logic: https://github.com/vllm-project/vllm/blob/main/vllm/entrypoints/chat_utils.py
🤖 Generated with Claude Code