fix(runtime): decouple summarization from state persistence to preserve full conversation history (#1393)#1430
Open
sjhddh wants to merge 1 commit intobytedance:mainfrom
Open
fix(runtime): decouple summarization from state persistence to preserve full conversation history (#1393)#1430sjhddh wants to merge 1 commit intobytedance:mainfrom
sjhddh wants to merge 1 commit intobytedance:mainfrom
Conversation
…ve full conversation history (bytedance#1393)
yanglinzhi
added a commit
to yanglinzhi/deer-flow
that referenced
this pull request
Mar 27, 2026
…estructive state overwriting
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Fixes #1393 and #1000.\n\nCurrently, the native
SummarizationMiddlewarereplaces the conversation history instate["messages"]with a generated summary plus a trailing window of messages. This permanently erases the original history from LangGraph checkpoints, causing the frontend to render truncated dialogue histories upon page refresh.\n\nThis patch introduces aSafeSummarizationMiddlewarewrapper that hijacksSummarizationMiddleware.abefore_model. Instead of returning a destructive state update withRemoveMessage(id=REMOVE_ALL_MESSAGES), it captures the summarized window and injects it directly into theModelRequest.messagesviaawrap_model_call. \n\nThis ensures that the token reduction strictly applies to the current inference window passed to the LLM, while leaving the authoritativestate["messages"]and DB checkpoints completely intact and immutable.