fix: normalize streamed Markdown text - #30
Merged
Conversation
There was a problem hiding this comment.
Pull request overview
This PR introduces a normalization layer for streamed / persisted message text so that Markdown rendering and related UI behaviors always receive plain strings, even when upstream providers return dynamic (non-string) chunks.
Changes:
- Wrap streamed chunk extraction (Anthropic SSE, Gemini, OpenRouter) with a new
stream-texthelper before concatenation/persistence. - Normalize
:content/:thinkingfrom existing and archived messages before JSON detection, Markdown rendering, and copy actions. - Add
stream-texthelper function to centralize the coercion/guard behavior.
💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Normalizes dynamic chunks from Gemini, OpenRouter, and Anthropic before concatenation and persistence. Existing or archived messages are normalized before Markdown rendering, JSON detection, and copy actions, preventing objects from reaching the Markdown string parser. Respo Markdown input guard is tracked separately in Respo/respo-markdown.calcit PR 33.