Skip to content

fix: parse complete cloud chat SSE frames - #544

Open
rudycelekli wants to merge 1 commit into
VectifyAI:mainfrom
rudycelekli:fix/cloud-chat-sse-frame-parsing-20261001
Open

rudycelekli wants to merge 1 commit into
VectifyAI:mainfrom
rudycelekli:fix/cloud-chat-sse-frame-parsing-20261001

Conversation

@rudycelekli

Copy link
Copy Markdown

Cloud chat currently parses individual lines only when they start with data: , so valid compact or multiline SSE frames lose answer text and can hide a midstream error after a partial answer.

Assemble data fields at blank event boundaries for both streaming views, following SSE's optional single space, newline joining, UTF-8/BOM and incomplete-EOF rules. Preserve citation metadata, [DONE], error propagation and response cleanup. Existing mocked SSE fixtures now include their event separators.

Validation: native localhost HTTP regressions failed before the fix (27 cases). The focused stream suite passes 29 tests; removing the text parser's error check makes both corresponding error regressions fail. The complete local python -m pytest -q passes 612 tests, 218 skipped on macOS/Python 3.12/PDFium 5 without optional agent frameworks. The repository's hosted Python 3.10/3.13/framework/PDFium matrix remains required. No live provider calls were used.

Signed-off-by: Rudy Celekli <47457359+rudycelekli@users.noreply.github.com>

This branch has not been deployed

No deployments
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant