Staging - #314
Merged
Merged
Conversation
Seed answers now include [[cite:n]] next to the claims, and citation payloads copy page numbers and page-image URLs from the resolved chunk so Notebook can render {title}/pN.
Co-authored-by: Cursor <cursoragent@cursor.com>
- Added new debug scripts for PDF/PPT track-split debugging and page memory staged debugging. - Removed the unified debug_parse.py script, consolidating its functionality into more specialized scripts. - Updated the pptx_to_pdf_api function documentation for clarity. - Enhanced the page_memory argument parser to include a clean option for output directory management.
- Updated the probing mechanism to ensure distinct printed pages are used for each regime, preventing duplicate entries. - Adjusted the scanning start logic to prioritize the printed page over the TOC end, ensuring offsets are non-negative. - Added new tests to validate the updated probing behavior and confirm that the scanning logic adheres to the new rules.
- Renamed and updated debug scripts for page memory stages, reflecting changes in the debugging process. - Removed deprecated Stage 3 and Stage 4 scripts, consolidating their functionality into more streamlined processes. - Adjusted documentation and argument parsing to align with the new script structure and improve clarity. - Updated references in shared utilities to reflect the removal of obsolete stages.
Keep GOOG, Meta earnings-call, and SpaceX demo artifacts as the API returned them. Citation chips pin a page on the original Root chunk instead of inventing per-page cards. Co-authored-by: Cursor <cursoragent@cursor.com>
Keep GOOG, Meta earnings-call, and SpaceX Official Library artifacts as the API returned them instead of locally splitting skeleton jobs into per-page cards. Co-authored-by: Cursor <cursoragent@cursor.com>
…demo-parse fix: restore original API parse for skeleton demo documents
…ation-chips feat: put page chips on demo catalog Q&A
Make page-memory node page lists a declared source field for results and referenced_chunks, drop unused attach_citation / source-passthrough, and fix related test isolation/mocks so contract suites stay green. Co-authored-by: Cursor <cursoragent@cursor.com>
…phase1-probe refactor: calibration phase1 probe and debug script cleanup
…-page-nums feat: expose page_nums on retrieval source for page-memory nodes
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
2 participants
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.
No description provided.