Implement per-page tagging and JSON writer enhancements#68
Merged
dpomian merged 2 commits intoAmadeusITGroup:mainfrom Mar 4, 2026
Merged
Implement per-page tagging and JSON writer enhancements#68dpomian merged 2 commits intoAmadeusITGroup:mainfrom
dpomian merged 2 commits intoAmadeusITGroup:mainfrom
Conversation
- scrollwordexporter: add per-URL tag overrides via page_url_tags/default_tag - json_writer: add SHA-256 checksum gate to skip unchanged downstream processing - teams_qna_loader: support per-item tag override from JSON - config_schema: add default_tag, page_url_tags, checksum_path, skip_downstream_if_unchanged, batch_size(exitsing parameter from vector skill) - docs: update indexer-skills.md accordingly
- json-writer: checksum per chunk (document_id → sha256(content)), not whole pipeline - confluence-faq-splitter: document_id from question only, stable across answer edits - indexer-skills.md: updated to reflect above points
dpomian
approved these changes
Mar 4, 2026
Comment on lines
+134
to
+135
| # Remove unchanged chunks from this document | ||
| if unchanged_chunks: |
Contributor
There was a problem hiding this comment.
There's no need for the if statement. You can say doc.chunks -= unchanged_chunks. If the unchanged_chunks is empty, it will be a no-op
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Per-chunk change detection (
JSONWriterSkill) — Newchecksum_path+skip_downstream_if_unchangedparams. SHA-256 checksums are stored perdocument_id; unchanged chunks are stripped so downstream embedding/indexing is skipped for them.Stable
document_id(ConfluenceFAQSplitter) — Chunk ID now hashed from question text only (was full Q+A), so the ID is stable when only the answer changes — enabling clean Azure AI Search upserts.Per-page tags (
ScrollWordExporter) —page_ids/page_urlsentries changed from bare strings to dicts ({id/url, tag?}), with per-page tag falling back to a top-leveltagparam.Per-item tags (
TeamsQnALoaderSkill) — Each Q&A object in the JSON can now include a"tag"field to override the skill-level default.Config schema + docs —
config_schema.yamlupdated to match;indexer-skills.mdadds a new use-case example (#6), a Writer Skills section, and updated YAML snippets throughout.