remove custom-metadata.md; canonical path is vdbs + notebooks#2195
remove custom-metadata.md; canonical path is vdbs + notebooks#2195kheiss-uwzoo wants to merge 10 commits into
Conversation
Greptile SummaryThis PR removes
|
| Filename | Overview |
|---|---|
| docs/docs/extraction/vdbs.md | Metadata section expanded with prose, parameter summary, service upload flow, and links; two stale cross-links to deleted custom-metadata.md replaced with notebook links and the in-page anchor. |
| docs/docs/extraction/notebooks/index.md | Both metadata notebooks now listed (metadata_and_filtered_search.ipynb added, nemo_retriever_retriever_query_metadata_filter.ipynb retained), resolving the prior gap raised in review. |
| docs/mkdocs.yml | Removed nav entries for both deleted pages, renumbered sections 7-12, added redirects for both deleted files. |
| nemo_retriever/tests/test_src_documentation_snippets.py | _PUBLIC_RETRIEVER_DOCS updated to reference vdbs.md instead of the deleted custom-metadata.md; path resolution logic handles docs/-prefixed paths correctly. |
| docs/docs/extraction/custom-metadata.md | Deleted; content consolidated into vdbs.md#metadata-and-filtering and notebook examples; mkdocs redirect in place. |
| docs/docs/extraction/integrations-langchain-llamaindex-haystack.md | Deleted; notebook links already present in notebooks/index.md; mkdocs redirect added. |
| docs/docs/extraction/workflow-agentic-retrieval.md | Cross-link updated from deleted custom-metadata.md to vdbs.md#metadata-and-filtering. |
| docs/docs/extraction/overview.md | Integration pathway link updated from deleted integrations-langchain-llamaindex-haystack.md to notebooks/index.md. |
Flowchart
%%{init: {'theme': 'neutral'}}%%
flowchart TD
A[User visits old URL custom-metadata.md] -->|MkDocs redirect| B[vdbs.md#metadata-and-filtering]
C[User visits old URL integrations page] -->|MkDocs redirect| D[notebooks/index.md]
B --> E[Prose: sidecar meta_* params, compact JSON, service flow]
B --> F[nemo_retriever_retriever_query_metadata_filter.ipynb]
B --> G[VDB README Metadata Filtering]
D --> F
D --> H[metadata_and_filtered_search.ipynb]
%%{init: {'theme': 'base', 'themeVariables': {"darkMode": true, "background": "#0d1117", "primaryColor": "#21262d", "primaryTextColor": "#e6edf3", "primaryBorderColor": "#8b949e", "lineColor": "#8b949e", "textColor": "#e6edf3", "edgeLabelBackground": "#161b22", "actorBkg": "#21262d", "actorBorder": "#8b949e", "actorTextColor": "#e6edf3", "actorLineColor": "#8b949e", "signalColor": "#8b949e", "signalTextColor": "#e6edf3", "noteBkgColor": "#373320", "noteBorderColor": "#d4a72c", "noteTextColor": "#f0e6c0", "labelBoxBkgColor": "#21262d", "labelBoxBorderColor": "#8b949e", "labelTextColor": "#e6edf3", "loopTextColor": "#e6edf3", "activationBkgColor": "#30363d", "activationBorderColor": "#8b949e"}}}%%
flowchart TD
A[User visits old URL custom-metadata.md] -->|MkDocs redirect| B[vdbs.md#metadata-and-filtering]
C[User visits old URL integrations page] -->|MkDocs redirect| D[notebooks/index.md]
B --> E[Prose: sidecar meta_* params, compact JSON, service flow]
B --> F[nemo_retriever_retriever_query_metadata_filter.ipynb]
B --> G[VDB README Metadata Filtering]
D --> F
D --> H[metadata_and_filtered_search.ipynb]
Reviews (16): Last reviewed commit: "Merge upstream/main into docs/consolidat..." | Re-trigger Greptile
Drop dead metadata_and_filtered_search notebook links; document retriever service sidecar upload on vdbs.md instead of delegating to VDB README.
Delete integrations-langchain-llamaindex-haystack.md, point inbound links at notebooks/index.md, and add a mkdocs redirect.
Replace duplicated metadata prose with a single notebook link per review.
Revert doc-snippet test list change; belongs outside this docs-only PR.
Users arriving via the deleted custom-metadata.md URL need a short overview of meta_* sidecar params and filter modes, plus links to the worked notebooks and VDB README—not a bare hyperlink alone.
Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com>
Restore vdbs.md metadata landing content with service sidecar guidance, drop dead metadata_and_filtered_search.ipynb links, and point the doc-snippet test registry at vdbs.md instead of deleted custom-metadata.md.
Remove reindex_example.ipynb entry (notebook removed on main in NVIDIA#2163). Rename framework cross-links to Starter kits to match mkdocs nav label.
a541334 to
fafaf61
Compare
Resolve modify/delete conflict on custom-metadata.md by keeping the PR deletion; canonical metadata docs live in vdbs.md with mkdocs redirect.
Summary
Follows Julio's NVBugs 6205401 guidance: VDB/metadata facts live on �dbs.md; runnable walkthroughs stay in notebooks.
Notebooks (canonical examples)
Operator/API reference remains in nemo_retriever/src/nemo_retriever/vdb/README.md.
Test plan