Repository navigation
fix: raise hotdata SDK floor to 0.7 - #35
Merged
Merged
Conversation
hotdata-dlt-destination's in-progress native load-modes PR bumps its own hotdata dependency to >=0.7,<0.8; its optional [ibis] extra pulls in hotdata-ibis, so the old <0.7 cap here made co-installation unsatisfiable. 0.7.0 is additive for the surface this backend uses, verified via the full offline suite run directly against hotdata==0.7.0.
There was a problem hiding this comment.
nit (not blocking): README.md:5 still advertises hotdata ≥0.6,<0.7 — worth bumping to ≥0.7,<0.8 so it matches the new pyproject.toml floor. (Couldn't attach inline since README isn't part of this diff.)
The bump itself is clean, well-documented, and consistent across pyproject.toml, uv.lock, and the changelog.
Review nit on #35: README.md still advertised the old >=0.6,<0.7 range.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Bumps the
hotdataSDK dependency floor from>=0.6,<0.7to>=0.7,<0.8, tracking its current minor per this project's SDK-pinning convention (the same "cap to current minor" approachhotdata-dlt-destinationdocuments).Why
hotdata-dlt-destination's in-progress native append/replace/upsert PR (#38) bumps its ownhotdatadependency to>=0.7.0,<0.8. That repo's optional[ibis]extra pulls inhotdata-ibis>=0.3,<0.4, which still cappedhotdata<0.7— makingpip install hotdata-dlt-destination[ibis]unsatisfiable once #38 lands (hotdata<0.7vshotdata>=0.7.0can't both hold).Is 0.7.0 safe?
Diffed the exact SDK surface
hotdata-ibisimports (DatabasesApi,ConnectionsApi,LoadManagedTableRequest,DatabaseDefaultTableDecl, etc.) between SDKv0.6.0andv0.7.0: purely additive — an unused optionalkeyfield on table declarations, andLoadManagedTableRequest.upload_idwidened from required to optional (we always pass it, so unaffected).Verified two ways:
hotdata==0.7.0directly into the venv and ran the full offline suite — 101 passed — before regeneratinguv.lockthe normal way.examples/04_ibis_table_workflows.py(project/filter/group_by/join/order_by/scalar-agg over a managed database) andexamples/05_roundtrip_demo.py(full create → write → read → drop lifecycle) both pass end-to-end.Testing
uv lockpicked uphotdata v0.6.0 -> v0.7.0.uv run pytest -q— 101 passed.examples/04_ibis_table_workflows.pyandexamples/05_roundtrip_demo.pyagainst hosted Hotdata — both pass.Checklist
CHANGELOG.mdupdated ([Unreleased])uv.lockregenerated