feat(indexes): add vector_precision field to index operations - #194
hotdata-automation[bot] wants to merge 1 commit into
Conversation
|
|
||
| ### Changed | ||
|
|
||
| - feat(indexes): add vector_precision field to index operations |
There was a problem hiding this comment.
This changelog does not record the removal of two public methods. ConnectionsApi.purge_connection_cache and ConnectionsApi.purge_table_cache are gone from hotdata/api/connections_api.py, along with their entries in docs/ConnectionsApi.md.
Add a **Breaking:** bullet for both methods under the existing ### Removed section, matching the entries at CHANGELOG.md:12-20.
Failure scenario: a user upgrades to the next release and calls connections_api.purge_connection_cache(connection_id). Python raises AttributeError. The changelog announces no removal, so the user has no way to learn that the endpoint retired or what replaces it.
| ### Changed | ||
|
|
||
| - feat(indexes): add vector_precision field to index operations | ||
| - chore(databases): clarify fork naming behavior |
There was a problem hiding this comment.
nit: this PR adds four public methods that no changelog entry names (not blocking). The new methods are DatabasesApi.get_database_lineage, DatabasesApi.lookup_database_by_name, DatabasesApi.set_database_table_constant_per_key, and ConnectionsApi.set_managed_table_constant_per_key. Six new models arrive with them.
Add an ### Added section listing the four methods. Consequence: users reading the release notes do not discover the new endpoints.
There was a problem hiding this comment.
Review
Context note: the full diff was omitted from the review prompt because of its size. gh pr diff 194 returned 403 KB, also too large to read in one pass. Review therefore combined the hunk list from that diff with reads of these files in the checkout: hotdata/models/create_index_request.py, hotdata/models/index_info_response.py, hotdata/models/index_entry_response.py, hotdata/models/load_managed_table_request.py, hotdata/models/database_lineage_response.py, hotdata/models/update_managed_table_request.py, hotdata/__init__.py, hotdata/models/__init__.py, CHANGELOG.md, tests/integration/test_connections_read.py, .github/workflows/integration-tests.yml, and .github/workflows/check-release.yml.
The vector_precision addition itself is consistent: the field, its enum validator, the two response models, the docs and the generated tests all agree. Two problems sit in the rest of the change, which the PR title and description do not mention.
Blocking Issues
-
tests/integration/test_connections_read.py:32callsconnections_api.purge_connection_cache(connection_id). This PR removes that method fromConnectionsApi. The test carries@pytest.mark.skiptoday, so pytest collects it and never runs the body, which is why CI stays green. The skip reason says "Re-enable once the seeded connection is restored", so the body is meant to run again. Whoever lifts the skip getsAttributeError: 'ConnectionsApi' object has no attribute 'purge_connection_cache'instead of the scenario the test covers. Delete lines 31-32 and drop "and cache purge" from the module docstring at line 4. The stale mention attests/integration/test_managed_tables_lifecycle.py:17is prose only and can stay. -
CHANGELOG.mdrecords no removal forpurge_connection_cacheandpurge_table_cache. See the inline comment on CHANGELOG.md line 24.
Action Required
- Remove the
purge_connection_cachecall and its comment fromtests/integration/test_connections_read.py, and update that file's docstring. - Add
**Breaking:**bullets for both removed methods under the Removed section inCHANGELOG.md.
CI state at review time: every check was queued or in progress, so this review asserts nothing about test results.
Auto-generated from the updated HotData OpenAPI spec.
Source: https://github.com/hotdata-dev/www/pull/428