feat: remote index stats (#1702)

BREAKING CHANGE: the return value of `index_stats` method has changed
and all `index_stats` APIs now take index name instead of UUID. Also
several deprecated index statistics methods were removed.

* Removes deprecated methods for individual index statistics
* Aligns public `IndexStatistics` struct with API response from LanceDB
Cloud.
* Implements `index_stats` for remote Rust SDK and Python async API.
This commit is contained in:
Will Jones
2024-09-27 12:10:00 -07:00
committed by GitHub
parent c1d9d6f70b
commit f958f4d2e8
16 changed files with 337 additions and 307 deletions

View File

@@ -66,6 +66,15 @@ async def test_create_bitmap_index(some_table: AsyncTable):
# TODO: Fix via https://github.com/lancedb/lance/issues/2039
# indices = await some_table.list_indices()
# assert str(indices) == '[Index(Bitmap, columns=["id"])]'
indices = await some_table.list_indices()
assert len(indices) == 1
index_name = indices[0].name
stats = await some_table.index_stats(index_name)
assert stats.index_type == "BITMAP"
assert stats.distance_type is None
assert stats.num_indexed_rows == await some_table.count_rows()
assert stats.num_unindexed_rows == 0
assert stats.num_indices == 1
@pytest.mark.asyncio
@@ -91,6 +100,14 @@ async def test_create_vector_index(some_table: AsyncTable):
assert len(indices) == 1
assert indices[0].index_type == "IvfPq"
assert indices[0].columns == ["vector"]
assert indices[0].name == "vector_idx"
stats = await some_table.index_stats("vector_idx")
assert stats.index_type == "IVF_PQ"
assert stats.distance_type == "l2"
assert stats.num_indexed_rows == await some_table.count_rows()
assert stats.num_unindexed_rows == 0
assert stats.num_indices == 1
@pytest.mark.asyncio