vastdb 2.0.1__tar.gz → 2.0.2__tar.gz

{vastdb-2.0.1 → vastdb-2.0.2}/CHANGELOG.md

@@ -4,13 +4,18 @@ All notable changes to this project will be documented in this file.
 
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/).
 
+## [2.0.2] (2025-10-22)
+[2.0.2]: https://github.com/vast-data/vastdb_sdk/compare/v2.0.1...v2.0.2
+
+### Fixed
+- Get correct rows count and size in bytes when asking for a table projections
+
 ## [2.0.1] (2025-10-22)
 [2.0.1]: https://github.com/vast-data/vastdb_sdk/compare/v2.0.0...v2.0.1
 
 ### Fixed
 - Performance improvement in query data flow - sending only required columns instead of entire schema
 
-
 ## [2.0.0] (2025-09-17)
 [2.0.0]: https://github.com/vast-data/vastdb_sdk/compare/v1.4.0...v2.0.0
 

{vastdb-2.0.1 → vastdb-2.0.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: vastdb
-Version: 2.0.1
+Version: 2.0.2
 Summary: VAST Data SDK
 Home-page: https://github.com/vast-data/vastdb_sdk
 Author: VAST DATA

{vastdb-2.0.1 → vastdb-2.0.2}/setup.py

@@ -29,7 +29,7 @@ setup(
     name='vastdb',
     python_requires='>=3.10.0',
     description='VAST Data SDK',
-    version='2.0.1' + suffix,
+    version='2.0.2' + suffix,
     url='https://github.com/vast-data/vastdb_sdk',
     author='VAST DATA',
     author_email='hello@vastdata.com',

{vastdb-2.0.1 → vastdb-2.0.2}/vastdb/table.py

@@ -231,11 +231,11 @@ class TableInTransaction(ITable):
             raise errors.NotSupportedCommand(
                 self.ref.bucket, self.ref.schema, self.ref.table)
 
-    def projection(self, name: str) -> "Projection":
+    def projection(self, name: str, include_stats: bool = True) -> "Projection":
         """Get a specific semi-sorted projection of this table."""
         self._assert_not_imports_table()
 
-        projs = tuple(self.projections(projection_name=name))
+        projs = tuple(self.projections(projection_name=name, include_stats=include_stats))
         if not projs:
             raise errors.MissingProjection(
                 self.ref.bucket, self.ref.schema, self.ref.table, name)
@@ -245,7 +245,7 @@ class TableInTransaction(ITable):
         log.debug("Found projection: %s", projs[0])
         return projs[0]
 
-    def projections(self, projection_name: str = "") -> Iterable["Projection"]:
+    def projections(self, projection_name: str = "", include_stats: bool = True) -> Iterable["Projection"]:
         """List all semi-sorted projections of this table if `projection_name` is empty.
 
         Otherwise, list only the specific projection (if exists).
@@ -260,7 +260,7 @@ class TableInTransaction(ITable):
             _bucket_name, _schema_name, _table_name, curr_projections, next_key, is_truncated, _ = \
                 self._tx._rpc.api.list_projections(
                     bucket=self.ref.bucket, schema=self.ref.schema, table=self.ref.table, next_key=next_key, txid=self._tx.active_txid,
-                    exact_match=exact_match, name_prefix=name_prefix)
+                    exact_match=exact_match, name_prefix=name_prefix, include_list_stats=include_stats)
             if not curr_projections:
                 break
             projections.extend(curr_projections)

{vastdb-2.0.1 → vastdb-2.0.2}/vastdb/tests/test_projections.py

@@ -8,6 +8,9 @@ from vastdb.table import QueryConfig
 log = logging.getLogger(__name__)
 
 
+DELAY_TO_LET_SERVER_UPDATE: int = 3
+
+
 def test_basic_projections(session, clean_bucket_name):
     with session.transaction() as tx:
         s = tx.bucket(clean_bucket_name).create_schema('s1')
@@ -94,7 +97,7 @@ def test_query_data_with_projection(session, clean_bucket_name):
         actual = pa.Table.from_batches(t.select(columns=['a', 'b', 's']))
         assert actual == expected
 
-    time.sleep(3)
+    time.sleep(DELAY_TO_LET_SERVER_UPDATE)
 
     with session.transaction() as tx:
         config = QueryConfig()
@@ -123,3 +126,48 @@ def test_query_data_with_projection(session, clean_bucket_name):
 
         t.drop()
         s.drop()
+
+
+def test_projection_stats(session, clean_bucket_name):
+    columns = pa.schema([
+        ('a', pa.int64()),
+        ('b', pa.int64()),
+    ])
+
+    # min size to be considered as a projection
+    GROUP_SIZE = 64 * 1024
+    expected = pa.table(schema=columns, data=[
+        [i for i in range(GROUP_SIZE)],
+        [i for i in reversed(range(GROUP_SIZE))],
+    ])
+
+    schema_name = "schema"
+    table_name = "table"
+    with session.transaction() as tx:
+        s = tx.bucket(clean_bucket_name).create_schema(schema_name)
+        t = s.create_table(table_name, expected.schema)
+
+        sorted_columns = ['b']
+        unsorted_columns = ['a']
+        t.create_projection('p1', sorted_columns, unsorted_columns)
+
+    with session.transaction() as tx:
+        s = tx.bucket(clean_bucket_name).schema(schema_name)
+        t = s.table(table_name)
+        t.insert(expected)
+        actual = pa.Table.from_batches(t.select(columns=['a', 'b']))
+        assert actual == expected
+
+    time.sleep(DELAY_TO_LET_SERVER_UPDATE)
+
+    with session.transaction() as tx:
+        s = tx.bucket(clean_bucket_name).schema(schema_name)
+        t = s.table(table_name)
+        projections = t.projections()
+        assert len(projections) == 1
+        stats = projections[0].stats
+        assert stats.num_rows == GROUP_SIZE
+        assert stats.size_in_bytes > 0
+
+        t.drop()
+        s.drop()

{vastdb-2.0.1 → vastdb-2.0.2}/vastdb.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: vastdb
-Version: 2.0.1
+Version: 2.0.2
 Summary: VAST Data SDK
 Home-page: https://github.com/vast-data/vastdb_sdk
 Author: VAST DATA