pyspiral 0.6.0__cp310-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl → 0.6.2__cp310-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl

{pyspiral-0.6.0.dist-info → pyspiral-0.6.2.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: pyspiral
-Version: 0.6.0
+Version: 0.6.2
 Classifier: Intended Audience :: Science/Research
 Classifier: Operating System :: OS Independent
 Classifier: Programming Language :: Python

{pyspiral-0.6.0.dist-info → pyspiral-0.6.2.dist-info}/RECORD

@@ -1,8 +1,8 @@
-pyspiral-0.6.0.dist-info/METADATA,sha256=ETkF1eW1JfgWeoRKxgF7eDX0K9LWDOOixTO8mWW7GlA,1836
-pyspiral-0.6.0.dist-info/WHEEL,sha256=PxcKzGLVtZeSnGJDErQ-Emkn2AvBXbmzIogfnaf7-q8,130
-pyspiral-0.6.0.dist-info/entry_points.txt,sha256=uft7u-a6g40NLt4Q6BleWbK4NY0M8nZuYPpP8DV0EOk,45
-spiral/__init__.py,sha256=iAicRWWphlRNKjIS_BFTSqIweCLwJTJTbyWF0BvqMLY,667
-spiral/_lib.abi3.so,sha256=SbX-oo3QcW2lvfAkklKzK8dqE_iy2cL7uiyP_2y5m4E,54872904
+pyspiral-0.6.2.dist-info/METADATA,sha256=n0a5SuYMybj-eUwB9No9IFxSk9Cn_pDZfjpx0HuDxRw,1836
+pyspiral-0.6.2.dist-info/WHEEL,sha256=PxcKzGLVtZeSnGJDErQ-Emkn2AvBXbmzIogfnaf7-q8,130
+pyspiral-0.6.2.dist-info/entry_points.txt,sha256=uft7u-a6g40NLt4Q6BleWbK4NY0M8nZuYPpP8DV0EOk,45
+spiral/__init__.py,sha256=5c0faqg-kHZBDwriQ7LzLAMcFolIucp-IA1EzNvCZ3k,711
+spiral/_lib.abi3.so,sha256=oOsM8f904rj2leVKR8cCOkLQ7uxcf36tKVIN0bCVOok,55184120
 spiral/adbc.py,sha256=7IxfWIeQN-fh0W5OdN_PP2x3pzQYg6ZUOLsHg3jktqw,14842
 spiral/api/__init__.py,sha256=ULBlVq3PnfNOO6T5naE_ULmmii-83--qTuN2PpAUQN0,2241
 spiral/api/admin.py,sha256=A1iVR1XYJSObZivPAD5UzmPuMgupXc9kaHNYYa_kwfs,585
@@ -24,13 +24,13 @@ spiral/cli/app.py,sha256=HWCjMJLzSz_JaiLF046jzC9A4-yvzS6506D3cOR2Vgc,1773
 spiral/cli/console.py,sha256=6JHbAQV6MFWz3P-VzqPOjhHpkIQagsCdzTMvmuDKMkU,2580
 spiral/cli/fs.py,sha256=UREIJhjr6MfIdcKK7pjUKICd0wsQULhQiWRVWUnQ0dc,4376
 spiral/cli/iceberg.py,sha256=Q14tcGcn1LixbFCYP0GhfYwFFXTmmi8tqBPYwalJEyE,3248
-spiral/cli/key_spaces.py,sha256=EEgn7Zjc16CkeQO-4vWdwEqCTddTMiUAdLh4vG4AoYk,2218
+spiral/cli/key_spaces.py,sha256=x3IFRP5d47pKiAHeWExYMOBaT2TwxbWjVM01SUqKrwI,2943
 spiral/cli/login.py,sha256=TgTr37ImgG1NKN8VbtqkxVAYaZFpMXMwPAb23gVldEw,649
 spiral/cli/orgs.py,sha256=fmOuLxpeIFfKqePRi292Gv9k-EF5pPn_tbKd2BLl2Ig,2869
 spiral/cli/printer.py,sha256=HcvSUpaMItzmhBUfIHROK1Z3SL8J8wDopS3Qo8H00uw,1781
 spiral/cli/projects.py,sha256=UYrBlLcFacuXExdLX1sZByfvkz9MRtk_0oRAZvqHa0w,5105
 spiral/cli/state.py,sha256=10wTIVQ0SJkY67Z6-KQ1LFlt3aVIPmZhoHFdTwp4kNA,130
-spiral/cli/tables.py,sha256=8-9ay0mXS1Ew7DMoYFfHqC-Ro0TWsOTTinusS7M1slE,4639
+spiral/cli/tables.py,sha256=48lZ0wPQSCTul1vn-Qx6Be5eGnw75Abtw2zxMK9dCPg,4613
 spiral/cli/telemetry.py,sha256=Uxo1Q1FkKJ6n6QNGOUmL3j_pRRWRx0qWIhoP-U9BuR0,589
 spiral/cli/text.py,sha256=DlWGe4JrkdERAiqyITNpk91Wqb63Re99rNYlIFsIamc,4031
 spiral/cli/types.py,sha256=XYzo1GgX7dBBItoBSrHI4vO5C2lLmS2sktb-2GnGH3E,1362
@@ -38,15 +38,15 @@ spiral/cli/workloads.py,sha256=2_SLfQTFN6y73R9H0i9dk8VIOVagKxSxOpHXC56yptY,2015
 spiral/client.py,sha256=Po9xgCH3NwVsCeRZMm3eJUPV77Rknyj-9MfCS1TbdTg,6623
 spiral/core/__init__.pyi,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 spiral/core/authn/__init__.pyi,sha256=Jw_8ywTMDTwgAtGxMtFED63rU0jOgrv-eZtaZ5sR5t4,757
-spiral/core/client/__init__.pyi,sha256=uONPrQbKvlNjnIDLT7c0wG9GMWwNveRd6aHJu6NuQ74,5228
-spiral/core/table/__init__.pyi,sha256=uZHXdm160fNAAoz3jnFPtbZl8EFEyLwS3wo0r7jEMOo,3807
-spiral/core/table/manifests/__init__.pyi,sha256=3V59-K1qr1z2dGfgRKXaHSVheK8NNw8Q8PFhfbeQd_4,1065
+spiral/core/client/__init__.pyi,sha256=iEhZgbySG5LScfrtkiiHW1iHghgehsrVmPP-v5Pv_vk,5740
+spiral/core/table/__init__.pyi,sha256=sjjShdgM_Uh8Roou1k02MnrqYpdAX4QuyRlIRlnyp1M,3073
+spiral/core/table/manifests/__init__.pyi,sha256=eVfDpmhYSjafIvvALqAkZe5baN3Y1HpKpxYEbjwd4gQ,1043
 spiral/core/table/metastore/__init__.pyi,sha256=rc3u9MwEKRvL2kxOc8lBorddFRnM8o_o1frqtae86a4,1697
 spiral/core/table/spec/__init__.pyi,sha256=0NyGeyEhV_ebwKWVU3sqSvdF2D9v8kEVwo6wYAHF99M,5579
 spiral/dataset.py,sha256=NNqG-oOrhbmNC2OMZ9AYAm4YkwwBozeRI6zXtz4cspA,8008
 spiral/datetime_.py,sha256=1TA1RYIRU22qcUuipIjVhAtGnPDVn2z9WttuhkmfkwY,964
 spiral/debug/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-spiral/debug/manifests.py,sha256=CGC5C1HG4XCMhlkNZI-woBAIO-EQSVSqXuDUsiV-d7g,2935
+spiral/debug/manifests.py,sha256=oaPB4534pQdqvPXCZetVNSvvhpdXTrv_1pN-_bAkeAo,2893
 spiral/debug/metrics.py,sha256=XdRDcjggtsLNGCAjam6IxG9072pz_d2C8iLApNRFUtk,2044
 spiral/debug/scan.py,sha256=9bMmVQFs5M6Rldm0fmrmmvn9LbSSTKBV5tIu37mEn78,8938
 spiral/expressions/__init__.py,sha256=T8PIb0_UB9kynK0dpWbUD4No5lKRTG-wKnao8xOcXjY,6381
@@ -84,16 +84,16 @@ spiral/protogen/_/substrait/extensions/__init__.py,sha256=sCMvwWCXWu2cSGiTEH0hRj
 spiral/protogen/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 spiral/protogen/util.py,sha256=smnvVo6nYH3FfDm9jqhNLaXz4bbTBaQezHQDCTvZyiQ,1486
 spiral/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-spiral/scan.py,sha256=nItu2SNqp5-f2LPMl4EXrEUxV5tJJGEVQEPMSko3STY,7044
+spiral/scan.py,sha256=20-NSGsoXYf6uKQ7yEdbbwT8ijIK7KxKTctycsl0AIk,7073
 spiral/server.py,sha256=ztBmB5lBnUz-smQxR_tC8AI5SOhz17wH0MI3GuzDUdM,600
 spiral/settings.py,sha256=Nap68xM-1ZvF3yDhkyRnNDIAVMIgxmIksglg_1iT0-0,3069
 spiral/snapshot.py,sha256=_l2wrqUXz2RARjIDxOWw4aQpegJohvggIoWuCllzStA,1506
 spiral/streaming_/__init__.py,sha256=s7MlW2ERsuZmZGExLFL6RcZon2e0tNBocBg5ANgki7k,61
-spiral/streaming_/reader.py,sha256=CahNNeJznRuUUTtWNexoEBZtKh9bikfaI6UCnER3Jhw,3451
-spiral/streaming_/stream.py,sha256=xFTtGB6CspEKstzBeyyaOeOR3KDiJc21m07ZpD1AXZQ,5669
+spiral/streaming_/reader.py,sha256=Kpqknv2jn12jUhHOEEDArj0JZwrWb8XjoOGs9HrdVyA,4047
+spiral/streaming_/stream.py,sha256=nxJEisPfZ2-Ebkm83hz_3v8NH27FxBku-1jw7UDlQuM,5881
 spiral/substrait_.py,sha256=AKeOD4KIXvz2J4TYxnIneOiHddtBIyOhuNxVO_uH0eg,12592
-spiral/table.py,sha256=Y5_FqZGjXwt7LT_SYzUA-M-zOBcOdJs7d_t919TAc1k,9605
+spiral/table.py,sha256=ZQFq5tuovDjQcpi38b5FUMuHNGI5XV0MnZbC6vbza1o,10312
 spiral/text_index.py,sha256=FQ9rgIEGLSJryS9lFdMhKtPFey18BXoWbPXyvZPJJ04,442
-spiral/transaction.py,sha256=O3vSaTc7zpeC5qbqnj-VWKwK6rrp_mYV2JuPHp2ZJ80,1464
+spiral/transaction.py,sha256=nSykH4UGs9hGtWuSWK9YyT9jfEuvzfkKoUgMM5Xt4zU,1841
 spiral/types_.py,sha256=W_jyO7F6rpPiH69jhgSgV7OxQZbOlb1Ho3InpKUP6Eo,155
-pyspiral-0.6.0.dist-info/RECORD,,
+pyspiral-0.6.2.dist-info/RECORD,,

spiral/__init__.py

@@ -7,9 +7,10 @@ from spiral import _lib
 assert _lib, "Spiral library"
 
 from spiral.client import Spiral  # noqa: E402
+from spiral.core.client import ShuffleStrategy  # noqa: E402
 from spiral.key_space_index import KeySpaceIndex  # noqa: E402
 from spiral.project import Project  # noqa: E402
-from spiral.scan import Scan, ShuffleStrategy  # noqa: E402
+from spiral.scan import Scan  # noqa: E402
 from spiral.snapshot import Snapshot  # noqa: E402
 from spiral.table import Table  # noqa: E402
 from spiral.text_index import TextIndex  # noqa: E402

spiral/cli/key_spaces.py

@@ -56,6 +56,27 @@ def ls(
     CONSOLE.print(rich_table)
 
 
+@app.command(help="Show index partitions.")
+def show(
+    project: ProjectArg,
+    name: Annotated[str | None, Option(help="Index name.")] = None,
+):
+    """Show index partitions."""
+    index_id = get_index_id(project, name)
+    index = state.spiral.key_space_index(index_id)
+    shards = state.spiral._ops().compute_shards(index.core)
+
+    rich_table = rich.table.Table("Begin", "End", "Cardinality", title=f"Index {index.name} Partitions")
+    for partition in shards:
+        rich_table.add_row(
+            # TODO(marko): This isn't really pretty...
+            repr(partition.key_range.begin),
+            repr(partition.key_range.end),
+            str(partition.cardinality),
+        )
+    CONSOLE.print(rich_table)
+
+
 @app.command(help="Trigger a sync job for an index.")
 def sync(
     project: ProjectArg,

spiral/cli/tables.py

@@ -111,12 +111,12 @@ def manifests(
     _, t = get_table(project, table, dataset)
     s = t.snapshot()
 
-    key_space_state = state.spiral._ops().key_space_state(t.core, asof=s.asof)  # pyright: ignore[reportPrivateUsage]
+    key_space_state = state.spiral._ops().key_space_state(s.core)  # pyright: ignore[reportPrivateUsage]
     key_space_manifest = key_space_state.manifest
 
-    column_groups_states = state.spiral._ops().column_groups_states(t.core, key_space_state, asof=s.asof)  # pyright: ignore[reportPrivateUsage]
+    column_groups_states = state.spiral._ops().column_groups_states(s.core, key_space_state)  # pyright: ignore[reportPrivateUsage]
 
-    display_manifests(key_space_manifest, ((x.column_group, x.manifest) for x in column_groups_states))
+    display_manifests(key_space_manifest, [(x.column_group, x.manifest) for x in column_groups_states])
 
 
 @app.command(help="Display the manifests which would be read by a scan of the given column group.")

spiral/core/client/__init__.pyi

@@ -124,6 +124,25 @@ class Shard:
     key_range: KeyRange
     cardinality: int
 
+    def __init__(self, key_range: KeyRange, cardinality: int): ...
+
+class ShuffleStrategy:
+    # Results are buffered in a pool of `shuffle_buffer_size` rows and shuffled again.
+    shuffle_buffer_size: int
+
+    # All randomness is derived from this seed. If None, a random seed is generated from the OS.
+    seed: int | None
+
+    # Externally provided shards to shuffle before reading rows.
+    shards: list[Shard] | None
+
+    def __init__(
+        self,
+        seed: int | None = None,
+        shuffle_buffer_size: int | None = None,
+        shards: list[Shard] | None = None,
+    ): ...
+
 class Operations:
     def flush_wal(self, table: Table) -> None:
         """
@@ -163,21 +182,19 @@ class Operations:
         Index table changes up to the given snapshot.
         """
         ...
-    def key_space_state(self, table: Table, *, asof: int | None = None) -> KeySpaceState:
+    def key_space_state(self, snapshot: Snapshot) -> KeySpaceState:
         """
         The key space state for the table.
         """
         ...
     def column_group_state(
-        self, table: Table, key_space_state: KeySpaceState, column_group: ColumnGroup, *, asof: int | None = None
+        self, snapshot: Snapshot, key_space_state: KeySpaceState, column_group: ColumnGroup
     ) -> ColumnGroupState:
         """
         The state the column group of the table.
         """
         ...
-    def column_groups_states(
-        self, table: Table, key_space_state: KeySpaceState, *, asof: int | None = None
-    ) -> list[ColumnGroupState]:
+    def column_groups_states(self, snapshot: Snapshot, key_space_state: KeySpaceState) -> list[ColumnGroupState]:
         """
         The state of each column group of the table.
         """

spiral/core/table/__init__.pyi

@@ -1,6 +1,7 @@
 from typing import Any
 
 import pyarrow as pa
+from spiral.core.client import ShuffleStrategy
 from spiral.expressions import Expr
 
 from .manifests import FragmentManifest
@@ -76,6 +77,7 @@ class Scan:
         output_path: str,
         key_range: KeyRange,
         expected_cardinality: int | None = None,
+        shard_row_block_size: int = 8192,
     ) -> None: ...
 
 class KeySpaceState:
@@ -93,27 +95,7 @@ class Transaction:
     status: str
 
     def write(self, expr: Expr, *, partition_size_bytes: int | None = None): ...
+    def drop_columns(self, column_paths: list[str]): ...
     def commit(self): ...
     def abort(self): ...
     def metrics(self) -> dict[str, Any]: ...
-
-class ShuffleStrategy:
-    # Results are buffered in a pool of `buffer_size` rows and shuffled again.
-    shuffle_buffer_size: int
-
-    # All randomness is derived from this seed. If None, a random seed is generated from the OS.
-    seed: int | None
-
-    # `approximate_batch_size` controls the maximum approximate size of each shard. Shards that
-    # are larger than this size are further split assuming uniform distribution of keys. Note
-    # that this is a best-effort and can be widely off. The purpose of this is to improve
-    # shuffling, rather than to support sharding. If not present, splits derived from the table
-    # are used in the attempt to minimize wasted reads.
-    approximate_buffer_size: int | None
-
-    def __init__(
-        self,
-        seed: int | None = None,
-        shard_size: int | None = None,
-        buffer_size: int | None = None,
-    ): ...

spiral/core/table/manifests/__init__.pyi

@@ -8,7 +8,7 @@ class FragmentManifest:
     def __getitem__(self, idx: int): ...
     def to_arrow(self) -> pa.RecordBatchReader: ...
     @staticmethod
-    def compute_schema(data_schema: pa.Schema) -> pa.Schema: ...
+    def compute_schema() -> pa.Schema: ...
     @staticmethod
     def from_fragment(fragment_file: FragmentFile) -> FragmentManifest: ...
     @staticmethod

spiral/debug/manifests.py

@@ -1,5 +1,3 @@
-from collections.abc import Iterable
-
 from spiral import datetime_
 from spiral.core.table import Scan
 from spiral.core.table.manifests import FragmentManifest
@@ -13,15 +11,15 @@ def display_scan_manifests(scan: Scan):
         raise NotImplementedError("Multiple table scans are not supported.")
     table_id = scan.table_ids()[0]
     key_space_manifest = scan.key_space_state(table_id).manifest
-    column_group_manifests = (
+    column_group_manifests = [
         (column_group, scan.column_group_state(column_group).manifest) for column_group in scan.column_groups()
-    )
+    ]
 
     display_manifests(key_space_manifest, column_group_manifests)
 
 
 def display_manifests(
-    key_space_manifest: FragmentManifest, column_group_manifests: Iterable[tuple[ColumnGroup, FragmentManifest]]
+    key_space_manifest: FragmentManifest, column_group_manifests: list[tuple[ColumnGroup, FragmentManifest]]
 ):
     _table_of_fragments(
         key_space_manifest,

spiral/scan.py

@@ -3,7 +3,8 @@ from typing import TYPE_CHECKING, Any
 
 import pyarrow as pa
 
-from spiral.core.table import KeyRange, ShuffleStrategy
+from spiral.core.client import ShuffleStrategy
+from spiral.core.table import KeyRange
 from spiral.core.table import Scan as CoreScan
 from spiral.core.table.spec import Schema
 from spiral.settings import CI, DEV
@@ -177,7 +178,7 @@ class Scan:
         # Print manifests in a human-readable format.
         from spiral.debug.manifests import display_scan_manifests
 
-        display_scan_manifests(self._core)
+        display_scan_manifests(self.core)
 
     def _dump_metrics(self):
         # Print metrics in a human-readable format.

spiral/streaming_/reader.py

@@ -70,6 +70,20 @@ class SpiralReader:
         """
         return [(FileInfo(basename=self.filename), None)]
 
+    def get_max_size(self) -> int:
+        """Get the full size of this shard.
+
+        "Max" in this case means both the raw (decompressed) and zip (compressed) versions are
+        resident (assuming it has a zip form). This is the maximum disk usage the shard can reach.
+        When compressed was used, even if keep_zip is ``False``, the zip form must still be
+        resident at the same time as the raw form during shard decompression.
+
+        Returns:
+            int: Size in bytes.
+        """
+        # TODO(marko): This is used to check cache limit is possible...
+        return 0
+
     @functools.cached_property
     def filename(self) -> str:
         """Used by SpiralStream to identify shard's file-on-disk, if it exists."""

spiral/streaming_/stream.py

@@ -24,10 +24,13 @@ class SpiralStream:
     Stream can be passed to MDS's StreamingDataset in `streams` argument.
     """
 
-    def __init__(self, scan: CoreScan, shards: list[Shard], cache_dir: str | None = None):
+    def __init__(
+        self, scan: CoreScan, shards: list[Shard], cache_dir: str | None = None, shard_row_block_size: int = 8192
+    ):
         self._scan = scan
         # TODO(marko): Read shards only on world.is_local_leader in `get_shards` and materialize on disk.
         self._shards = shards
+        self.shard_row_block_size = shard_row_block_size
 
         if cache_dir is not None:
             if not os.path.exists(cache_dir):
@@ -92,7 +95,12 @@ class SpiralStream:
             return 0
 
         # This method exists but it's hidden.
-        self._scan._prepare_shard(shard_path, shard.shard.key_range, expected_cardinality=shard.shard.cardinality)
+        self._scan._prepare_shard(
+            shard_path,
+            shard.shard.key_range,
+            expected_cardinality=shard.shard.cardinality,
+            shard_row_block_size=self.shard_row_block_size,
+        )
 
         # Get the size of the file on disk.
         stat = os.stat(shard_path)

spiral/table.py

@@ -109,6 +109,17 @@ class Table(Expr):
                 partition_size_bytes=partition_size_bytes,
             )
 
+    def drop_columns(self, column_paths: list[str]) -> None:
+        """
+        Drops the specified columns from the table.
+
+
+        :param column_paths: Fully qualified column names. (e.g., "column_name" or "nested.field").
+            All columns must exist, if a a column doesn't exist the function will return an error.
+        """
+        with self.txn() as txn:
+            txn.drop_columns(column_paths)
+
     def snapshot(self, asof: datetime | int | None = None) -> Snapshot:
         """Returns a snapshot of the table at the given timestamp."""
         if isinstance(asof, datetime):
@@ -183,7 +194,9 @@ class Table(Expr):
         *,
         index: "KeySpaceIndex",
         batch_size: int | None = None,
+        cache_dir: str | None = None,
         cache_limit: int | str | None = None,
+        predownload: int | None = None,
         sampling_method: str = "balanced",
         sampling_granularity: int = 1,
         partition_algo: str = "relaxed",
@@ -208,12 +221,13 @@ class Table(Expr):
         """
         from streaming import StreamingDataset
 
-        stream = self.to_streaming(index=index)
+        stream = self.to_streaming(index=index, cache_dir=cache_dir)
 
         return StreamingDataset(
             streams=[stream],
             batch_size=batch_size,
             cache_limit=cache_limit,
+            predownload=predownload,
             sampling_method=sampling_method,
             sampling_granularity=sampling_granularity,
             partition_algo=partition_algo,
@@ -226,13 +240,14 @@ class Table(Expr):
             replication=replication,
         )
 
-    def to_streaming(self, index: "KeySpaceIndex") -> "streaming.Stream":
+    def to_streaming(self, index: "KeySpaceIndex", *, cache_dir: str | None = None) -> "streaming.Stream":
         """Returns a stream to be used with MosaicML's StreamingDataset.
 
         Requires `streaming` package to be installed.
 
         Args:
             index: Prebuilt KeysIndex to use when creating the stream. The index's `asof` will be used when scanning.
+            cache_dir: Directory to use for caching data. If None, a temporary directory will be used.
         """
         from spiral.streaming_ import SpiralStream
 
@@ -254,4 +269,4 @@ class Table(Expr):
         #   We have a world there and can compute shards only on leader.
         shards = self.spiral._core._ops().compute_shards(index=index.core)
 
-        return SpiralStream(scan=scan.core, shards=shards)  # type: ignore[return-value]
+        return SpiralStream(scan=scan.core, shards=shards, cache_dir=cache_dir)  # type: ignore[return-value]

spiral/transaction.py

@@ -39,6 +39,16 @@ class Transaction:
 
         self._core.write(expr.__expr__, partition_size_bytes=partition_size_bytes)
 
+    def drop_columns(self, column_paths: list[str]):
+        """
+        Drops the specified columns from the table.
+
+
+        :param column_paths: Fully qualified column names. (e.g., "column_name" or "nested.field").
+            All columns must exist, if a a column doesn't exist the function will return an error.
+        """
+        self._core.drop_columns(column_paths)
+
     def commit(self):
         """Commit the transaction."""
         self._core.commit()