datachain 0.24.3__py3-none-any.whl → 0.24.5__py3-none-any.whl

datachain/catalog/catalog.py

@@ -1120,6 +1120,14 @@ class Catalog:
         pull_dataset: bool = False,
         update: bool = False,
     ) -> DatasetRecord:
+        # Intentionally ignore update flag is version is provided. Here only exact
+        # version can be provided and update then doesn't make sense.
+        # It corresponds to a query like this for example:
+        #
+        #    dc.read_dataset("some.remote.dataset", version="1.0.0", update=True)
+        if version:
+            update = False
+
         if self.metastore.is_local_dataset(namespace_name) or not update:
             try:
                 project = self.metastore.get_project(project_name, namespace_name)

datachain/dataset.py

@@ -32,7 +32,7 @@ QUERY_DATASET_PREFIX = "ds_query_"
 LISTING_PREFIX = "lst__"
 
 DEFAULT_DATASET_VERSION = "1.0.0"
-DATASET_NAME_RESERVED_CHARS = ["."]
+DATASET_NAME_RESERVED_CHARS = [".", "@"]
 DATASET_NAME_REPLACEMENT_CHAR = "_"
 
 

datachain/lib/dc/datachain.py

@@ -21,6 +21,7 @@ from typing import (
 import orjson
 import sqlalchemy
 from pydantic import BaseModel
+from sqlalchemy.sql.elements import ColumnElement
 from tqdm import tqdm
 
 from datachain import semver
@@ -806,11 +807,35 @@ class DataChain:
             chain.save("new_dataset")
             ```
         """
+        # Convert string partition_by parameters to Column objects
+        processed_partition_by = partition_by
+        if partition_by is not None:
+            if isinstance(partition_by, (str, Function, ColumnElement)):
+                list_partition_by = [partition_by]
+            else:
+                list_partition_by = list(partition_by)
+
+            processed_partition_columns: list[ColumnElement] = []
+            for col in list_partition_by:
+                if isinstance(col, str):
+                    col_db_name = ColumnMeta.to_db_name(col)
+                    col_type = self.signals_schema.get_column_type(col_db_name)
+                    column = Column(col_db_name, python_to_sql(col_type))
+                    processed_partition_columns.append(column)
+                elif isinstance(col, Function):
+                    column = col.get_column(self.signals_schema)
+                    processed_partition_columns.append(column)
+                else:
+                    # Assume it's already a ColumnElement
+                    processed_partition_columns.append(col)
+
+            processed_partition_by = processed_partition_columns
+
         udf_obj = self._udf_to_obj(Aggregator, func, params, output, signal_map)
         return self._evolve(
             query=self._query.generate(
                 udf_obj.to_udf_wrapper(),
-                partition_by=partition_by,
+                partition_by=processed_partition_by,
                 **self._settings.to_dict(),
             ),
             signal_schema=udf_obj.output,

datachain/lib/dc/datasets.py

@@ -189,6 +189,7 @@ def read_dataset(
         namespace_name=namespace_name,
         version=version,  #  type: ignore[arg-type]
         session=session,
+        update=update,
     )
 
     signals_schema = SignalSchema({"sys": Sys})

datachain/namespace.py

@@ -6,7 +6,7 @@ from typing import Any, Optional, TypeVar
 from datachain.error import InvalidNamespaceNameError
 
 N = TypeVar("N", bound="Namespace")
-NAMESPACE_NAME_RESERVED_CHARS = ["."]
+NAMESPACE_NAME_RESERVED_CHARS = [".", "@"]
 
 
 @dataclass(frozen=True)

datachain/project.py

@@ -7,7 +7,7 @@ from datachain.error import InvalidProjectNameError
 from datachain.namespace import Namespace
 
 P = TypeVar("P", bound="Project")
-PROJECT_NAME_RESERVED_CHARS = ["."]
+PROJECT_NAME_RESERVED_CHARS = [".", "@"]
 
 
 @dataclass(frozen=True)

datachain/query/dataset.py

@@ -82,7 +82,10 @@ if TYPE_CHECKING:
 INSERT_BATCH_SIZE = 10000
 
 PartitionByType = Union[
-    Function, ColumnElement, Sequence[Union[Function, ColumnElement]]
+    str,
+    Function,
+    ColumnElement,
+    Sequence[Union[str, Function, ColumnElement]],
 ]
 JoinPredicateType = Union[str, ColumnClause, ColumnElement]
 DatasetDependencyType = tuple["DatasetRecord", str]
@@ -1142,6 +1145,7 @@ class DatasetQuery:
                     project_name=project_name,
                     version=version,
                     pull_dataset=True,
+                    update=update,
                 )
             )
 

{datachain-0.24.3.dist-info → datachain-0.24.5.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: datachain
-Version: 0.24.3
+Version: 0.24.5
 Summary: Wrangle unstructured AI data at scale
 Author-email: Dmitry Petrov <support@dvc.org>
 License-Expression: Apache-2.0

{datachain-0.24.3.dist-info → datachain-0.24.5.dist-info}/RECORD

@@ -3,17 +3,17 @@ datachain/__main__.py,sha256=hG3Y4ARGEqe1AWwNMd259rBlqtphx1Wk39YbueQ0yV8,91
 datachain/asyn.py,sha256=RH_jFwJcTXxhEFomaI9yL6S3Onau6NZ6FSKfKFGtrJE,9689
 datachain/cache.py,sha256=ESVRaCJXEThMIfGEFVHx6wJPOZA7FYk9V6WxjyuqUBY,3626
 datachain/config.py,sha256=g8qbNV0vW2VEKpX-dGZ9pAn0DAz6G2ZFcr7SAV3PoSM,4272
-datachain/dataset.py,sha256=wDrukmkDnYP0X8bAGY-7O1NDE3DWCFqrH8VVDpXM9Ok,25263
+datachain/dataset.py,sha256=ATGa-CBTFoZeTN2V40-zHEzfMBcdYK0WuoJ6H2yEAvo,25268
 datachain/delta.py,sha256=fTEhCedseUsHuH_Ek52NXFhFPyFD_6MioEH5sCilNgo,9897
 datachain/error.py,sha256=OWwWMkzZYJrkcoEDGhJHMf7SfKvxcsOLRF94mjPf29I,1609
 datachain/job.py,sha256=x5PB6d5sqx00hePNNkirESlOVAvnmkEM5ygUgQmAhsk,1262
 datachain/listing.py,sha256=GuUlj3oYJQOIbLYr3IqNCSeuiU5pfOtRAQQl9uBEFQU,7458
-datachain/namespace.py,sha256=r7YwpXFc7LdMS2jhyLMTrAHffALFG2bAIiYUSuZa5Rc,1786
+datachain/namespace.py,sha256=MozcXYxedIbamzY56YKy9r9fgSpOm2VryhWfIf6stYk,1791
 datachain/node.py,sha256=KWDT0ClYXB7FYI-QOvzAa-UDkLJErUI2eWm5FBteYuU,5577
 datachain/nodes_fetcher.py,sha256=_wgaKyqEjkqdwJ_Hj6D8vUYz7hnU7g6xhm0H6ZnYxmE,1095
 datachain/nodes_thread_pool.py,sha256=mdo0s-VybuSZkRUARcUO4Tjh8KFfZr9foHqmupx2SmM,3989
 datachain/progress.py,sha256=lRzxoYP4Qv2XBwD78sOkmYRzHFpZ2ExVNJF8wAeICtY,770
-datachain/project.py,sha256=kX5urIAnkHcZJ8m_IIzrUtHEytMBeiceVTrqgCXc_4E,2275
+datachain/project.py,sha256=90D4GpJSA3t0fayYZbzrL3sk4U7EJhQo8psnWvdI7_o,2280
 datachain/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 datachain/script_meta.py,sha256=V-LaFOZG84pD0Zc0NvejYdzwDgzITv6yHvAHggDCnuY,4978
 datachain/semver.py,sha256=UB8GHPBtAP3UJGeiuJoInD7SK-DnB93_Xd1qy_CQ9cU,2074
@@ -21,7 +21,7 @@ datachain/studio.py,sha256=bLok-eJNFRHQScEyAyA_Fas52dmijd5r-73KudWxV4k,13337
 datachain/telemetry.py,sha256=0A4IOPPp9VlP5pyW9eBfaTK3YhHGzHl7dQudQjUAx9A,994
 datachain/utils.py,sha256=DNqOi-Ydb7InyWvD9m7_yailxz6-YGpZzh00biQaHNo,15305
 datachain/catalog/__init__.py,sha256=cMZzSz3VoUi-6qXSVaHYN-agxQuAcz2XSqnEPZ55crE,353
-datachain/catalog/catalog.py,sha256=7N4QmG6mpj8WaWcJSp7LijFQlVHbz_B6DLUnMKXmx6k,65644
+datachain/catalog/catalog.py,sha256=3QwiljkEHWm5xNmvqT3ey_JvKS8viiJgGWwhbqWIH0M,65996
 datachain/catalog/datasource.py,sha256=IkGMh0Ttg6Q-9DWfU_H05WUnZepbGa28HYleECi6K7I,1353
 datachain/catalog/loader.py,sha256=UXjYD6BNRoupPvkiz3-b04jepXhtLHCA4gzKFnXxOtQ,5987
 datachain/cli/__init__.py,sha256=WvBqnwjG8Wp9xGCn-4eqfoZ3n7Sj1HJemCi4MayJh_c,8221
@@ -103,8 +103,8 @@ datachain/lib/convert/values_to_tuples.py,sha256=j5yZMrVUH6W7b-7yUvdCTGI7JCUAYUO
 datachain/lib/dc/__init__.py,sha256=HD0NYrdy44u6kkpvgGjJcvGz-UGTHui2azghcT8ZUg0,838
 datachain/lib/dc/csv.py,sha256=q6a9BpapGwP6nwy6c5cklxQumep2fUp9l2LAjtTJr6s,4411
 datachain/lib/dc/database.py,sha256=g5M6NjYR1T0vKte-abV-3Ejnm-HqxTIMir5cRi_SziE,6051
-datachain/lib/dc/datachain.py,sha256=dOPtNOYx6ocFr61YHTDrKGoMTDWDY0AZt8MLh79EJkc,85876
-datachain/lib/dc/datasets.py,sha256=U4xqAfs6FdW8HIJjeayQaIg1dunaIsVXYGqfq_sDSv0,13274
+datachain/lib/dc/datachain.py,sha256=_FJnpgNN_b2xz39MsgeS0NTto0hzpcFPbJlaUBLcqTs,87094
+datachain/lib/dc/datasets.py,sha256=MzM7MTn90Q-dZYuMNUzJXRW1YHOpHeFHspadRjQoI70,13297
 datachain/lib/dc/hf.py,sha256=PJl2wiLjdRsMz0SYbLT-6H8b-D5i2WjeH7li8HHOk_0,2145
 datachain/lib/dc/json.py,sha256=dNijfJ-H92vU3soyR7X1IiDrWhm6yZIGG3bSnZkPdAE,2733
 datachain/lib/dc/listings.py,sha256=V379Cb-7ZyquM0w7sWArQZkzInZy4GB7QQ1ZfowKzQY,4544
@@ -125,7 +125,7 @@ datachain/model/ultralytics/pose.py,sha256=pBlmt63Qe68FKmexHimUGlNbNOoOlMHXG4fzX
 datachain/model/ultralytics/segment.py,sha256=63bDCj43E6iZ0hFI5J6uQfksdCmjEp6sEm1XzVaE8pw,2986
 datachain/query/__init__.py,sha256=7DhEIjAA8uZJfejruAVMZVcGFmvUpffuZJwgRqNwe-c,263
 datachain/query/batch.py,sha256=-goxLpE0EUvaDHu66rstj53UnfHpYfBUGux8GSpJ93k,4306
-datachain/query/dataset.py,sha256=mKee4PkQHYPT96utPjM1DocURU4TghAR7AHtYkzdqwY,61292
+datachain/query/dataset.py,sha256=t9EWZkJGPRPcBvKOsFO7ZiaTeUXc8YuTZydRbcv83_E,61350
 datachain/query/dispatch.py,sha256=A0nPxn6mEN5d9dDo6S8m16Ji_9IvJLXrgF2kqXdi4fs,15546
 datachain/query/metrics.py,sha256=DOK5HdNVaRugYPjl8qnBONvTkwjMloLqAr7Mi3TjCO0,858
 datachain/query/params.py,sha256=O_j89mjYRLOwWNhYZl-z7mi-rkdP7WyFmaDufsdTryE,863
@@ -157,9 +157,9 @@ datachain/sql/sqlite/vector.py,sha256=ncW4eu2FlJhrP_CIpsvtkUabZlQdl2D5Lgwy_cbfqR
 datachain/toolkit/__init__.py,sha256=eQ58Q5Yf_Fgv1ZG0IO5dpB4jmP90rk8YxUWmPc1M2Bo,68
 datachain/toolkit/split.py,sha256=ktGWzY4kyzjWyR86dhvzw-Zhl0lVk_LOX3NciTac6qo,2914
 datachain/torch/__init__.py,sha256=gIS74PoEPy4TB3X6vx9nLO0Y3sLJzsA8ckn8pRWihJM,579
-datachain-0.24.3.dist-info/licenses/LICENSE,sha256=8DnqK5yoPI_E50bEg_zsHKZHY2HqPy4rYN338BHQaRA,11344
-datachain-0.24.3.dist-info/METADATA,sha256=qlqP_kAxEtkEJvV-ZzEPikT2OwW9S7Yqx4ly-iIOMBY,13281
-datachain-0.24.3.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-datachain-0.24.3.dist-info/entry_points.txt,sha256=0GMJS6B_KWq0m3VT98vQI2YZodAMkn4uReZ_okga9R4,49
-datachain-0.24.3.dist-info/top_level.txt,sha256=lZPpdU_2jJABLNIg2kvEOBi8PtsYikbN1OdMLHk8bTg,10
-datachain-0.24.3.dist-info/RECORD,,
+datachain-0.24.5.dist-info/licenses/LICENSE,sha256=8DnqK5yoPI_E50bEg_zsHKZHY2HqPy4rYN338BHQaRA,11344
+datachain-0.24.5.dist-info/METADATA,sha256=Z2iA5yN6lZ_rNmi3YqSOj-j5jvJ-HCIpy8sVQvb3vR4,13281
+datachain-0.24.5.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+datachain-0.24.5.dist-info/entry_points.txt,sha256=0GMJS6B_KWq0m3VT98vQI2YZodAMkn4uReZ_okga9R4,49
+datachain-0.24.5.dist-info/top_level.txt,sha256=lZPpdU_2jJABLNIg2kvEOBi8PtsYikbN1OdMLHk8bTg,10
+datachain-0.24.5.dist-info/RECORD,,