datachain 0.22.0__tar.gz → 0.24.0__tar.gz

{datachain-0.22.0 → datachain-0.24.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: datachain
-Version: 0.22.0
+Version: 0.24.0
 Summary: Wrangle unstructured AI data at scale
 Author-email: Dmitry Petrov <support@dvc.org>
 License-Expression: Apache-2.0

{datachain-0.22.0 → datachain-0.24.0}/docs/guide/env.md

@@ -15,4 +15,8 @@ List of environment variables used to configure DataChain behavior.
 - `DATACHAIN_STUDIO_TOKEN` – Authentication token for Studio.
 - `DATACHAIN_STUDIO_TEAM` – Studio team name.
 
+### Namespaces and projects
+- `DATACHAIN_NAMESPACE` – Namespace name to use as default.
+- `DATACHAIN_PROJECT` – Project name or combination of namespace name and project name separated by `.` to use as default, example: `DATACHAIN_PROJECT=dev.analytics`
+
 Note: Some environment variables are used internally and may not be documented here. For the most up-to-date list, refer to the source code.

{datachain-0.22.0 → datachain-0.24.0}/docs/guide/namespaces.md

@@ -82,6 +82,49 @@ This is equivalent to saving to `dev.analytics.metrics`.
 
 In CLI, `.settings()` is only supported when both `namespace` and `project` are set to `"local"`.
 
+## Setting Namespace and Project via Environment Variables
+
+In addition to using `.settings()`, you can configure the namespace and project using environment variables:
+
+- `DATACHAIN_NAMESPACE` sets the namespace.
+- `DATACHAIN_PROJECT` sets the project name, or both the namespace and project using the format `namespace.project`.
+
+### Examples
+
+```
+# Set namespace only
+export DATACHAIN_NAMESPACE=dev
+
+# Set project only
+export DATACHAIN_PROJECT=analytics
+
+# Set both namespace and project
+export DATACHAIN_PROJECT=dev.analytics
+```
+
+##  How Namespace and Project Are Resolved
+
+When determining which namespace and project to use, Datachain applies the following precedence:
+
+1. **Fully qualified dataset name**
+   If the dataset name includes both the namespace and project, these values take highest precedence.
+   ```python
+   dc.read_dataset("dev.analytics.metrics")
+
+2. **Explicit settings in code**
+   Values provided via `.settings()` or passed directly to `read_dataset()` or similar methods.
+   ```python
+   dc.settings(namespace="dev", project="analytics")
+   dc.read_dataset("metrics", namespace="dev", project="analytics")
+   ```
+3. **Environment variables**
+   Namespace and project set using environment variables:
+   ```console
+   export DATACHAIN_PROJECT=dev.analytics
+   ```
+4. **Defaults**
+If none of the above are provided, Datachain falls back to the default namespace and project.
+
 ## Reading a Dataset from a Project
 
 To read a dataset from a specific namespace and project:
@@ -116,4 +159,3 @@ dc.read_values(scores=[0.8, 1.5, 2.1]).save("metrics")
 
 ds = dc.read_dataset("local.local.metrics")
 ds.show()
-```

{datachain-0.22.0 → datachain-0.24.0}/src/datachain/catalog/catalog.py

@@ -49,6 +49,7 @@ from datachain.error import (
     DatasetInvalidVersionError,
     DatasetNotFoundError,
     DatasetVersionNotFoundError,
+    NamespaceNotFoundError,
     ProjectNotFoundError,
     QueryScriptCancelError,
     QueryScriptRunError,
@@ -1059,6 +1060,39 @@ class Catalog:
 
         return self.get_dataset(name, project)
 
+    def get_full_dataset_name(
+        self,
+        name: str,
+        project_name: Optional[str] = None,
+        namespace_name: Optional[str] = None,
+    ) -> tuple[str, str, str]:
+        """
+        Returns dataset name together with separated namespace and project name.
+        It takes into account all the ways namespace and project can be added.
+        """
+        parsed_namespace_name, parsed_project_name, name = parse_dataset_name(name)
+
+        namespace_env = os.environ.get("DATACHAIN_NAMESPACE")
+        project_env = os.environ.get("DATACHAIN_PROJECT")
+        if project_env and len(project_env.split(".")) == 2:
+            # we allow setting both namespace and project in DATACHAIN_PROJECT
+            namespace_env, project_env = project_env.split(".")
+
+        namespace_name = (
+            parsed_namespace_name
+            or namespace_name
+            or namespace_env
+            or self.metastore.default_namespace_name
+        )
+        project_name = (
+            parsed_project_name
+            or project_name
+            or project_env
+            or self.metastore.default_project_name
+        )
+
+        return namespace_name, project_name, name
+
     def get_dataset(
         self, name: str, project: Optional[Project] = None
     ) -> DatasetRecord:
@@ -1074,21 +1108,26 @@ class Catalog:
         namespace_name: str,
         project_name: str,
         version: Optional[str] = None,
+        pull_dataset: bool = False,
+        update: bool = False,
     ) -> DatasetRecord:
-        try:
-            project = self.metastore.get_project(project_name, namespace_name)
-            ds = self.get_dataset(name, project)
-            if version and not ds.has_version(version):
-                raise DatasetVersionNotFoundError(
-                    f"Dataset {name} does not have version {version}"
-                )
-            return ds
+        if self.metastore.is_local_dataset(namespace_name) or not update:
+            try:
+                project = self.metastore.get_project(project_name, namespace_name)
+                ds = self.get_dataset(name, project)
+                if not version or ds.has_version(version):
+                    return ds
+            except (NamespaceNotFoundError, ProjectNotFoundError, DatasetNotFoundError):
+                pass
+
+        if self.metastore.is_local_dataset(namespace_name):
+            raise DatasetNotFoundError(
+                f"Dataset {name}"
+                + (f" version {version} " if version else " ")
+                + "not found"
+            )
 
-        except (
-            ProjectNotFoundError,
-            DatasetNotFoundError,
-            DatasetVersionNotFoundError,
-        ):
+        if pull_dataset:
             print("Dataset not found in local catalog, trying to get from studio")
             remote_ds_uri = create_dataset_uri(
                 name, namespace_name, project_name, version
@@ -1103,6 +1142,8 @@ class Catalog:
                 name, self.metastore.get_project(project_name, namespace_name)
             )
 
+        return self.get_remote_dataset(namespace_name, project_name, name)
+
     def get_dataset_with_version_uuid(self, uuid: str) -> DatasetRecord:
         """Returns dataset that contains version with specific uuid"""
         for dataset in self.ls_datasets():
@@ -1119,6 +1160,10 @@ class Catalog:
 
         info_response = studio_client.dataset_info(namespace, project, name)
         if not info_response.ok:
+            if info_response.status == 404:
+                raise DatasetNotFoundError(
+                    f"Dataset {namespace}.{project}.{name} not found"
+                )
             raise DataChainError(info_response.message)
 
         dataset_info = info_response.data

{datachain-0.22.0 → datachain-0.24.0}/src/datachain/cli/commands/datasets.py

@@ -8,7 +8,6 @@ if TYPE_CHECKING:
 
 from datachain.cli.utils import determine_flavors
 from datachain.config import Config
-from datachain.dataset import parse_dataset_name
 from datachain.error import DataChainError, DatasetNotFoundError
 from datachain.studio import list_datasets as list_datasets_studio
 
@@ -106,9 +105,8 @@ def list_datasets_local(catalog: "Catalog", name: Optional[str] = None):
 
 
 def list_datasets_local_versions(catalog: "Catalog", name: str):
-    namespace_name, project_name, name = parse_dataset_name(name)
-    namespace_name = namespace_name or catalog.metastore.default_namespace_name
-    project_name = project_name or catalog.metastore.default_project_name
+    namespace_name, project_name, name = catalog.get_full_dataset_name(name)
+
     project = catalog.metastore.get_project(project_name, namespace_name)
     ds = catalog.get_dataset(name, project)
     for v in ds.versions:
@@ -137,9 +135,7 @@ def rm_dataset(
     studio: Optional[bool] = False,
     team: Optional[str] = None,
 ):
-    namespace_name, project_name, name = parse_dataset_name(name)
-    namespace_name = namespace_name or catalog.metastore.default_namespace_name
-    project_name = project_name or catalog.metastore.default_project_name
+    namespace_name, project_name, name = catalog.get_full_dataset_name(name)
 
     if not catalog.metastore.is_local_dataset(namespace_name) and studio:
         from datachain.studio import remove_studio_dataset
@@ -166,9 +162,7 @@ def edit_dataset(
     attrs: Optional[list[str]] = None,
     team: Optional[str] = None,
 ):
-    namespace_name, project_name, name = parse_dataset_name(name)
-    namespace_name = namespace_name or catalog.metastore.default_namespace_name
-    project_name = project_name or catalog.metastore.default_project_name
+    namespace_name, project_name, name = catalog.get_full_dataset_name(name)
 
     if catalog.metastore.is_local_dataset(namespace_name):
         try:

{datachain-0.22.0 → datachain-0.24.0}/src/datachain/data_storage/metastore.py

@@ -132,6 +132,7 @@ class AbstractMetastore(ABC, Serializable):
         description: Optional[str] = None,
         uuid: Optional[str] = None,
         ignore_if_exists: bool = True,
+        validate: bool = True,
         **kwargs,
     ) -> Namespace:
         """Creates new namespace"""
@@ -192,6 +193,7 @@ class AbstractMetastore(ABC, Serializable):
         description: Optional[str] = None,
         uuid: Optional[str] = None,
         ignore_if_exists: bool = True,
+        validate: bool = True,
         **kwargs,
     ) -> Project:
         """Creates new project in specific namespace"""
@@ -725,8 +727,11 @@ class AbstractDBMetastore(AbstractMetastore):
         description: Optional[str] = None,
         uuid: Optional[str] = None,
         ignore_if_exists: bool = True,
+        validate: bool = True,
         **kwargs,
     ) -> Namespace:
+        if validate:
+            Namespace.validate_name(name)
         query = self._namespaces_insert().values(
             name=name,
             uuid=uuid or str(uuid4()),
@@ -775,12 +780,15 @@ class AbstractDBMetastore(AbstractMetastore):
         description: Optional[str] = None,
         uuid: Optional[str] = None,
         ignore_if_exists: bool = True,
+        validate: bool = True,
         **kwargs,
     ) -> Project:
+        if validate:
+            Project.validate_name(name)
         try:
             namespace = self.get_namespace(namespace_name)
         except NamespaceNotFoundError:
-            namespace = self.create_namespace(namespace_name)
+            namespace = self.create_namespace(namespace_name, validate=validate)
 
         query = self._projects_insert().values(
             namespace_id=namespace.id,
@@ -817,11 +825,14 @@ class AbstractDBMetastore(AbstractMetastore):
         """Gets a single project inside some namespace by name"""
         n = self._namespaces
         p = self._projects
+        validate = True
+
         if self._is_listing_project(name, namespace_name) or self._is_default_project(
             name, namespace_name
         ):
             # we are always creating default and listing projects if they don't exist
             create = True
+            validate = False
 
         query = self._projects_select(
             *(getattr(n.c, f) for f in self._namespaces_fields),
@@ -834,7 +845,7 @@ class AbstractDBMetastore(AbstractMetastore):
         rows = list(self.db.execute(query, conn=conn))
         if not rows:
             if create:
-                return self.create_project(namespace_name, name)
+                return self.create_project(namespace_name, name, validate=validate)
             raise ProjectNotFoundError(
                 f"Project {name} in namespace {namespace_name} not found."
             )

{datachain-0.22.0 → datachain-0.24.0}/src/datachain/data_storage/sqlite.py

@@ -468,8 +468,12 @@ class SQLiteMetastore(AbstractDBMetastore):
         be created implicitly though, to keep the same fully qualified name with
         Studio dataset.
         """
-        system_namespace = self.create_namespace(Namespace.system(), "System namespace")
-        self.create_project(system_namespace.name, Project.listing(), "Listing project")
+        system_namespace = self.create_namespace(
+            Namespace.system(), "System namespace", validate=False
+        )
+        self.create_project(
+            system_namespace.name, Project.listing(), "Listing project", validate=False
+        )
 
     def _check_schema_version(self) -> None:
         """

{datachain-0.22.0 → datachain-0.24.0}/src/datachain/dataset.py

@@ -12,6 +12,9 @@ from typing import (
 )
 from urllib.parse import urlparse
 
+from packaging.specifiers import SpecifierSet
+from packaging.version import Version
+
 from datachain import semver
 from datachain.error import DatasetVersionNotFoundError, InvalidDatasetNameError
 from datachain.namespace import Namespace
@@ -81,8 +84,10 @@ def create_dataset_uri(
 def parse_dataset_name(name: str) -> tuple[Optional[str], Optional[str], str]:
     """Parses dataset name and returns namespace, project and name"""
     if not name:
-        raise ValueError("Name must be defined to parse it")
+        raise InvalidDatasetNameError("Name must be defined to parse it")
     split = name.split(".")
+    if len(split) > 3:
+        raise InvalidDatasetNameError(f"Invalid dataset name {name}")
     name = split[-1]
     project_name = split[-2] if len(split) > 1 else None
     namespace_name = split[-3] if len(split) > 2 else None
@@ -659,13 +664,39 @@ class DatasetRecord:
             return None
         return max(versions).version
 
-    @property
-    def prev_version(self) -> Optional[str]:
-        """Returns previous version of a dataset"""
-        if len(self.versions) == 1:
+    def latest_compatible_version(self, version_spec: str) -> Optional[str]:
+        """
+        Returns the latest version that matches the given version specifier.
+
+        Supports Python version specifiers like:
+        - ">=1.0.0,<2.0.0" (compatible release range)
+        - "~=1.4.2" (compatible release clause)
+        - "==1.2.*" (prefix matching)
+        - ">1.0.0" (exclusive ordered comparison)
+        - ">=1.0.0" (inclusive ordered comparison)
+        - "!=1.3.0" (version exclusion)
+
+        Args:
+            version_spec: Version specifier string following PEP 440
+
+        Returns:
+            Latest compatible version string, or None if no compatible version found
+        """
+        spec_set = SpecifierSet(version_spec)
+
+        # Convert dataset versions to packaging.Version objects
+        # and filter compatible ones
+        compatible_versions = []
+        for v in self.versions:
+            pkg_version = Version(v.version)
+            if spec_set.contains(pkg_version):
+                compatible_versions.append(v)
+
+        if not compatible_versions:
             return None
 
-        return sorted(self.versions)[-2].version
+        # Return the latest compatible version
+        return max(compatible_versions).version
 
     @classmethod
     def from_dict(cls, d: dict[str, Any]) -> "DatasetRecord":

{datachain-0.22.0 → datachain-0.24.0}/src/datachain/lib/dc/datachain.py

@@ -24,7 +24,7 @@ from pydantic import BaseModel
 from tqdm import tqdm
 
 from datachain import semver
-from datachain.dataset import DatasetRecord, parse_dataset_name
+from datachain.dataset import DatasetRecord
 from datachain.delta import delta_disabled
 from datachain.error import ProjectCreateNotAllowedError, ProjectNotFoundError
 from datachain.func import literal
@@ -557,6 +557,7 @@ class DataChain:
             update_version: which part of the dataset version to automatically increase.
                 Available values: `major`, `minor` or `patch`. Default is `patch`.
         """
+        catalog = self.session.catalog
         if version is not None:
             semver.validate(version)
 
@@ -570,17 +571,10 @@ class DataChain:
                 " patch"
             )
 
-        namespace_name, project_name, name = parse_dataset_name(name)
-
-        namespace_name = (
-            namespace_name
-            or self._settings.namespace
-            or self.session.catalog.metastore.default_namespace_name
-        )
-        project_name = (
-            project_name
-            or self._settings.project
-            or self.session.catalog.metastore.default_project_name
+        namespace_name, project_name, name = catalog.get_full_dataset_name(
+            name,
+            namespace_name=self._settings.namespace,
+            project_name=self._settings.project,
         )
 
         try:

{datachain-0.22.0 → datachain-0.24.0}/src/datachain/lib/dc/datasets.py

@@ -1,16 +1,12 @@
 from collections.abc import Sequence
 from typing import TYPE_CHECKING, Optional, Union, get_origin, get_type_hints
 
-from datachain.dataset import parse_dataset_name
 from datachain.error import (
     DatasetNotFoundError,
     DatasetVersionNotFoundError,
     ProjectNotFoundError,
 )
 from datachain.lib.dataset_info import DatasetInfo
-from datachain.lib.file import (
-    File,
-)
 from datachain.lib.projects import get as get_project
 from datachain.lib.settings import Settings
 from datachain.lib.signal_schema import SignalSchema
@@ -35,7 +31,6 @@ def read_dataset(
     version: Optional[Union[str, int]] = None,
     session: Optional[Session] = None,
     settings: Optional[dict] = None,
-    fallback_to_studio: bool = True,
     delta: Optional[bool] = False,
     delta_on: Optional[Union[str, Sequence[str]]] = (
         "file.path",
@@ -45,6 +40,7 @@ def read_dataset(
     delta_result_on: Optional[Union[str, Sequence[str]]] = None,
     delta_compare: Optional[Union[str, Sequence[str]]] = None,
     delta_retry: Optional[Union[bool, str]] = None,
+    update: bool = False,
 ) -> "DataChain":
     """Get data from a saved Dataset. It returns the chain itself.
     If dataset or version is not found locally, it will try to pull it from Studio.
@@ -56,11 +52,12 @@ def read_dataset(
             set; otherwise, default values will be applied.
         namespace : optional name of namespace in which dataset to read is created
         project : optional name of project in which dataset to read is created
-        version : dataset version
+        version : dataset version. Supports:
+            - Exact version strings: "1.2.3"
+            - Legacy integer versions: 1, 2, 3 (finds latest major version)
+            - Version specifiers (PEP 440): ">=1.0.0,<2.0.0", "~=1.4.2", "==1.2.*", etc.
         session : Session to use for the chain.
         settings : Settings to use for the chain.
-        fallback_to_studio : Try to pull dataset from Studio if not found locally.
-            Default is True.
         delta: If True, only process new or changed files instead of reprocessing
             everything. This saves time by skipping files that were already processed in
             previous versions. The optimization is working when a new version of the
@@ -80,6 +77,10 @@ def read_dataset(
               (error mode)
             - True: Reprocess records missing from the result dataset (missing mode)
             - None: No retry processing (default)
+        update: If True always checks for newer versions available on Studio, even if
+            some version of the dataset exists locally already. If False (default), it
+            will only fetch the dataset from Studio if it is not found locally.
+
 
     Example:
         ```py
@@ -93,11 +94,22 @@ def read_dataset(
         ```
 
         ```py
-        chain = dc.read_dataset("my_cats", fallback_to_studio=False)
+        chain = dc.read_dataset("my_cats", version="1.0.0")
         ```
 
         ```py
-        chain = dc.read_dataset("my_cats", version="1.0.0")
+        # Using version specifiers (PEP 440)
+        chain = dc.read_dataset("my_cats", version=">=1.0.0,<2.0.0")
+        ```
+
+        ```py
+        # Legacy integer version support (finds latest in major version)
+        chain = dc.read_dataset("my_cats", version=1)  # Latest 1.x.x version
+        ```
+
+        ```py
+        # Always check for newer versions matching a version specifier from Studio
+        chain = dc.read_dataset("my_cats", version=">=1.0.0", update=True)
         ```
 
         ```py
@@ -114,7 +126,6 @@ def read_dataset(
             version="1.0.0",
             session=session,
             settings=settings,
-            fallback_to_studio=True,
         )
         ```
     """
@@ -122,41 +133,49 @@ def read_dataset(
 
     from .datachain import DataChain
 
+    telemetry.send_event_once("class", "datachain_init", name=name, version=version)
+
     session = Session.get(session)
     catalog = session.catalog
 
-    namespace_name, project_name, name = parse_dataset_name(name)
-    namespace_name = (
-        namespace_name or namespace or catalog.metastore.default_namespace_name
+    namespace_name, project_name, name = catalog.get_full_dataset_name(
+        name,
+        project_name=project,
+        namespace_name=namespace,
     )
-    project_name = project_name or project or catalog.metastore.default_project_name
 
     if version is not None:
+        dataset = session.catalog.get_dataset_with_remote_fallback(
+            name, namespace_name, project_name, update=update
+        )
+
+        # Convert legacy integer versions to version specifiers
+        # For backward compatibility we still allow users to put version as integer
+        # in which case we convert it to a version specifier that finds the latest
+        # version where major part is equal to that input version.
+        # For example if user sets version=2, we convert it to ">=2.0.0,<3.0.0"
+        # which will find something like 2.4.3 (assuming 2.4.3 is the biggest among
+        # all 2.* dataset versions)
+        if isinstance(version, int):
+            version_spec = f">={version}.0.0,<{version + 1}.0.0"
+        else:
+            version_spec = str(version)
+
+        from packaging.specifiers import InvalidSpecifier, SpecifierSet
+
         try:
-            # for backward compatibility we still allow users to put version as integer
-            # in which case we are trying to find latest version where major part is
-            # equal to that input version. For example if user sets version=2, we could
-            # continue with something like 2.4.3 (assuming 2.4.3 is the biggest among
-            # all 2.* dataset versions). If dataset doesn't have any versions where
-            # major part is equal to that input, exception is thrown.
-            major = int(version)
-            try:
-                ds_project = get_project(project_name, namespace_name, session=session)
-            except ProjectNotFoundError:
-                raise DatasetNotFoundError(
-                    f"Dataset {name} not found in namespace {namespace_name} and",
-                    f" project {project_name}",
-                ) from None
-
-            dataset = session.catalog.get_dataset(name, ds_project)
-            latest_major = dataset.latest_major_version(major)
-            if not latest_major:
+            # Try to parse as version specifier
+            SpecifierSet(version_spec)
+            # If it's a valid specifier set, find the latest compatible version
+            latest_compatible = dataset.latest_compatible_version(version_spec)
+            if not latest_compatible:
                 raise DatasetVersionNotFoundError(
-                    f"Dataset {name} does not have version {version}"
+                    f"No dataset {name} version matching specifier {version_spec}"
                 )
-            version = latest_major
-        except ValueError:
-            # version is in new semver string format, continuing as normal
+            version = latest_compatible
+        except InvalidSpecifier:
+            # If not a valid specifier, treat as exact version string
+            # This handles cases like "1.2.3" which are exact versions, not specifiers
             pass
 
     if settings:
@@ -170,11 +189,8 @@ def read_dataset(
         namespace_name=namespace_name,
         version=version,  #  type: ignore[arg-type]
         session=session,
-        indexing_column_types=File._datachain_column_types,
-        fallback_to_studio=fallback_to_studio,
     )
 
-    telemetry.send_event_once("class", "datachain_init", name=name, version=version)
     signals_schema = SignalSchema({"sys": Sys})
     if query.feature_schema:
         signals_schema |= SignalSchema.deserialize(query.feature_schema)
@@ -320,11 +336,11 @@ def delete_dataset(
     session = Session.get(session, in_memory=in_memory)
     catalog = session.catalog
 
-    namespace_name, project_name, name = parse_dataset_name(name)
-    namespace_name = (
-        namespace_name or namespace or catalog.metastore.default_namespace_name
+    namespace_name, project_name, name = catalog.get_full_dataset_name(
+        name,
+        project_name=project,
+        namespace_name=namespace,
     )
-    project_name = project_name or project or catalog.metastore.default_project_name
 
     if not catalog.metastore.is_local_dataset(namespace_name) and studio:
         return remove_studio_dataset(

{datachain-0.22.0 → datachain-0.24.0}/src/datachain/lib/dc/listings.py

@@ -127,12 +127,8 @@ def read_listing_dataset(
     if version is None:
         version = dataset.latest_version
 
-    query = DatasetQuery(
-        name=name,
-        session=session,
-        indexing_column_types=File._datachain_column_types,
-        fallback_to_studio=False,
-    )
+    query = DatasetQuery(name=name, session=session)
+
     if settings:
         cfg = {**settings}
         if "prefetch" not in cfg:

{datachain-0.22.0 → datachain-0.24.0}/src/datachain/lib/dc/records.py

@@ -97,4 +97,4 @@ def read_records(
     for chunk in batched(records, INSERT_BATCH_SIZE):
         warehouse.insert_rows(table, chunk)
     warehouse.insert_rows_done(table)
-    return read_dataset(name=dsr.name, session=session, settings=settings)
+    return read_dataset(name=dsr.full_name, session=session, settings=settings)

{datachain-0.22.0 → datachain-0.24.0}/src/datachain/lib/projects.py

@@ -54,7 +54,7 @@ def get(name: str, namespace: str, session: Optional[Session]) -> Project:
         ```py
         import datachain as dc
         from datachain.lib.projects import get as get_project
-        project  = get_project("my-project", "local")
+        project = get_project("my-project", "local")
         ```
     """
     return Session.get(session).catalog.metastore.get_project(name, namespace)

{datachain-0.22.0 → datachain-0.24.0}/src/datachain/lib/signal_schema.py

@@ -25,6 +25,7 @@ from pydantic import BaseModel, Field, create_model
 from sqlalchemy import ColumnElement
 from typing_extensions import Literal as LiteralEx
 
+from datachain.func import literal
 from datachain.func.func import Func
 from datachain.lib.convert.python_to_sql import python_to_sql
 from datachain.lib.convert.sql_to_python import sql_to_python
@@ -659,6 +660,7 @@ class SignalSchema:
 
     def mutate(self, args_map: dict) -> "SignalSchema":
         new_values = self.values.copy()
+        primitives = (bool, str, int, float)
 
         for name, value in args_map.items():
             if isinstance(value, Column) and value.name in self.values:
@@ -679,6 +681,12 @@ class SignalSchema:
                 # adding new signal with function
                 new_values[name] = value.get_result_type(self)
                 continue
+            if isinstance(value, primitives):
+                # For primitives, store the type, not the value
+                val = literal(value)
+                val.type = python_to_sql(type(value))()
+                new_values[name] = sql_to_python(val)
+                continue
             if isinstance(value, ColumnElement):
                 # adding new signal
                 new_values[name] = sql_to_python(value)