datachain 0.16.3__tar.gz → 0.16.5__tar.gz

{datachain-0.16.3 → datachain-0.16.5}/.github/workflows/benchmarks.yml

@@ -25,11 +25,14 @@ jobs:
           python-version: '3.13'
 
       - name: Setup uv
-        uses: astral-sh/setup-uv@v5
+        uses: astral-sh/setup-uv@v6
         with:
           enable-cache: true
           cache-suffix: benchmarks
           cache-dependency-glob: pyproject.toml
+          # revert after this is fixed
+          # https://github.com/wntrblm/nox/issues/953
+          version: ">=0.6,<0.7"
 
       - name: Install nox and dvc
         run: uv pip install dvc[gs] nox --system

{datachain-0.16.3 → datachain-0.16.5}/.github/workflows/release.yml

@@ -27,7 +27,7 @@ jobs:
           python-version: '3.13'
 
       - name: Setup uv
-        uses: astral-sh/setup-uv@v5
+        uses: astral-sh/setup-uv@v6
 
       - name: Install nox
         run: uv pip install nox --system

{datachain-0.16.3 → datachain-0.16.5}/.github/workflows/tests-studio.yml

@@ -84,7 +84,7 @@ jobs:
           python-version: ${{ matrix.pyv }}
 
       - name: Setup uv
-        uses: astral-sh/setup-uv@v5
+        uses: astral-sh/setup-uv@v6
         with:
           enable-cache: true
           cache-suffix: studio
@@ -98,6 +98,7 @@ jobs:
       - name: Run tests
         # Generate `.test_durations` file with `pytest --store-durations --durations-path ../.github/.test_durations ...`
         run: >
+          PYTHONPATH="$(pwd)/..:${PYTHONPATH}"
           pytest
           --config-file=pyproject.toml -rs
           --splits=6 --group=${{ matrix.group }} --durations-path=../../.github/.test_durations

{datachain-0.16.3 → datachain-0.16.5}/.github/workflows/tests.yml

@@ -29,11 +29,14 @@ jobs:
           python-version: '3.9'
 
       - name: Setup uv
-        uses: astral-sh/setup-uv@v5
+        uses: astral-sh/setup-uv@v6
         with:
           enable-cache: true
           cache-suffix: lint
           cache-dependency-glob: pyproject.toml
+          # revert after this is fixed
+          # https://github.com/wntrblm/nox/issues/953
+          version: ">=0.6,<0.7"
 
       - name: Install nox
         run: uv pip install nox --system
@@ -87,11 +90,14 @@ jobs:
           python-version: ${{ matrix.pyv }}
 
       - name: Setup uv
-        uses: astral-sh/setup-uv@v5
+        uses: astral-sh/setup-uv@v6
         with:
           enable-cache: true
           cache-suffix: tests-${{ matrix.pyv }}
           cache-dependency-glob: pyproject.toml
+          # revert after this is fixed
+          # https://github.com/wntrblm/nox/issues/953
+          version: ">=0.6,<0.7"
 
       - name: Install nox
         run: uv pip install nox --system
@@ -154,11 +160,14 @@ jobs:
           python-version: ${{ matrix.pyv }}
 
       - name: Setup uv
-        uses: astral-sh/setup-uv@v5
+        uses: astral-sh/setup-uv@v6
         with:
           enable-cache: true
           cache-suffix: examples-${{ matrix.pyv }}
           cache-dependency-glob: pyproject.toml
+          # revert after this is fixed
+          # https://github.com/wntrblm/nox/issues/953
+          version: ">=0.6,<0.7"
 
       - name: Install nox
         run: uv pip install nox --system

{datachain-0.16.3 → datachain-0.16.5}/.pre-commit-config.yaml

@@ -24,7 +24,7 @@ repos:
       - id: trailing-whitespace
         exclude: '^LICENSES/'
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: 'v0.11.6'
+    rev: 'v0.11.8'
     hooks:
       - id: ruff
         args: [--fix, --exit-non-zero-on-fix]

{datachain-0.16.3/src/datachain.egg-info → datachain-0.16.5}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: datachain
-Version: 0.16.3
+Version: 0.16.5
 Summary: Wrangle unstructured AI data at scale
 Author-email: Dmitry Petrov <support@dvc.org>
 License-Expression: Apache-2.0

{datachain-0.16.3 → datachain-0.16.5}/docs/commands/job/run.md

@@ -60,6 +60,11 @@ datachain job run --workers 4 --files utils.py config.json query.py
 datachain job run --env API_KEY=123 --req pandas numpy query.py
 ```
 
+6. Run a job with a repository (will be cloned in the job working directory):
+```bash
+datachain job run --repository https://github.com/iterative/datachain query.py
+```
+
 ## Notes
 
 * Closing the logs command (e.g., with Ctrl+C) will only stop displaying the logs but will not cancel the job execution

{datachain-0.16.3 → datachain-0.16.5}/src/datachain/catalog/catalog.py

@@ -79,6 +79,7 @@ DATASET_INTERNAL_ERROR_MESSAGE = "Internal error on creating dataset"
 QUERY_SCRIPT_INVALID_LAST_STATEMENT_EXIT_CODE = 10
 # exit code we use if query script was canceled
 QUERY_SCRIPT_CANCELED_EXIT_CODE = 11
+QUERY_SCRIPT_SIGTERM_EXIT_CODE = -15  # if query script was terminated by SIGTERM
 
 # dataset pull
 PULL_DATASET_MAX_THREADS = 5
@@ -1645,7 +1646,10 @@ class Catalog:
                     thread.join()  # wait for the reader thread
 
         logger.info("Process %s exited with return code %s", proc.pid, proc.returncode)
-        if proc.returncode == QUERY_SCRIPT_CANCELED_EXIT_CODE:
+        if proc.returncode in (
+            QUERY_SCRIPT_CANCELED_EXIT_CODE,
+            QUERY_SCRIPT_SIGTERM_EXIT_CODE,
+        ):
             raise QueryScriptCancelError(
                 "Query script was canceled by user",
                 return_code=proc.returncode,

{datachain-0.16.3 → datachain-0.16.5}/src/datachain/cli/__init__.py

@@ -34,8 +34,10 @@ def main(argv: Optional[list[str]] = None) -> int:
     datachain_parser = get_parser()
     args = datachain_parser.parse_args(argv)
 
-    if args.command in ("internal-run-udf", "internal-run-udf-worker"):
-        return handle_udf(args.command)
+    if args.command == "internal-run-udf":
+        return handle_udf()
+    if args.command == "internal-run-udf-worker":
+        return handle_udf_runner(args.fd)
 
     if args.command is None:
         datachain_parser.print_help(sys.stderr)
@@ -303,13 +305,13 @@ def handle_general_exception(exc, args, logging_level):
     return error, 1
 
 
-def handle_udf(command):
-    if command == "internal-run-udf":
-        from datachain.query.dispatch import udf_entrypoint
+def handle_udf() -> int:
+    from datachain.query.dispatch import udf_entrypoint
 
-        return udf_entrypoint()
+    return udf_entrypoint()
 
-    if command == "internal-run-udf-worker":
-        from datachain.query.dispatch import udf_worker_entrypoint
 
-        return udf_worker_entrypoint()
+def handle_udf_runner(fd: Optional[int] = None) -> int:
+    from datachain.query.dispatch import udf_worker_entrypoint
+
+    return udf_worker_entrypoint(fd)

{datachain-0.16.3 → datachain-0.16.5}/src/datachain/cli/commands/query.py

@@ -29,6 +29,7 @@ def query(
         name=os.path.basename(script),
         query=script_content,
         query_type=JobQueryType.PYTHON,
+        status=JobStatus.RUNNING,
         python_version=python_version,
         params=params,
     )

{datachain-0.16.3 → datachain-0.16.5}/src/datachain/cli/parser/__init__.py

@@ -549,7 +549,15 @@ def get_parser() -> ArgumentParser:  # noqa: PLR0915
     add_anon_arg(parse_gc)
 
     subp.add_parser("internal-run-udf", parents=[parent_parser])
-    subp.add_parser("internal-run-udf-worker", parents=[parent_parser])
+    run_udf_worker = subp.add_parser("internal-run-udf-worker", parents=[parent_parser])
+    run_udf_worker.add_argument(
+        "--fd",
+        type=int,
+        action="store",
+        default=None,
+        help="File descriptor to write results to",
+    )
+
     add_completion_parser(subp, [parent_parser])
     return parser
 

{datachain-0.16.3 → datachain-0.16.5}/src/datachain/cli/parser/job.py

@@ -13,7 +13,7 @@ def add_jobs_parser(subparsers, parent_parser) -> None:
     )
     jobs_subparser = jobs_parser.add_subparsers(
         dest="cmd",
-        help="Use `datachain auth CMD --help` to display command-specific help",
+        help="Use `datachain job CMD --help` to display command-specific help",
     )
 
     studio_run_help = "Run a job in Studio"
@@ -66,6 +66,11 @@ def add_jobs_parser(subparsers, parent_parser) -> None:
         action="store",
         help="Python version for the job (e.g., 3.9, 3.10, 3.11)",
     )
+    studio_run_parser.add_argument(
+        "--repository",
+        action="store",
+        help="Repository URL to clone before running the job",
+    )
     studio_run_parser.add_argument(
         "--req-file",
         action="store",

{datachain-0.16.3 → datachain-0.16.5}/src/datachain/data_storage/job.py

@@ -3,6 +3,7 @@ from enum import Enum
 
 class JobStatus(int, Enum):
     CREATED = 1
+    SCHEDULED = 10
     QUEUED = 2
     INIT = 3
     RUNNING = 4

{datachain-0.16.3 → datachain-0.16.5}/src/datachain/data_storage/metastore.py

@@ -254,6 +254,7 @@ class AbstractMetastore(ABC, Serializable):
         name: str,
         query: str,
         query_type: JobQueryType = JobQueryType.PYTHON,
+        status: JobStatus = JobStatus.CREATED,
         workers: int = 1,
         python_version: Optional[str] = None,
         params: Optional[dict[str, str]] = None,
@@ -264,33 +265,35 @@ class AbstractMetastore(ABC, Serializable):
         """
 
     @abstractmethod
-    def set_job_status(
+    def get_job(self, job_id: str) -> Optional[Job]:
+        """Returns the job with the given ID."""
+
+    @abstractmethod
+    def update_job(
         self,
         job_id: str,
-        status: JobStatus,
+        status: Optional[JobStatus] = None,
+        exit_code: Optional[int] = None,
         error_message: Optional[str] = None,
         error_stack: Optional[str] = None,
+        finished_at: Optional[datetime] = None,
         metrics: Optional[dict[str, Any]] = None,
-    ) -> None:
-        """Set the status of the given job."""
+    ) -> Optional["Job"]:
+        """Updates job fields."""
 
     @abstractmethod
-    def get_job_status(self, job_id: str) -> Optional[JobStatus]:
-        """Returns the status of the given job."""
-
-    @abstractmethod
-    def set_job_and_dataset_status(
+    def set_job_status(
         self,
         job_id: str,
-        job_status: JobStatus,
-        dataset_status: DatasetStatus,
+        status: JobStatus,
+        error_message: Optional[str] = None,
+        error_stack: Optional[str] = None,
     ) -> None:
-        """Set the status of the given job and dataset."""
+        """Set the status of the given job."""
 
     @abstractmethod
-    def get_job_dataset_versions(self, job_id: str) -> list[tuple[str, int]]:
-        """Returns dataset names and versions for the job."""
-        raise NotImplementedError
+    def get_job_status(self, job_id: str) -> Optional[JobStatus]:
+        """Returns the status of the given job."""
 
 
 class AbstractDBMetastore(AbstractMetastore):
@@ -651,30 +654,31 @@ class AbstractDBMetastore(AbstractMetastore):
         dataset_version = dataset.get_version(version)
 
         values = {}
+        version_values: dict = {}
         for field, value in kwargs.items():
             if field in self._dataset_version_fields[1:]:
                 if field == "schema":
-                    dataset_version.update(**{field: DatasetRecord.parse_schema(value)})
                     values[field] = json.dumps(value) if value else None
+                    version_values[field] = DatasetRecord.parse_schema(value)
                 elif field == "feature_schema":
                     values[field] = json.dumps(value) if value else None
+                    version_values[field] = value
                 elif field == "preview" and isinstance(value, list):
                     values[field] = json.dumps(value, cls=JSONSerialize)
+                    version_values[field] = value
                 else:
                     values[field] = value
-                    dataset_version.update(**{field: value})
-
-        if not values:
-            # Nothing to update
-            return dataset_version
+                    version_values[field] = value
 
-        dv = self._datasets_versions
-        self.db.execute(
-            self._datasets_versions_update()
-            .where(dv.c.id == dataset_version.id)
-            .values(values),
-            conn=conn,
-        )  # type: ignore [attr-defined]
+        if values:
+            dv = self._datasets_versions
+            self.db.execute(
+                self._datasets_versions_update()
+                .where(dv.c.dataset_id == dataset.id and dv.c.version == version)
+                .values(values),
+                conn=conn,
+            )  # type: ignore [attr-defined]
+            dataset_version.update(**version_values)
 
         return dataset_version
 
@@ -702,7 +706,7 @@ class AbstractDBMetastore(AbstractMetastore):
         dataset_fields: list[str],
         dataset_version_fields: list[str],
         isouter: bool = True,
-    ):
+    ) -> "Select":
         if not (
             self.db.has_table(self._datasets.name)
             and self.db.has_table(self._datasets_versions.name)
@@ -719,12 +723,12 @@ class AbstractDBMetastore(AbstractMetastore):
         j = d.join(dv, d.c.id == dv.c.dataset_id, isouter=isouter)
         return query.select_from(j)
 
-    def _base_dataset_query(self):
+    def _base_dataset_query(self) -> "Select":
         return self._get_dataset_query(
             self._dataset_fields, self._dataset_version_fields
         )
 
-    def _base_list_datasets_query(self):
+    def _base_list_datasets_query(self) -> "Select":
         return self._get_dataset_query(
             self._dataset_list_fields, self._dataset_list_version_fields, isouter=False
         )
@@ -1018,6 +1022,7 @@ class AbstractDBMetastore(AbstractMetastore):
         name: str,
         query: str,
         query_type: JobQueryType = JobQueryType.PYTHON,
+        status: JobStatus = JobStatus.CREATED,
         workers: int = 1,
         python_version: Optional[str] = None,
         params: Optional[dict[str, str]] = None,
@@ -1032,7 +1037,7 @@ class AbstractDBMetastore(AbstractMetastore):
             self._jobs_insert().values(
                 id=job_id,
                 name=name,
-                status=JobStatus.CREATED,
+                status=status,
                 created_at=datetime.now(timezone.utc),
                 query=query,
                 query_type=query_type.value,
@@ -1047,25 +1052,65 @@ class AbstractDBMetastore(AbstractMetastore):
         )
         return job_id
 
+    def get_job(self, job_id: str, conn=None) -> Optional[Job]:
+        """Returns the job with the given ID."""
+        query = self._jobs_select(self._jobs).where(self._jobs.c.id == job_id)
+        results = list(self.db.execute(query, conn=conn))
+        if not results:
+            return None
+        return self._parse_job(results[0])
+
+    def update_job(
+        self,
+        job_id: str,
+        status: Optional[JobStatus] = None,
+        exit_code: Optional[int] = None,
+        error_message: Optional[str] = None,
+        error_stack: Optional[str] = None,
+        finished_at: Optional[datetime] = None,
+        metrics: Optional[dict[str, Any]] = None,
+        conn: Optional[Any] = None,
+    ) -> Optional["Job"]:
+        """Updates job fields."""
+        values: dict = {}
+        if status is not None:
+            values["status"] = status
+        if exit_code is not None:
+            values["exit_code"] = exit_code
+        if error_message is not None:
+            values["error_message"] = error_message
+        if error_stack is not None:
+            values["error_stack"] = error_stack
+        if finished_at is not None:
+            values["finished_at"] = finished_at
+        if metrics:
+            values["metrics"] = json.dumps(metrics)
+
+        if values:
+            j = self._jobs
+            self.db.execute(
+                self._jobs_update().where(j.c.id == job_id).values(**values),
+                conn=conn,
+            )  # type: ignore [attr-defined]
+
+        return self.get_job(job_id, conn=conn)
+
     def set_job_status(
         self,
         job_id: str,
         status: JobStatus,
         error_message: Optional[str] = None,
         error_stack: Optional[str] = None,
-        metrics: Optional[dict[str, Any]] = None,
         conn: Optional[Any] = None,
     ) -> None:
         """Set the status of the given job."""
-        values: dict = {"status": status.value}
-        if status.value in JobStatus.finished():
+        values: dict = {"status": status}
+        if status in JobStatus.finished():
             values["finished_at"] = datetime.now(timezone.utc)
         if error_message:
             values["error_message"] = error_message
         if error_stack:
             values["error_stack"] = error_stack
-        if metrics:
-            values["metrics"] = json.dumps(metrics)
         self.db.execute(
             self._jobs_update(self._jobs.c.id == job_id).values(**values),
             conn=conn,
@@ -1086,37 +1131,3 @@ class AbstractDBMetastore(AbstractMetastore):
         if not results:
             return None
         return results[0][0]
-
-    def set_job_and_dataset_status(
-        self,
-        job_id: str,
-        job_status: JobStatus,
-        dataset_status: DatasetStatus,
-    ) -> None:
-        """Set the status of the given job and dataset."""
-        with self.db.transaction() as conn:
-            self.set_job_status(job_id, status=job_status, conn=conn)
-            dv = self._datasets_versions
-            query = (
-                self._datasets_versions_update()
-                .where(
-                    (dv.c.job_id == job_id) & (dv.c.status != DatasetStatus.COMPLETE)
-                )
-                .values(status=dataset_status)
-            )
-            self.db.execute(query, conn=conn)  # type: ignore[attr-defined]
-
-    def get_job_dataset_versions(self, job_id: str) -> list[tuple[str, int]]:
-        """Returns dataset names and versions for the job."""
-        dv = self._datasets_versions
-        ds = self._datasets
-
-        join_condition = dv.c.dataset_id == ds.c.id
-
-        query = (
-            self._datasets_versions_select(ds.c.name, dv.c.version)
-            .select_from(dv.join(ds, join_condition))
-            .where(dv.c.job_id == job_id)
-        )
-
-        return list(self.db.execute(query))