mlrun 1.7.0rc14__py3-none-any.whl → 1.7.0rc16__py3-none-any.whl

mlrun/datastore/datastore_profile.py

@@ -185,6 +185,18 @@ class DatastoreProfileS3(DatastoreProfile):
     assume_role_arn: typing.Optional[str] = None
     access_key_id: typing.Optional[str] = None
     secret_key: typing.Optional[str] = None
+    bucket: typing.Optional[str] = None
+
+    @pydantic.validator("bucket")
+    @classmethod
+    def check_bucket(cls, v):
+        if not v:
+            warnings.warn(
+                "The 'bucket' attribute will be mandatory starting from version 1.9",
+                FutureWarning,
+                stacklevel=2,
+            )
+        return v
 
     def secrets(self) -> dict:
         res = {}
@@ -203,7 +215,13 @@ class DatastoreProfileS3(DatastoreProfile):
         return res
 
     def url(self, subpath):
-        return f"s3:/{subpath}"
+        # TODO: There is an inconsistency with DatastoreProfileGCS. In DatastoreProfileGCS,
+        # we assume that the subpath can begin without a '/' character,
+        # while here we assume it always starts with one.
+        if self.bucket:
+            return f"s3://{self.bucket}{subpath}"
+        else:
+            return f"s3:/{subpath}"
 
 
 class DatastoreProfileRedis(DatastoreProfile):
@@ -272,6 +290,18 @@ class DatastoreProfileGCS(DatastoreProfile):
     _private_attributes = ("gcp_credentials",)
     credentials_path: typing.Optional[str] = None  # path to file.
     gcp_credentials: typing.Optional[typing.Union[str, dict]] = None
+    bucket: typing.Optional[str] = None
+
+    @pydantic.validator("bucket")
+    @classmethod
+    def check_bucket(cls, v):
+        if not v:
+            warnings.warn(
+                "The 'bucket' attribute will be mandatory starting from version 1.9",
+                FutureWarning,
+                stacklevel=2,
+            )
+        return v
 
     @pydantic.validator("gcp_credentials", pre=True, always=True)
     def convert_dict_to_json(cls, v):
@@ -280,10 +310,15 @@ class DatastoreProfileGCS(DatastoreProfile):
         return v
 
     def url(self, subpath) -> str:
+        # TODO: but there's something wrong with the subpath being assumed to not start with a slash here,
+        # but the opposite assumption is made in S3.
         if subpath.startswith("/"):
             #  in gcs the path after schema is starts with bucket, wherefore it should not start with "/".
             subpath = subpath[1:]
-        return f"gcs://{subpath}"
+        if self.bucket:
+            return f"gcs://{self.bucket}/{subpath}"
+        else:
+            return f"gcs://{subpath}"
 
     def secrets(self) -> dict:
         res = {}
@@ -311,12 +346,27 @@ class DatastoreProfileAzureBlob(DatastoreProfile):
     client_secret: typing.Optional[str] = None
     sas_token: typing.Optional[str] = None
     credential: typing.Optional[str] = None
+    container: typing.Optional[str] = None
+
+    @pydantic.validator("container")
+    @classmethod
+    def check_container(cls, v):
+        if not v:
+            warnings.warn(
+                "The 'container' attribute will be mandatory starting from version 1.9",
+                FutureWarning,
+                stacklevel=2,
+            )
+        return v
 
     def url(self, subpath) -> str:
         if subpath.startswith("/"):
-            #  in azure the path after schema is starts with bucket, wherefore it should not start with "/".
+            #  in azure the path after schema is starts with container, wherefore it should not start with "/".
             subpath = subpath[1:]
-        return f"az://{subpath}"
+        if self.container:
+            return f"az://{self.container}/{subpath}"
+        else:
+            return f"az://{subpath}"
 
     def secrets(self) -> dict:
         res = {}

mlrun/datastore/inmem.py

@@ -80,8 +80,8 @@ class InMemoryStore(DataStore):
             reader = df_module.read_json
         else:
             raise mlrun.errors.MLRunInvalidArgumentError(f"file type unhandled {url}")
-        # InMemoryStore store do not filter on time
-        for field in ["time_column", "start_time", "end_time"]:
+        # InMemoryStore store – don't pass filters
+        for field in ["time_column", "start_time", "end_time", "additional_filters"]:
             kwargs.pop(field, None)
 
         return reader(item, **kwargs)

mlrun/datastore/sources.py

@@ -102,8 +102,12 @@ class BaseSourceDriver(DataSource):
         start_time=None,
         end_time=None,
         time_field=None,
+        additional_filters=None,
     ):
         """return the source data as dataframe"""
+        mlrun.utils.helpers.additional_filters_warning(
+            additional_filters, self.__class__
+        )
         return mlrun.store_manager.object(url=self.path).as_df(
             columns=columns,
             df_module=df_module,
@@ -245,7 +249,11 @@ class CSVSource(BaseSourceDriver):
         start_time=None,
         end_time=None,
         time_field=None,
+        additional_filters=None,
     ):
+        mlrun.utils.helpers.additional_filters_warning(
+            additional_filters, self.__class__
+        )
         reader_args = self.attributes.get("reader_args", {})
         return mlrun.store_manager.object(url=self.path).as_df(
             columns=columns,
@@ -281,6 +289,12 @@ class ParquetSource(BaseSourceDriver):
     :parameter start_time: filters out data before this time
     :parameter end_time: filters out data after this time
     :parameter attributes: additional parameters to pass to storey.
+    :param additional_filters: List of additional_filter conditions as tuples.
+                               Each tuple should be in the format (column_name, operator, value).
+                               Supported operators: "=", ">=", "<=", ">", "<".
+                               Example: [("Product", "=", "Computer")]
+                               For all supported filters, please see:
+                               https://arrow.apache.org/docs/python/generated/pyarrow.parquet.ParquetDataset.html
     """
 
     kind = "parquet"
@@ -297,6 +311,7 @@ class ParquetSource(BaseSourceDriver):
         schedule: str = None,
         start_time: Optional[Union[datetime, str]] = None,
         end_time: Optional[Union[datetime, str]] = None,
+        additional_filters: Optional[list[tuple]] = None,
     ):
         super().__init__(
             name,
@@ -308,6 +323,7 @@ class ParquetSource(BaseSourceDriver):
             start_time,
             end_time,
         )
+        self.additional_filters = additional_filters
 
     @property
     def start_time(self):
@@ -341,6 +357,7 @@ class ParquetSource(BaseSourceDriver):
         start_time=None,
         end_time=None,
         context=None,
+        additional_filters=None,
     ):
         import storey
 
@@ -358,6 +375,7 @@ class ParquetSource(BaseSourceDriver):
             end_filter=self.end_time,
             start_filter=self.start_time,
             filter_column=self.time_field or time_field,
+            additional_filters=self.additional_filters or additional_filters,
             **attributes,
         )
 
@@ -380,6 +398,7 @@ class ParquetSource(BaseSourceDriver):
         start_time=None,
         end_time=None,
         time_field=None,
+        additional_filters=None,
     ):
         reader_args = self.attributes.get("reader_args", {})
         return mlrun.store_manager.object(url=self.path).as_df(
@@ -389,6 +408,7 @@ class ParquetSource(BaseSourceDriver):
             end_time=end_time or self.end_time,
             time_column=time_field or self.time_field,
             format="parquet",
+            additional_filters=additional_filters or self.additional_filters,
             **reader_args,
         )
 
@@ -519,10 +539,15 @@ class BigQuerySource(BaseSourceDriver):
         start_time=None,
         end_time=None,
         time_field=None,
+        additional_filters=None,
     ):
         from google.cloud import bigquery
         from google.cloud.bigquery_storage_v1 import BigQueryReadClient
 
+        mlrun.utils.helpers.additional_filters_warning(
+            additional_filters, self.__class__
+        )
+
         def schema_to_dtypes(schema):
             from mlrun.data_types.data_types import gbq_to_pandas_dtype
 
@@ -562,7 +587,6 @@ class BigQuerySource(BaseSourceDriver):
         else:
             df = rows_iterator.to_dataframe(dtypes=dtypes)
 
-        # TODO : filter as part of the query
         return select_columns_from_df(
             filter_df_start_end_time(
                 df,
@@ -740,7 +764,19 @@ class DataFrameSource:
             context=self.context or context,
         )
 
-    def to_dataframe(self, **kwargs):
+    def to_dataframe(
+        self,
+        columns=None,
+        df_module=None,
+        entities=None,
+        start_time=None,
+        end_time=None,
+        time_field=None,
+        additional_filters=None,
+    ):
+        mlrun.utils.helpers.additional_filters_warning(
+            additional_filters, self.__class__
+        )
         return self._df
 
     def is_iterator(self):
@@ -935,6 +971,7 @@ class KafkaSource(OnlineSource):
         start_time=None,
         end_time=None,
         time_field=None,
+        additional_filters=None,
     ):
         raise mlrun.MLRunInvalidArgumentError(
             "KafkaSource does not support batch processing"
@@ -1075,9 +1112,13 @@ class SQLSource(BaseSourceDriver):
         start_time=None,
         end_time=None,
         time_field=None,
+        additional_filters=None,
     ):
         import sqlalchemy as sqlalchemy
 
+        mlrun.utils.helpers.additional_filters_warning(
+            additional_filters, self.__class__
+        )
         db_path = self.attributes.get("db_path")
         table_name = self.attributes.get("table_name")
         parse_dates = self.attributes.get("parse_dates")

mlrun/datastore/store_resources.py

@@ -17,7 +17,7 @@
 import mlrun
 import mlrun.artifacts
 from mlrun.config import config
-from mlrun.utils.helpers import is_legacy_artifact, parse_artifact_uri
+from mlrun.utils.helpers import parse_artifact_uri
 
 from ..common.helpers import parse_versioned_object_uri
 from ..platforms.iguazio import parse_path
@@ -167,11 +167,7 @@ def get_store_resource(
         )
         if resource.get("kind", "") == "link":
             # todo: support other link types (not just iter, move this to the db/api layer
-            link_iteration = (
-                resource.get("link_iteration", 0)
-                if is_legacy_artifact(resource)
-                else resource["spec"].get("link_iteration", 0)
-            )
+            link_iteration = resource["spec"].get("link_iteration", 0)
 
             resource = db.read_artifact(
                 key,

mlrun/datastore/targets.py

@@ -656,6 +656,29 @@ class BaseStoreTarget(DataTargetBase):
     def _target_path_object(self):
         """return the actual/computed target path"""
         is_single_file = hasattr(self, "is_single_file") and self.is_single_file()
+
+        if self._resource and self.path:
+            parsed_url = urlparse(self.path)
+            # When the URL consists only from scheme and endpoint and no path,
+            # make a default path for DS and redis targets.
+            # Also ignore KafkaTarget when it uses the ds scheme (no default path for KafkaTarget)
+            if (
+                not isinstance(self, KafkaTarget)
+                and parsed_url.scheme in ["ds", "redis", "rediss"]
+                and (not parsed_url.path or parsed_url.path == "/")
+            ):
+                return TargetPathObject(
+                    _get_target_path(
+                        self,
+                        self._resource,
+                        self.run_id is not None,
+                        netloc=parsed_url.netloc,
+                        scheme=parsed_url.scheme,
+                    ),
+                    self.run_id,
+                    is_single_file,
+                )
+
         return self.get_path() or (
             TargetPathObject(
                 _get_target_path(self, self._resource, self.run_id is not None),
@@ -714,9 +737,13 @@ class BaseStoreTarget(DataTargetBase):
         start_time=None,
         end_time=None,
         time_column=None,
+        additional_filters=None,
         **kwargs,
     ):
         """return the target data as dataframe"""
+        mlrun.utils.helpers.additional_filters_warning(
+            additional_filters, self.__class__
+        )
         return mlrun.get_dataitem(self.get_target_path()).as_df(
             columns=columns,
             df_module=df_module,
@@ -961,6 +988,7 @@ class ParquetTarget(BaseStoreTarget):
         start_time=None,
         end_time=None,
         time_column=None,
+        additional_filters=None,
         **kwargs,
     ):
         """return the target data as dataframe"""
@@ -971,6 +999,7 @@ class ParquetTarget(BaseStoreTarget):
             start_time=start_time,
             end_time=end_time,
             time_column=time_column,
+            additional_filters=additional_filters,
             **kwargs,
         )
         if not columns:
@@ -1101,8 +1130,12 @@ class CSVTarget(BaseStoreTarget):
         start_time=None,
         end_time=None,
         time_column=None,
+        additional_filters=None,
         **kwargs,
     ):
+        mlrun.utils.helpers.additional_filters_warning(
+            additional_filters, self.__class__
+        )
         df = super().as_df(
             columns=columns,
             df_module=df_module,
@@ -1209,6 +1242,7 @@ class SnowflakeTarget(BaseStoreTarget):
         start_time=None,
         end_time=None,
         time_column=None,
+        additional_filters=None,
         **kwargs,
     ):
         raise NotImplementedError()
@@ -1275,7 +1309,17 @@ class NoSqlBaseTarget(BaseStoreTarget):
     def get_dask_options(self):
         return {"format": "csv"}
 
-    def as_df(self, columns=None, df_module=None, **kwargs):
+    def as_df(
+        self,
+        columns=None,
+        df_module=None,
+        entities=None,
+        start_time=None,
+        end_time=None,
+        time_column=None,
+        additional_filters=None,
+        **kwargs,
+    ):
         raise NotImplementedError()
 
     def write_dataframe(
@@ -1390,39 +1434,6 @@ class RedisNoSqlTarget(NoSqlBaseTarget):
     support_spark = True
     writer_step_name = "RedisNoSqlTarget"
 
-    @property
-    def _target_path_object(self):
-        url = self.path or mlrun.mlconf.redis.url
-        if self._resource and url:
-            parsed_url = urlparse(url)
-            if not parsed_url.path or parsed_url.path == "/":
-                kind_prefix = (
-                    "sets"
-                    if self._resource.kind
-                    == mlrun.common.schemas.ObjectKind.feature_set
-                    else "vectors"
-                )
-                kind = self.kind
-                name = self._resource.metadata.name
-                project = (
-                    self._resource.metadata.project or mlrun.mlconf.default_project
-                )
-                data_prefix = get_default_prefix_for_target(kind).format(
-                    ds_profile_name=parsed_url.netloc,
-                    authority=parsed_url.netloc,
-                    project=project,
-                    kind=kind,
-                    name=name,
-                )
-                if url.startswith("rediss://"):
-                    data_prefix = data_prefix.replace("redis://", "rediss://", 1)
-                if not self.run_id:
-                    version = self._resource.metadata.tag or "latest"
-                    name = f"{name}-{version}"
-                url = f"{data_prefix}/{kind_prefix}/{name}"
-                return TargetPathObject(url, self.run_id, False)
-        return super()._target_path_object
-
     # Fetch server url from the RedisNoSqlTarget::__init__() 'path' parameter.
     # If not set fetch it from 'mlrun.mlconf.redis.url' (MLRUN_REDIS__URL environment variable).
     # Then look for username and password at REDIS_xxx secrets
@@ -1544,7 +1555,17 @@ class StreamTarget(BaseStoreTarget):
             **self.attributes,
         )
 
-    def as_df(self, columns=None, df_module=None, **kwargs):
+    def as_df(
+        self,
+        columns=None,
+        df_module=None,
+        entities=None,
+        start_time=None,
+        end_time=None,
+        time_column=None,
+        additional_filters=None,
+        **kwargs,
+    ):
         raise NotImplementedError()
 
 
@@ -1649,7 +1670,17 @@ class KafkaTarget(BaseStoreTarget):
             **attributes,
         )
 
-    def as_df(self, columns=None, df_module=None, **kwargs):
+    def as_df(
+        self,
+        columns=None,
+        df_module=None,
+        entities=None,
+        start_time=None,
+        end_time=None,
+        time_column=None,
+        additional_filters=None,
+        **kwargs,
+    ):
         raise NotImplementedError()
 
     def purge(self):
@@ -1696,7 +1727,17 @@ class TSDBTarget(BaseStoreTarget):
             **self.attributes,
         )
 
-    def as_df(self, columns=None, df_module=None, **kwargs):
+    def as_df(
+        self,
+        columns=None,
+        df_module=None,
+        entities=None,
+        start_time=None,
+        end_time=None,
+        time_column=None,
+        additional_filters=None,
+        **kwargs,
+    ):
         raise NotImplementedError()
 
     def write_dataframe(
@@ -1807,11 +1848,16 @@ class DFTarget(BaseStoreTarget):
         self,
         columns=None,
         df_module=None,
+        entities=None,
         start_time=None,
         end_time=None,
         time_column=None,
+        additional_filters=None,
         **kwargs,
     ):
+        mlrun.utils.helpers.additional_filters_warning(
+            additional_filters, self.__class__
+        )
         return select_columns_from_df(
             filter_df_start_end_time(
                 self._df,
@@ -1986,6 +2032,7 @@ class SQLTarget(BaseStoreTarget):
         start_time=None,
         end_time=None,
         time_column=None,
+        additional_filters=None,
         **kwargs,
     ):
         try:
@@ -1994,6 +2041,10 @@ class SQLTarget(BaseStoreTarget):
         except (ModuleNotFoundError, ImportError) as exc:
             self._raise_sqlalchemy_import_error(exc)
 
+        mlrun.utils.helpers.additional_filters_warning(
+            additional_filters, self.__class__
+        )
+
         db_path, table_name, _, _, _, _ = self._parse_url()
         engine = sqlalchemy.create_engine(db_path)
         parse_dates: Optional[list[str]] = self.attributes.get("parse_dates")
@@ -2140,7 +2191,7 @@ kind_to_driver = {
 }
 
 
-def _get_target_path(driver, resource, run_id_mode=False):
+def _get_target_path(driver, resource, run_id_mode=False, netloc=None, scheme=""):
     """return the default target path given the resource and target kind"""
     kind = driver.kind
     suffix = driver.suffix
@@ -2157,11 +2208,27 @@ def _get_target_path(driver, resource, run_id_mode=False):
     )
     name = resource.metadata.name
     project = resource.metadata.project or mlrun.mlconf.default_project
-    data_prefix = get_default_prefix_for_target(kind).format(
+
+    default_kind_name = kind
+    if scheme == "ds":
+        # "dsnosql" is not an actual target like Parquet or Redis; rather, it serves
+        # as a placeholder that can be used in any specified target
+        default_kind_name = "dsnosql"
+    if scheme == "redis" or scheme == "rediss":
+        default_kind_name = TargetTypes.redisnosql
+
+    netloc = netloc or ""
+    data_prefix = get_default_prefix_for_target(default_kind_name).format(
+        ds_profile_name=netloc,  # In case of ds profile, set its the name
+        authority=netloc,  # In case of redis, replace {authority} with netloc
         project=project,
         kind=kind,
         name=name,
     )
+
+    if scheme == "rediss":
+        data_prefix = data_prefix.replace("redis://", "rediss://", 1)
+
     # todo: handle ver tag changes, may need to copy files?
     if not run_id_mode:
         version = resource.metadata.tag

mlrun/db/base.py

@@ -16,6 +16,7 @@ import datetime
 from abc import ABC, abstractmethod
 from typing import Optional, Union
 
+import mlrun.alerts
 import mlrun.common.schemas
 import mlrun.model_monitoring
 
@@ -117,7 +118,18 @@ class RunDBInterface(ABC):
         pass
 
     @abstractmethod
-    def del_artifact(self, key, tag="", project="", tree=None, uid=None):
+    def del_artifact(
+        self,
+        key,
+        tag="",
+        project="",
+        tree=None,
+        uid=None,
+        deletion_strategy: mlrun.common.schemas.artifact.ArtifactsDeletionStrategies = (
+            mlrun.common.schemas.artifact.ArtifactsDeletionStrategies.metadata_only
+        ),
+        secrets: dict = None,
+    ):
         pass
 
     @abstractmethod
@@ -543,7 +555,7 @@ class RunDBInterface(ABC):
         end: Optional[str] = None,
         metrics: Optional[list[str]] = None,
         features: bool = False,
-    ):
+    ) -> mlrun.model_monitoring.ModelEndpoint:
         pass
 
     @abstractmethod
@@ -664,7 +676,7 @@ class RunDBInterface(ABC):
     def store_alert_config(
         self,
         alert_name: str,
-        alert_data: Union[dict, mlrun.common.schemas.AlertConfig],
+        alert_data: Union[dict, mlrun.alerts.alert.AlertConfig],
         project="",
     ):
         pass
@@ -685,6 +697,14 @@ class RunDBInterface(ABC):
     def reset_alert_config(self, alert_name: str, project=""):
         pass
 
+    @abstractmethod
+    def get_alert_template(self, template_name: str):
+        pass
+
+    @abstractmethod
+    def list_alert_templates(self):
+        pass
+
     @abstractmethod
     def get_builder_status(
         self,