ddeutil-workflow 0.0.4__py3-none-any.whl → 0.0.6__py3-none-any.whl

ddeutil/workflow/__about__.py

@@ -1 +1 @@
-__version__: str = "0.0.4"
+__version__: str = "0.0.6"

ddeutil/workflow/__init__.py

@@ -0,0 +1,9 @@
+# ------------------------------------------------------------------------------
+# Copyright (c) 2022 Korawich Anuttra. All rights reserved.
+# Licensed under the MIT License. See LICENSE in the project root for
+# license information.
+# ------------------------------------------------------------------------------
+from .exceptions import StageException
+from .on import On
+from .pipeline import Pipeline
+from .stage import Stage

ddeutil/workflow/__types.py

@@ -5,8 +5,50 @@
 # ------------------------------------------------------------------------------
 from __future__ import annotations
 
-from typing import Any
+import re
+from re import (
+    IGNORECASE,
+    MULTILINE,
+    UNICODE,
+    VERBOSE,
+    Pattern,
+)
+from typing import Any, Union
 
 TupleStr = tuple[str, ...]
 DictData = dict[str, Any]
 DictStr = dict[str, str]
+Matrix = dict[str, Union[list[str], list[int]]]
+MatrixInclude = list[dict[str, Union[str, int]]]
+MatrixExclude = list[dict[str, Union[str, int]]]
+
+
+class Re:
+    """Regular expression config."""
+
+    # NOTE: Search caller
+    __re_caller: str = r"""
+        \$
+        {{
+            \s*(?P<caller>
+                [a-zA-Z0-9_.\s'\"\[\]\(\)\-\{}]+?
+            )\s*
+        }}
+    """
+    RE_CALLER: Pattern = re.compile(
+        __re_caller, MULTILINE | IGNORECASE | UNICODE | VERBOSE
+    )
+
+    # NOTE: Search task
+    __re_task_fmt: str = r"""
+        ^
+            (?P<path>[^/@]+)
+            /
+            (?P<func>[^@]+)
+            @
+            (?P<tag>.+)
+        $
+    """
+    RE_TASK_FMT: Pattern = re.compile(
+        __re_task_fmt, MULTILINE | IGNORECASE | UNICODE | VERBOSE
+    )

ddeutil/workflow/exceptions.py

@@ -9,4 +9,16 @@ Define Errors Object for Node package
 from __future__ import annotations
 
 
-class TaskException(Exception): ...
+class WorkflowException(Exception): ...
+
+
+class UtilException(WorkflowException): ...
+
+
+class StageException(WorkflowException): ...
+
+
+class JobException(WorkflowException): ...
+
+
+class PipelineException(WorkflowException): ...

ddeutil/workflow/loader.py

@@ -6,46 +6,17 @@
 from __future__ import annotations
 
 from functools import cached_property
-from typing import Any, ClassVar, TypeVar
+from typing import TypeVar
 
-from ddeutil.core import (
-    getdot,
-    hasdot,
-    import_string,
-)
-from ddeutil.io import (
-    PathData,
-    PathSearch,
-    YamlEnvFl,
-)
-from pydantic import BaseModel, Field
-from pydantic.functional_validators import model_validator
+from ddeutil.core import import_string
+from ddeutil.io import PathSearch, YamlFlResolve
+from pydantic import BaseModel
 
-from .__regex import RegexConf
 from .__types import DictData
+from .utils import ConfParams, config
 
-T = TypeVar("T")
-BaseModelType = type[BaseModel]
 AnyModel = TypeVar("AnyModel", bound=BaseModel)
-
-
-class Engine(BaseModel):
-    """Engine Model"""
-
-    paths: PathData = Field(default_factory=PathData)
-    registry: list[str] = Field(default_factory=lambda: ["ddeutil.workflow"])
-
-    @model_validator(mode="before")
-    def __prepare_registry(cls, values: DictData) -> DictData:
-        if (_regis := values.get("registry")) and isinstance(_regis, str):
-            values["registry"] = [_regis]
-        return values
-
-
-class Params(BaseModel):
-    """Params Model"""
-
-    engine: Engine = Field(default_factory=Engine)
+AnyModelType = type[AnyModel]
 
 
 class SimLoad:
@@ -63,27 +34,25 @@ class SimLoad:
     def __init__(
         self,
         name: str,
-        params: Params,
+        params: ConfParams,
         externals: DictData,
     ) -> None:
         self.data: DictData = {}
         for file in PathSearch(params.engine.paths.conf).files:
             if any(file.suffix.endswith(s) for s in ("yml", "yaml")) and (
-                data := YamlEnvFl(file).read().get(name, {})
+                data := YamlFlResolve(file).read().get(name, {})
             ):
                 self.data = data
         if not self.data:
             raise ValueError(f"Config {name!r} does not found on conf path")
-        self.__conf_params: Params = params
+        self.conf_params: ConfParams = params
         self.externals: DictData = externals
 
-    @property
-    def conf_params(self) -> Params:
-        return self.__conf_params
-
     @cached_property
-    def type(self) -> BaseModelType:
-        """Return object type which implement in `config_object` key."""
+    def type(self) -> AnyModelType:
+        """Return object of string type which implement on any registry. The
+        object type
+        """
         if not (_typ := self.data.get("type")):
             raise ValueError(
                 f"the 'type' value: {_typ} does not exists in config data."
@@ -99,76 +68,13 @@ class SimLoad:
                     continue
             return import_string(f"{_typ}")
 
-    def load(self) -> AnyModel:
-        return self.type.model_validate(self.data)
-
 
 class Loader(SimLoad):
-    """Main Loader Object that get the config `yaml` file from current path.
+    """Loader Object that get the config `yaml` file from current path.
 
     :param name: A name of config data that will read by Yaml Loader object.
     :param externals: An external parameters
     """
 
-    conf_name: ClassVar[str] = "workflows-conf"
-
-    def __init__(
-        self,
-        name: str,
-        externals: DictData,
-        *,
-        path: str | None = None,
-    ) -> None:
-        self.data: DictData = {}
-
-        # NOTE: import params object from specific config file
-        params: Params = self.config(path)
-
-        super().__init__(name, params, externals)
-
-    @classmethod
-    def config(cls, path: str | None = None) -> Params:
-        """Load Config data from ``workflows-conf.yaml`` file."""
-        return Params.model_validate(
-            YamlEnvFl(path or f"./{cls.conf_name}.yaml").read()
-        )
-
-
-def map_params(value: Any, params: dict[str, Any]) -> Any:
-    """Map caller value that found from ``RE_CALLER`` regular expression.
-
-    :param value: A value that want to mapped with an params
-    :param params: A parameter value that getting with matched regular
-        expression.
-
-    :rtype: Any
-    :returns: An any getter value from the params input.
-    """
-    if isinstance(value, dict):
-        return {k: map_params(value[k], params) for k in value}
-    elif isinstance(value, (list, tuple, set)):
-        return type(value)([map_params(i, params) for i in value])
-    elif not isinstance(value, str):
-        return value
-
-    if not (found := RegexConf.RE_CALLER.search(value)):
-        return value
-
-    # NOTE: get caller value that setting inside; ``${{ <caller-value> }}``
-    caller: str = found.group("caller")
-    if not hasdot(caller, params):
-        raise ValueError(f"params does not set caller: {caller!r}")
-    getter: Any = getdot(caller, params)
-
-    # NOTE: check type of vars
-    if isinstance(getter, (str, int)):
-        return value.replace(found.group(0), str(getter))
-
-    # NOTE:
-    #   If type of getter caller does not formatting, it will return origin
-    #   value.
-    if value.replace(found.group(0), "") != "":
-        raise ValueError(
-            "Callable variable should not pass other outside ${{ ... }}"
-        )
-    return getter
+    def __init__(self, name: str, externals: DictData) -> None:
+        super().__init__(name, config(), externals)

ddeutil/workflow/on.py

@@ -0,0 +1,195 @@
+# ------------------------------------------------------------------------------
+# Copyright (c) 2022 Korawich Anuttra. All rights reserved.
+# Licensed under the MIT License. See LICENSE in the project root for
+# license information.
+# ------------------------------------------------------------------------------
+from __future__ import annotations
+
+from datetime import datetime
+from typing import Annotated, Literal
+from zoneinfo import ZoneInfo, ZoneInfoNotFoundError
+
+from pydantic import BaseModel, ConfigDict, Field
+from pydantic.functional_validators import field_validator, model_validator
+from typing_extensions import Self
+
+try:
+    from .__types import DictData, DictStr
+    from .loader import Loader
+    from .scheduler import WEEKDAYS, CronJob, CronJobYear, CronRunner
+except ImportError:
+    from ddeutil.workflow.__types import DictData, DictStr
+    from ddeutil.workflow.loader import Loader
+    from ddeutil.workflow.scheduler import (
+        WEEKDAYS,
+        CronJob,
+        CronJobYear,
+        CronRunner,
+    )
+
+
+def interval2crontab(
+    interval: Literal["daily", "weekly", "monthly"],
+    day: str | None = None,
+    time: str = "00:00",
+) -> str:
+    """Return the crontab string that was generated from specific values.
+
+    :param interval: A interval value that is one of 'daily', 'weekly', or
+        'monthly'.
+    :param day: A day value that will be day of week. The default value is
+        monday if it be weekly interval.
+    :param time: A time value that passing with format '%H:%M'.
+
+    Examples:
+        >>> interval2crontab(interval='daily', time='01:30')
+        '1 30 * * *'
+        >>> interval2crontab(interval='weekly', day='friday', time='18:30')
+        '18 30 * * 5'
+        >>> interval2crontab(interval='monthly', time='00:00')
+        '0 0 1 * *'
+        >>> interval2crontab(interval='monthly', day='tuesday', time='12:00')
+        '12 0 1 * 2'
+    """
+    d: str = "*"
+    if interval == "weekly":
+        d = WEEKDAYS[(day or "monday")[:3].title()]
+    elif interval == "monthly" and day:
+        d = WEEKDAYS[day[:3].title()]
+
+    h, m = tuple(
+        i.lstrip("0") if i != "00" else "0" for i in time.split(":", maxsplit=1)
+    )
+    return f"{h} {m} {'1' if interval == 'monthly' else '*'} * {d}"
+
+
+class On(BaseModel):
+    """On Model (Schedule)
+
+    See Also:
+        * ``generate()`` is the main usecase of this schedule object.
+    """
+
+    model_config = ConfigDict(arbitrary_types_allowed=True)
+
+    # NOTE: This is fields of the base schedule.
+    cronjob: Annotated[
+        CronJob,
+        Field(description="Cron job of this schedule"),
+    ]
+    tz: Annotated[
+        str,
+        Field(
+            description="A timezone string value",
+            alias="timezone",
+        ),
+    ] = "Etc/UTC"
+    extras: Annotated[
+        DictData,
+        Field(
+            default_factory=dict,
+            description="An extras mapping parameters",
+        ),
+    ]
+
+    @classmethod
+    def from_value(cls, value: DictStr, externals: DictData) -> Self:
+        """Constructor from values that will generate crontab by function.
+
+        :param value: A mapping value that will generate crontab before create
+            schedule model.
+        :param externals: A extras external parameter that will keep in extras.
+        """
+        passing: DictStr = {}
+        if "timezone" in value:
+            passing["tz"] = value.pop("timezone")
+        passing["cronjob"] = interval2crontab(
+            **{v: value[v] for v in value if v in ("interval", "day", "time")}
+        )
+        return cls(extras=externals, **passing)
+
+    @classmethod
+    def from_loader(
+        cls,
+        name: str,
+        externals: DictData,
+    ) -> Self:
+        """Constructor from the name of config that will use loader object for
+        getting the data.
+
+        :param name: A name of config that will getting from loader.
+        :param externals: A extras external parameter that will keep in extras.
+        """
+        loader: Loader = Loader(name, externals=externals)
+        # NOTE: Validate the config type match with current connection model
+        if loader.type != cls:
+            raise ValueError(f"Type {loader.type} does not match with {cls}")
+
+        loader_data: DictData = loader.data
+        if "interval" in loader_data:
+            return cls.model_validate(
+                obj=dict(
+                    cronjob=interval2crontab(
+                        **{
+                            v: loader_data[v]
+                            for v in loader_data
+                            if v in ("interval", "day", "time")
+                        }
+                    ),
+                    extras=externals,
+                    **loader_data,
+                )
+            )
+        if "cronjob" not in loader_data:
+            raise ValueError("Config does not set ``cronjob`` key")
+        return cls.model_validate(
+            obj=dict(
+                cronjob=loader_data.pop("cronjob"),
+                extras=externals,
+                **loader_data,
+            )
+        )
+
+    @model_validator(mode="before")
+    def __prepare_values(cls, values):
+        if tz := values.pop("tz", None):
+            values["timezone"] = tz
+        return values
+
+    @field_validator("tz")
+    def __validate_tz(cls, value: str):
+        """Validate timezone value that able to initialize with ZoneInfo after
+        it passing to this model in before mode."""
+        try:
+            _ = ZoneInfo(value)
+            return value
+        except ZoneInfoNotFoundError as err:
+            raise ValueError(f"Invalid timezone: {value}") from err
+
+    @field_validator("cronjob", mode="before")
+    def __prepare_cronjob(cls, value: str | CronJob) -> CronJob:
+        """Prepare crontab value that able to receive with string type."""
+        return CronJob(value) if isinstance(value, str) else value
+
+    def generate(self, start: str | datetime) -> CronRunner:
+        """Return Cron runner object."""
+        if not isinstance(start, datetime):
+            start: datetime = datetime.fromisoformat(start)
+        return self.cronjob.schedule(date=start, tz=self.tz)
+
+
+class AwsOn(On):
+    """Implement On AWS Schedule for AWS Service like AWS Glue."""
+
+    model_config = ConfigDict(arbitrary_types_allowed=True)
+
+    # NOTE: This is fields of the base schedule.
+    cronjob: Annotated[
+        CronJobYear,
+        Field(description="Cron job of this schedule"),
+    ]
+
+    @field_validator("cronjob", mode="before")
+    def __prepare_cronjob(cls, value: str | CronJobYear) -> CronJobYear:
+        """Prepare crontab value that able to receive with string type."""
+        return CronJobYear(value) if isinstance(value, str) else value