ob-metaflow 2.12.36.3__py2.py3-none-any.whl → 2.13.0.1__py2.py3-none-any.whl

metaflow/user_configs/config_parameters.py

@@ -0,0 +1,405 @@
+import collections.abc
+import json
+import os
+import re
+
+from typing import Any, Callable, Dict, Optional, TYPE_CHECKING, Union
+
+
+from ..exception import MetaflowException
+
+from ..parameters import (
+    Parameter,
+    ParameterContext,
+    current_flow,
+)
+
+if TYPE_CHECKING:
+    from metaflow import FlowSpec
+
+# _tracefunc_depth = 0
+
+
+# def tracefunc(func):
+#     """Decorates a function to show its trace."""
+
+#     @functools.wraps(func)
+#     def tracefunc_closure(*args, **kwargs):
+#         global _tracefunc_depth
+#         """The closure."""
+#         print(f"{_tracefunc_depth}: {func.__name__}(args={args}, kwargs={kwargs})")
+#         _tracefunc_depth += 1
+#         result = func(*args, **kwargs)
+#         _tracefunc_depth -= 1
+#         print(f"{_tracefunc_depth} => {result}")
+#         return result
+
+#     return tracefunc_closure
+
+CONFIG_FILE = os.path.join(
+    os.path.dirname(os.path.abspath(__file__)), "CONFIG_PARAMETERS"
+)
+
+ID_PATTERN = re.compile(r"^[a-zA-Z_][a-zA-Z0-9_]*$")
+
+UNPACK_KEY = "_unpacked_delayed_"
+
+
+def dump_config_values(flow: "FlowSpec"):
+    from ..flowspec import _FlowState  # Prevent circular import
+
+    configs = flow._flow_state.get(_FlowState.CONFIGS)
+    if configs:
+        return {"user_configs": configs}
+    return {}
+
+
+class ConfigValue(collections.abc.Mapping):
+    """
+    ConfigValue is a thin wrapper around an arbitrarily nested dictionary-like
+    configuration object. It allows you to access elements of this nested structure
+    using either a "." notation or a [] notation. As an example, if your configuration
+    object is:
+    {"foo": {"bar": 42}}
+    you can access the value 42 using either config["foo"]["bar"] or config.foo.bar.
+
+    All "keys"" need to be valid Python identifiers
+    """
+
+    # Thin wrapper to allow configuration values to be accessed using a "." notation
+    # as well as a [] notation.
+
+    def __init__(self, data: Dict[str, Any]):
+        self._data = data
+
+    def __getattr__(self, key: str) -> Any:
+        """
+        Access an element of this configuration
+
+        Parameters
+        ----------
+        key : str
+            Element to access
+
+        Returns
+        -------
+        Any
+            Element of the configuration
+        """
+        if key == "_data":
+            # Called during unpickling. Special case to not run into infinite loop
+            # below.
+            raise AttributeError(key)
+
+        if key in self._data:
+            return self[key]
+        raise AttributeError(key)
+
+    def __setattr__(self, name: str, value: Any) -> None:
+        # Prevent configuration modification
+        if name == "_data":
+            return super().__setattr__(name, value)
+        raise TypeError("ConfigValue is immutable")
+
+    def __getitem__(self, key: Any) -> Any:
+        """
+        Access an element of this configuration
+
+        Parameters
+        ----------
+        key : Any
+            Element to access
+
+        Returns
+        -------
+        Any
+            Element of the configuration
+        """
+        value = self._data[key]
+        if isinstance(value, dict):
+            value = ConfigValue(value)
+        return value
+
+    def __len__(self):
+        return len(self._data)
+
+    def __iter__(self):
+        return iter(self._data)
+
+    def __repr__(self):
+        return repr(self._data)
+
+    def __str__(self):
+        return json.dumps(self._data)
+
+    def to_dict(self) -> Dict[Any, Any]:
+        """
+        Returns a dictionary representation of this configuration object.
+
+        Returns
+        -------
+        Dict[Any, Any]
+            Dictionary equivalent of this configuration object.
+        """
+        return dict(self._data)
+
+
+class DelayEvaluator(collections.abc.Mapping):
+    """
+    Small wrapper that allows the evaluation of a Config() value in a delayed manner.
+    This is used when we want to use config.* values in decorators for example.
+
+    It also allows the following "delayed" access on an obj that is a DelayEvaluation
+      - obj.x.y.z (ie: accessing members of DelayEvaluator; accesses will be delayed until
+        the DelayEvaluator is evaluated)
+      - **obj (ie: unpacking the DelayEvaluator as a dictionary). Note that this requires
+        special handling in whatever this is being unpacked into, specifically the handling
+        of _unpacked_delayed_*
+    """
+
+    def __init__(self, ex: str):
+        self._config_expr = ex
+        if ID_PATTERN.match(self._config_expr):
+            # This is a variable only so allow things like config_expr("config").var
+            self._is_var_only = True
+            self._access = []
+        else:
+            self._is_var_only = False
+            self._access = None
+
+    def __iter__(self):
+        yield "%s%d" % (UNPACK_KEY, id(self))
+
+    def __getitem__(self, key):
+        if key == "%s%d" % (UNPACK_KEY, id(self)):
+            return self
+        if self._access is None:
+            raise KeyError(key)
+        self._access.append(key)
+        return self
+
+    def __len__(self):
+        return 1
+
+    def __getattr__(self, name):
+        if self._access is None:
+            raise AttributeError()
+        self._access.append(name)
+        return self
+
+    def __call__(self, ctx=None, deploy_time=False):
+        from ..flowspec import _FlowState  # Prevent circular import
+
+        # Two additional arguments are only used by DeployTimeField which will call
+        # this function with those two additional arguments. They are ignored.
+        flow_cls = getattr(current_flow, "flow_cls", None)
+        if flow_cls is None:
+            # We are not executing inside a flow (ie: not the CLI)
+            raise MetaflowException(
+                "Config object can only be used directly in the FlowSpec defining them. "
+                "If using outside of the FlowSpec, please use ConfigEval"
+            )
+        if self._access is not None:
+            # Build the final expression by adding all the fields in access as . fields
+            self._config_expr = ".".join([self._config_expr] + self._access)
+        # Evaluate the expression setting the config values as local variables
+        try:
+            return eval(
+                self._config_expr,
+                globals(),
+                {
+                    k: ConfigValue(v)
+                    for k, v in flow_cls._flow_state.get(_FlowState.CONFIGS, {}).items()
+                },
+            )
+        except NameError as e:
+            potential_config_name = self._config_expr.split(".")[0]
+            if potential_config_name not in flow_cls._flow_state.get(
+                _FlowState.CONFIGS, {}
+            ):
+                raise MetaflowException(
+                    "Config '%s' not found in the flow (maybe not required and not "
+                    "provided?)" % potential_config_name
+                ) from e
+            raise
+
+
+def config_expr(expr: str) -> DelayEvaluator:
+    """
+    Function to allow you to use an expression involving a config parameter in
+    places where it may not be directory accessible or if you want a more complicated
+    expression than just a single variable.
+
+    You can use it as follows:
+      - When the config is not directly accessible:
+
+            @project(name=config_expr("config").project.name)
+            class MyFlow(FlowSpec):
+                config = Config("config")
+                ...
+      - When you want a more complex expression:
+            class MyFlow(FlowSpec):
+                config = Config("config")
+
+                @environment(vars={"foo": config_expr("config.bar.baz.lower()")})
+                @step
+                def start(self):
+                    ...
+
+    Parameters
+    ----------
+    expr : str
+        Expression using the config values.
+    """
+    return DelayEvaluator(expr)
+
+
+class Config(Parameter, collections.abc.Mapping):
+    """
+    Includes a configuration for this flow.
+
+    `Config` is a special type of `Parameter` but differs in a few key areas:
+      - it is immutable and determined at deploy time (or prior to running if not deploying
+        to a scheduler)
+      - as such, it can be used anywhere in your code including in Metaflow decorators
+
+    The value of the configuration is determines as follows:
+      - use the user-provided file path or value. It is an error to provide both
+      - if none are present:
+        - if a default file path (default) is provided, attempt to read this file
+            - if the file is present, use that value. Note that the file will be used
+              even if it has an invalid syntax
+            - if the file is not present, and a default value is present, use that
+      - if still None and is required, this is an error.
+
+    Parameters
+    ----------
+    name : str
+        User-visible configuration name.
+    default : Union[str, Callable[[ParameterContext], str], optional, default None
+        Default path from where to read this configuration. A function implies that the
+        value will be computed using that function.
+        You can only specify default or default_value.
+    default_value : Union[str, Dict[str, Any], Callable[[ParameterContext, Union[str, Dict[str, Any]]], Any], optional, default None
+        Default value for the parameter. A function
+        implies that the value will be computed using that function.
+        You can only specify default or default_value.
+    help : str, optional, default None
+        Help text to show in `run --help`.
+    required : bool, optional, default None
+        Require that the user specified a value for the configuration. Note that if
+        a default is provided, the required flag is ignored. A value of None is
+        equivalent to False.
+    parser : Union[str, Callable[[str], Dict[Any, Any]]], optional, default None
+        If a callable, it is a function that can parse the configuration string
+        into an arbitrarily nested dictionary. If a string, the string should refer to
+        a function (like "my_parser_package.my_parser.my_parser_function") which should
+        be able to parse the configuration string into an arbitrarily nested dictionary.
+        If the name starts with a ".", it is assumed to be relative to "metaflow".
+    show_default : bool, default True
+        If True, show the default value in the help text.
+    """
+
+    IS_CONFIG_PARAMETER = True
+
+    def __init__(
+        self,
+        name: str,
+        default: Optional[Union[str, Callable[[ParameterContext], str]]] = None,
+        default_value: Optional[
+            Union[
+                str,
+                Dict[str, Any],
+                Callable[[ParameterContext], Union[str, Dict[str, Any]]],
+            ]
+        ] = None,
+        help: Optional[str] = None,
+        required: Optional[bool] = None,
+        parser: Optional[Union[str, Callable[[str], Dict[Any, Any]]]] = None,
+        **kwargs: Dict[str, str]
+    ):
+
+        if default and default_value:
+            raise MetaflowException(
+                "For config '%s', you can only specify default or default_value, not both"
+                % name
+            )
+        self._default_is_file = default is not None
+        kwargs["default"] = default or default_value
+        super(Config, self).__init__(
+            name, required=required, help=help, type=str, **kwargs
+        )
+        super(Config, self).init()
+
+        if isinstance(kwargs.get("default", None), str):
+            kwargs["default"] = json.dumps(kwargs["default"])
+        self.parser = parser
+        self._computed_value = None
+
+    def load_parameter(self, v):
+        if v is None:
+            return None
+        return ConfigValue(v)
+
+    def _store_value(self, v: Any) -> None:
+        self._computed_value = v
+
+    # Support <config>.<var> syntax
+    def __getattr__(self, name):
+        return DelayEvaluator(self.name.lower()).__getattr__(name)
+
+    # Next three methods are to implement mapping to support **<config> syntax
+    def __iter__(self):
+        return iter(DelayEvaluator(self.name.lower()))
+
+    def __len__(self):
+        return len(DelayEvaluator(self.name.lower()))
+
+    def __getitem__(self, key):
+        return DelayEvaluator(self.name.lower())[key]
+
+
+def resolve_delayed_evaluator(v: Any, ignore_errors: bool = False) -> Any:
+    try:
+        if isinstance(v, DelayEvaluator):
+            return v()
+        if isinstance(v, dict):
+            return {
+                resolve_delayed_evaluator(k): resolve_delayed_evaluator(v)
+                for k, v in v.items()
+            }
+        if isinstance(v, list):
+            return [resolve_delayed_evaluator(x) for x in v]
+        if isinstance(v, tuple):
+            return tuple(resolve_delayed_evaluator(x) for x in v)
+        if isinstance(v, set):
+            return {resolve_delayed_evaluator(x) for x in v}
+        return v
+    except Exception as e:
+        if ignore_errors:
+            # Assumption is that default value of None is always allowed.
+            # This code path is *only* used when evaluating Parameters AND they
+            # use configs in their attributes AND the runner/deployer is being used
+            # AND CLICK_API_PROCESS_CONFIG is False. In those cases, all attributes in
+            # Parameter can be set to None except for required and show_default
+            # and even in those cases, a wrong value will have very limited consequence.
+            return None
+        raise e
+
+
+def unpack_delayed_evaluator(
+    to_unpack: Dict[str, Any], ignore_errors: bool = False
+) -> Dict[str, Any]:
+    result = {}
+    for k, v in to_unpack.items():
+        if not isinstance(k, str) or not k.startswith(UNPACK_KEY):
+            result[k] = v
+        else:
+            # k.startswith(UNPACK_KEY)
+            try:
+                result.update(resolve_delayed_evaluator(v[k]))
+            except Exception as e:
+                if ignore_errors:
+                    continue
+                raise e
+    return result

metaflow/util.py

@@ -296,6 +296,9 @@ def get_metaflow_root():
 
 
 def dict_to_cli_options(params):
+    # Prevent circular imports
+    from .user_configs.config_options import ConfigInput
+
     for k, v in params.items():
         # Omit boolean options set to false or None, but preserve options with an empty
         # string argument.
@@ -304,6 +307,20 @@ def dict_to_cli_options(params):
             # keyword in Python, so we call it 'decospecs' in click args
             if k == "decospecs":
                 k = "with"
+            if k in ("config_file", "config_value"):
+                # Special handling here since we gather them all in one option but actually
+                # need to send them one at a time using --config-value <name> kv.<name>
+                # Note it can be either config_file or config_value depending
+                # on click processing order.
+                for config_name in v.keys():
+                    yield "--config-value"
+                    yield to_unicode(config_name)
+                    yield to_unicode(ConfigInput.make_key_name(config_name))
+                continue
+            if k == "local_config_file":
+                # Skip this value -- it should only be used locally and never when
+                # forming another command line
+                continue
             k = k.replace("_", "-")
             v = v if isinstance(v, (list, tuple, set)) else [v]
             for value in v:

metaflow/version.py

@@ -1 +1 @@
-metaflow_version = "2.12.36.3"
+metaflow_version = "2.13.0.1"

{ob_metaflow-2.12.36.3.dist-info → ob_metaflow-2.13.0.1.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: ob-metaflow
-Version: 2.12.36.3
+Version: 2.13.0.1
 Summary: Metaflow: More Data Science, Less Engineering
 Author: Netflix, Outerbounds & the Metaflow Community
 Author-email: help@outerbounds.co
@@ -12,7 +12,7 @@ Requires-Dist: boto3
 Requires-Dist: pylint
 Requires-Dist: kubernetes
 Provides-Extra: stubs
-Requires-Dist: metaflow-stubs==2.12.36.3; extra == "stubs"
+Requires-Dist: metaflow-stubs==2.13.0.1; extra == "stubs"
 
 ![Metaflow_Logo_Horizontal_FullColor_Ribbon_Dark_RGB](https://user-images.githubusercontent.com/763451/89453116-96a57e00-d713-11ea-9fa6-82b29d4d6eff.png)
 
@@ -83,3 +83,4 @@ There are several ways to get in touch with us:
 
 ## Contributing
 We welcome contributions to Metaflow. Please see our [contribution guide](https://docs.metaflow.org/introduction/contributing-to-metaflow) for more details.
+