Types

zenml.types

Custom ZenML types.

Attributes

HookSpecification = Union[str, Source, FunctionType, Callable[..., None]] module-attribute

Classes

CSVString

Bases: str

Special string class to indicate a CSV string.

HTMLString

Bases: str

Special string class to indicate an HTML string.

JSONString

Bases: str

Special string class to indicate a JSON string.

MarkdownString

Bases: str

Special string class to indicate a Markdown string.

Source

Bases: BaseModel

Source specification.

A source specifies a module name as well as an optional attribute of that module. These values can be used to import the module and get the value of the attribute inside the module.

Example

The source Source(module="zenml.config.source", attribute="Source") references the class that this docstring is describing. This class is defined in the zenml.config.source module and the name of the attribute is the class name Source.

Attributes:

Name	Type	Description
module	str	The module name.
attribute	Optional[str]	Optional name of the attribute inside the module.
type	SourceType	The type of the source.

Attributes

import_path: str property

The import path of the source.

Returns:

Type	Description
str	The import path of the source.

is_internal: bool property

If the source is internal (=from the zenml package).

Returns:

Type	Description
bool	True if the source is internal, False otherwise

is_module_source: bool property

If the source is a module source.

Returns:

Type	Description
bool	If the source is a module source.

Functions

from_import_path(import_path: str, is_module_path: bool = False) -> Source classmethod

Creates a source from an import path.

Parameters:

Name	Type	Description	Default
import_path	str	The import path.	required
is_module_path	bool	If the import path points to a module or not.	False

Raises:

Type	Description
ValueError	If the import path is empty.

Returns:

Type	Description
Source	The source.

Source code in src/zenml/config/source.py

@classmethod
def from_import_path(
    cls, import_path: str, is_module_path: bool = False
) -> "Source":
    """Creates a source from an import path.

    Args:
        import_path: The import path.
        is_module_path: If the import path points to a module or not.

    Raises:
        ValueError: If the import path is empty.

    Returns:
        The source.
    """
    if not import_path:
        raise ValueError(
            "Invalid empty import path. The import path needs to refer "
            "to a Python module and an optional attribute of that module."
        )

    # Remove internal version pins for backwards compatibility
    if "@" in import_path:
        import_path = import_path.split("@", 1)[0]

    if is_module_path or "." not in import_path:
        module = import_path
        attribute = None
    else:
        module, attribute = import_path.rsplit(".", maxsplit=1)

    return Source(
        module=module, attribute=attribute, type=SourceType.UNKNOWN
    )

model_dump(**kwargs: Any) -> Dict[str, Any]

Dump the source as a dictionary.

Parameters:

Name	Type	Description	Default
**kwargs	Any	Additional keyword arguments.	{}

Returns:

Type	Description
Dict[str, Any]	The source as a dictionary.

Source code in src/zenml/config/source.py

def model_dump(self, **kwargs: Any) -> Dict[str, Any]:
    """Dump the source as a dictionary.

    Args:
        **kwargs: Additional keyword arguments.

    Returns:
        The source as a dictionary.
    """
    return super().model_dump(serialize_as_any=True, **kwargs)

model_dump_json(**kwargs: Any) -> str

Dump the source as a JSON string.

Parameters:

Name	Type	Description	Default
**kwargs	Any	Additional keyword arguments.	{}

Returns:

Type	Description
str	The source as a JSON string.

Source code in src/zenml/config/source.py

def model_dump_json(self, **kwargs: Any) -> str:
    """Dump the source as a JSON string.

    Args:
        **kwargs: Additional keyword arguments.

    Returns:
        The source as a JSON string.
    """
    return super().model_dump_json(serialize_as_any=True, **kwargs)