contract.mux

MapFromPathsParams dataclass

MapFromPathsParams(
    paths: List[PathLike],
    include_glob_pattern: List[str],
    inner_data_stream: Type[_TDataStream],
    inner_param_factory: Callable[[str], TReaderParams],
    as_collection: bool = True,
    exclude_glob_pattern: List[str] = list(),
    inner_descriptions: dict[str, Optional[str]] = dict(),
)

Bases: Generic[_TDataStream]

Parameters for creating multiple data streams from file paths.

Defines parameters for locating files and creating data streams for each one.

Attributes:

Name	Type	Description
paths	List[PathLike]	List of directory paths to search for files.
include_glob_pattern	List[str]	List of glob patterns to match files to include.
inner_data_stream	Type[_TDataStream]	Type of DataStream to create for each matched file.
inner_param_factory	Callable[[str], TReaderParams]	Function that creates reader params from file paths.
as_collection	bool	Whether to return results as a collection. Defaults to True.
exclude_glob_pattern	List[str]	List of glob patterns for files to exclude.
inner_descriptions	dict[str, Optional[str]]	Dictionary mapping file stems to descriptions for streams.

MapFromPaths

MapFromPaths(
    name: str,
    *,
    description: Optional[str] = None,
    reader_params: Optional[TReaderParams] = None,
    **kwargs,
)

Bases: DataStreamCollectionBase[_TDataStream, MapFromPathsParams]

File path mapper data stream provider.

A data stream implementation for creating multiple child data streams by searching for files matching glob patterns and creating a stream for each.

Parameters:

Name	Type	Description	Default
DataStreamCollectionBase		Base class for data stream collection providers.	required

Examples:

from contraqctor.contract import mux, text

# Define a factory function for TextParams
def create_text_params(file_path):
    return text.TextParams(path=file_path)

# Create and load a text file collection
params = mux.MapFromPathsParams(
    paths=["documents/"],
    include_glob_pattern=["*.txt"],
    inner_data_stream=text.Text,
    inner_param_factory=create_text_params
)

docs = mux.MapFromPaths("documents", reader_params=params).load()
readme = docs["readme"].data

Source code in src/contraqctor/contract/base.py

def __init__(
    self: Self,
    name: str,
    *,
    description: Optional[str] = None,
    reader_params: Optional[_typing.TReaderParams] = None,
    **kwargs,
) -> None:
    super().__init__(name=name, description=description, reader_params=reader_params, **kwargs)
    self._hashmap: Dict[str, TDataStream] = {}
    self._update_hashmap()

name property

name: str

Get the name of the data stream.

Returns:

Name	Type	Description
str	str	Name identifier of the data stream.

resolved_name property

resolved_name: str

Get the full hierarchical name of the data stream.

Generates a path-like name showing the stream's position in the hierarchy, using '::' as a separator between parent and child names.

Returns:

Name	Type	Description
str	str	The fully resolved name including all parent names.

description property

description: Optional[str]

Get the description of the data stream.

Returns:

Type	Description
Optional[str]	Optional[str]: Description of the data stream, or None if not provided.

parent property

parent: Optional[DataStream]

Get the parent data stream.

Returns:

Type	Description
Optional[DataStream]	Optional[DataStream]: Parent data stream, or None if this is a root stream.

is_collection property

is_collection: bool

Check if this data stream is a collection of other streams.

Returns:

Name	Type	Description
bool	bool	True if this is a collection stream, False otherwise.

reader_params property

reader_params: TReaderParams

Get the parameters for the data reader.

Returns:

Name	Type	Description
TReaderParams	TReaderParams	Parameters for the data reader.

has_data property

has_data: bool

Check if the data stream has loaded data.

Returns:

Name	Type	Description
bool	bool	True if data has been loaded, False otherwise.

data property

data: TData

Get the loaded data.

Returns:

Name	Type	Description
TData	TData	The loaded data.

Raises:

Type	Description
ValueError	If data has not been loaded yet.

read

read(
    reader_params: Optional[TReaderParams] = None,
) -> TData

Read data using the configured reader.

Parameters:

Name	Type	Description	Default
reader_params	Optional[TReaderParams]	Optional parameters to override the default reader parameters.	None

Returns:

Name	Type	Description
TData	TData	Data read from the source.

Raises:

Type	Description
ValueError	If reader parameters are not set.

Source code in src/contraqctor/contract/base.py

def read(self, reader_params: Optional[_typing.TReaderParams] = None) -> _typing.TData:
    """Read data using the configured reader.

    Args:
        reader_params: Optional parameters to override the default reader parameters.

    Returns:
        TData: Data read from the source.

    Raises:
        ValueError: If reader parameters are not set.
    """
    reader_params = reader_params if reader_params is not None else self._reader_params
    if _typing.is_unset(reader_params):
        raise ValueError("Reader parameters are not set. Cannot read data.")
    return self._reader(reader_params)

bind_reader_params

bind_reader_params(params: TReaderParams) -> Self

Bind reader parameters to the data stream.

Parameters:

Name	Type	Description	Default
params	TReaderParams	Parameters to bind to the data stream's reader.	required

Returns:

Name	Type	Description
Self	Self	The data stream instance for method chaining.

Raises:

Type	Description
ValueError	If reader parameters have already been set.

Source code in src/contraqctor/contract/base.py

def bind_reader_params(self, params: _typing.TReaderParams) -> Self:
    """Bind reader parameters to the data stream.

    Args:
        params: Parameters to bind to the data stream's reader.

    Returns:
        Self: The data stream instance for method chaining.

    Raises:
        ValueError: If reader parameters have already been set.
    """
    if not _typing.is_unset(self._reader_params):
        raise ValueError("Reader parameters are already set. Cannot bind again.")
    self._reader_params = params
    return self

at

at(name: str) -> TDataStream

Get a child data stream by name.

Parameters:

Name	Type	Description	Default
name	str	Name of the child data stream to retrieve.	required

Returns:

Name	Type	Description
TDataStream	TDataStream	The child data stream with the given name.

Raises:

Type	Description
ValueError	If data has not been loaded yet.
KeyError	If no child stream with the given name exists.

Source code in src/contraqctor/contract/base.py

@override
def at(self, name: str) -> TDataStream:
    """Get a child data stream by name.

    Args:
        name: Name of the child data stream to retrieve.

    Returns:
        TDataStream: The child data stream with the given name.

    Raises:
        ValueError: If data has not been loaded yet.
        KeyError: If no child stream with the given name exists.
    """
    if not self.has_data:
        raise ValueError("data streams have not been read yet. Cannot access data streams.")
    if name in self._hashmap:
        return self._hashmap[name]
    else:
        raise KeyError(f"Stream with name: '{name}' not found in data streams.")

load

load()

Load data for this collection.

Overrides the base method to add validation that loaded data is a list of DataStreams.

Returns:

Name	Type	Description
Self		The collection instance for method chaining.

Raises:

Type	Description
ValueError	If loaded data is not a list of DataStreams.

Source code in src/contraqctor/contract/base.py

@override
def load(self):
    """Load data for this collection.

    Overrides the base method to add validation that loaded data is a list of DataStreams.

    Returns:
        Self: The collection instance for method chaining.

    Raises:
        ValueError: If loaded data is not a list of DataStreams.
    """
    super().load()
    if not isinstance(self._data, list):
        self._data = _typing.UnsetData
        raise ValueError("Data must be a list of DataStreams.")
    self._update_hashmap()
    return self

load_all

load_all(
    strict: bool = False,
) -> list[tuple[DataStream, Exception], None, None]

Recursively load this data stream and all child streams.

Performs depth-first traversal to load all streams in the hierarchy.

Parameters:

Name	Type	Description	Default
strict	bool	If True, raises exceptions immediately; otherwise collects and returns them.	False

Returns:

Name	Type	Description
list	list[tuple[DataStream, Exception], None, None]	List of tuples containing streams and exceptions that occurred during loading.

Raises:

Type	Description
Exception	If strict is True and an exception occurs during loading.

Examples:

# Load all streams and handle errors
errors = collection.load_all(strict=False)

if errors:
    for stream, error in errors:
        print(f"Error loading {stream.name}: {error}")

Source code in src/contraqctor/contract/base.py

def load_all(self, strict: bool = False) -> list[tuple["DataStream", Exception], None, None]:
    """Recursively load this data stream and all child streams.

    Performs depth-first traversal to load all streams in the hierarchy.

    Args:
        strict: If True, raises exceptions immediately; otherwise collects and returns them.

    Returns:
        list: List of tuples containing streams and exceptions that occurred during loading.

    Raises:
        Exception: If strict is True and an exception occurs during loading.

    Examples:
        ```python
        # Load all streams and handle errors
        errors = collection.load_all(strict=False)

        if errors:
            for stream, error in errors:
                print(f"Error loading {stream.name}: {error}")
        ```
    """
    self.load()
    exceptions = []
    for stream in self:
        if stream is None:
            continue
        try:
            exceptions += stream.load_all(strict=strict)
        except Exception as e:
            if strict:
                raise e
            exceptions.append((stream, e))
    return exceptions

iter_all

iter_all() -> Generator[DataStream, None, None]

Iterator for all child data streams, including nested collections.

Implements a depth-first traversal of the stream hierarchy.

Yields:

Name	Type	Description
DataStream	DataStream	All recursively yielded child data streams.

Source code in src/contraqctor/contract/base.py

def iter_all(self) -> Generator[DataStream, None, None]:
    """Iterator for all child data streams, including nested collections.

    Implements a depth-first traversal of the stream hierarchy.

    Yields:
        DataStream: All recursively yielded child data streams.
    """
    for value in self:
        if isinstance(value, DataStream):
            yield value
        if isinstance(value, DataStreamCollectionBase):
            yield from value.iter_all()