contract.csv

CsvParams dataclass

CsvParams(
    path: PathLike,
    delimiter: Optional[str] = None,
    strict_header: bool = True,
    index: Optional[str] = None,
)

Bases: FilePathBaseParam

Parameters for CSV file processing.

Extends the base file path parameters with CSV-specific options.

Attributes:

Name	Type	Description
delimiter	Optional[str]	Custom delimiter character for CSV parsing. If None, the default comma delimiter is used.
strict_header	bool	If True, treats the first row as a header. Otherwise, no header is assumed.
index	Optional[str]	Column name to set as the DataFrame index. If None, default numeric indices are used.

Csv

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

Bases: DataStream[DataFrame, CsvParams]

CSV file data stream provider.

A data stream implementation for reading CSV files into pandas DataFrames with configurable parameters for delimiter, header handling, and indexing.

Parameters:

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

Examples:

from contraqctor.contract.csv import Csv, CsvParams

# Create and load a CSV stream
params = CsvParams(path="data/european_data.csv", delimiter=";")
csv_stream = Csv("measurements", reader_params=params)
csv_stream.load()

# Access the DataFrame
df = csv_stream.data
filtered = df[df["temperature"] > 25]

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

def __init__(
    self: Self,
    name: str,
    *,
    description: Optional[str] = None,
    reader_params: _typing.TReaderParams = _typing.UnsetParams,
    **kwargs,
) -> None:
    if "::" in name:
        raise ValueError("Name cannot contain '::' character.")
    self._name = name

    self._description = description
    self._reader_params = reader_params
    self._data = _typing.UnsetData
    self._parent: Optional["DataStream"] = None

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.

at property

at: _AtProtocol

Get a child data stream by name.

Parameters:

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

Returns:

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

Raises:

Type	Description
NotImplementedError	If the data stream does not support child access.

Examples:

# Access stream in a collection
collection = data_collection.load()
temp_stream = collection.at("temperature")

# Or using dictionary-style syntax
humidity_stream = collection["humidity"]

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.

has_error property

has_error: bool

Check if the data stream encountered an error during loading.

Returns:

Name	Type	Description
bool	bool	True if an error occurred, 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.

set_parent

set_parent(parent: DataStream) -> None

Set the parent data stream.

Parameters:

Name	Type	Description	Default
parent	DataStream	The parent data stream to set.	required

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

def set_parent(self, parent: "DataStream") -> None:
    """Set the parent data stream.

    Args:
        parent: The parent data stream to set.
    """
    self._parent = parent

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

clear

clear() -> Self

Clear the loaded data from the data stream.

Resets the data to an unset state, allowing for reloading.

Returns:

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

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

def clear(self) -> Self:
    """Clear the loaded data from the data stream.

    Resets the data to an unset state, allowing for reloading.

    Returns:
        Self: The data stream instance for method chaining.
    """
    self._data = _typing.UnsetData
    return self

load

load() -> Self

Load data into the data stream.

Reads data from the source and stores it in the data stream.

Returns:

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

Examples:

from contraqctor.contract import csv

# Create and load a CSV stream
params = csv.CsvParams(path="data/measurements.csv")
csv_stream = csv.Csv("measurements", reader_params=params)
csv_stream.load()

# Access the data
df = csv_stream.data
print(f"Loaded {len(df)} rows")

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

def load(self) -> Self:
    """Load data into the data stream.

    Reads data from the source and stores it in the data stream.

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

    Examples:
        ```python
        from contraqctor.contract import csv

        # Create and load a CSV stream
        params = csv.CsvParams(path="data/measurements.csv")
        csv_stream = csv.Csv("measurements", reader_params=params)
        csv_stream.load()

        # Access the data
        df = csv_stream.data
        print(f"Loaded {len(df)} rows")
        ```
    """
    try:
        self._data = self.read()
    except Exception as e:  # pylint: disable=broad-except
        self._data = _typing.ErrorOnLoad(self, exception=e)
    return self

collect_errors

collect_errors() -> List[ErrorOnLoad]

Collect all errors from this stream and its children.

Performs a depth-first traversal to gather all ErrorOnLoad instances.

Returns:

Type	Description
List[ErrorOnLoad]	List[ErrorOnLoad]: List of all errors raised on load encountered in the hierarchy.

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

def collect_errors(self) -> List[_typing.ErrorOnLoad]:
    """Collect all errors from this stream and its children.

    Performs a depth-first traversal to gather all ErrorOnLoad instances.

    Returns:
        List[ErrorOnLoad]: List of all errors raised on load encountered in the hierarchy.
    """
    errors = []
    if self.has_error:
        errors.append(cast(_typing.ErrorOnLoad, self._data))
    for stream in self:
        if stream is None:
            continue
        errors.extend(stream.collect_errors())
    return errors

load_all

load_all(strict: bool = False) -> Self

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	Self	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) -> Self:
    """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()
    for stream in self:
        if stream is None:
            continue
        stream.load_all(strict=strict)
        if stream.has_error and strict:
            cast(_typing.ErrorOnLoad, stream.data).raise_from_error()
    return self