Attention

You’re viewing an old version of the L’ART Research Assistant documentation. The project has moved to https://lart.readthedocs.io/projects/research-assistant.

research_client.datavalidator.schemas

Prototypes to conveniently define data classes with built-in validation.

Classes

`CField`	Defines a manually validated data field for DataSchema classes.
`CFieldList`	Defines a field containing an arbitrary number of CField data.
`DataField`	Base class for data schema fields.
`DataFieldList`	Defines a field containing an arbitrary number of DataField data.
`DataGroup`	Defines a data group for DataSchema classes.
`DataSchema`	Abstract base class to define auto-validating data classes.
`VField`	Defines an auto-validated data field for DataSchema classes.
`VFieldList`	Defines a field containing an arbitrary number of VField data.

class CField[source]

Bases: DataField

Defines a manually validated data field for DataSchema classes.

__init__(name, type_, typedesc, vmethod, forcecast=None, required=True)[source]

Instantiates a new CField.

Parameters:

name (str)
type_ (str)
typedesc (str)
vmethod (Callable[[Any, Any], Any])
forcecast (Optional[bool])
required (bool)

_fieldparams = [('name', <class 'str'>, 'The name of the DataField'), ('type_', typing.Any, 'The data type for the DataField'), ('typedesc', <class 'str'>, 'A user-intelligible description of the data type'), ('vmethod', typing.Callable[[typing.Any, typing.Any], typing.Any], 'A callable accepting a two arguments: the first is the type_ of the DataField and the second the value. The callable should raise either a TypeError or a DataValidationError if the value is invalid, and must return a (possibly processed) version of the value it was passed which fits the type_ it was passed.'), ('forcecast', typing.Optional[bool], 'Whether to force casting of data during validation', None), ('required', <class 'bool'>, 'Whether the field is required', True)]: Type: list[Union[tuple[str, Any, str], tuple[str, Any, str, Any]]]

forcecast: Type: Optional[bool]

vmethod: Type: Union[str, Callable[[Any, Any], Any]]

class CFieldList[source]

Bases: CField, DataFieldList

Defines a field containing an arbitrary number of CField data.

class DataField[source]

Bases: object

Base class for data schema fields.

To instantiate, use VField, CField, or shorthand notation in a DataSchema declaration.

__init__(name, type_, typedesc, required=True)[source]

Instantiates a new DataField.

Parameters:

name (str)
type_ (Any)
typedesc (str)
required (bool)

classmethod fieldparams()[source]

Returns the parameter list for a DataField of this type.

Return type:: list[Union[tuple[str, Any, str], tuple[str, Any, str, Any]]]

fieldspecs()[source]

Returns the values for each parameter of the DataField.

Return type:: dict[str, Any]

_fieldparams = [('name', <class 'str'>, 'The name of the DataField'), ('type_', typing.Any, 'The data type for the DataField'), ('typedesc', <class 'str'>, 'A user-intelligible description of the data type'), ('required', <class 'bool'>, 'Whether the field is required', True)]: Type: list[Union[tuple[str, Any, str], tuple[str, Any, str, Any]]]

name: Type: str

required: Type: bool

type_: Type: Any

typedesc: Type: str

class DataFieldList[source]

Bases: DataField

Defines a field containing an arbitrary number of DataField data.

class DataGroup[source]

Bases: object

Defines a data group for DataSchema classes.

__init__(name, fields)[source]

Instantiates a new DataGroup.

Parameters:

name (str)
fields (dict[str, Union[DataField, dict[str, Any]]])

getfield(key)[source]

Gets the data field with the name indicated by key.

Return type:: Union[dict[str, Any], DataField]
Parameters:: key (str)

class DataSchema[source]

Bases: object

Abstract base class to define auto-validating data classes.

@TODO: - __getattr__(self, name) - to retreive data

__setattr__(self, name, value) - to set data (with validation)
__delattr__(self, name) - to remove/clear a datapoint
Move some __new__ stuff to __init_subclass__(cls)?
JSON import/export

__delfieldfactory(fieldname, fieldspecs)

Return type:

Callable[[DataSchema], None]

Parameters:

fieldname (str)
fieldspecs (dict[str, Any])

__delfieldlistfactory(fieldname, fieldspec)

Return type:

Callable[[DataSchema], None]

Parameters:

fieldname (str)
fieldspec (dict[str, Any])

__delgroupfactory(gname, fieldspecs)

Return type:

Callable[[DataSchema], None]

Parameters:

gname (str)
fieldspecs (dict[str, dict[str, Any]])

classmethod __functionalize()

Dynamically creates and attaches methods to get/set values.

Return type:: None

__getfieldfactory(fieldname, fieldspecs)

Return type:

Callable[[DataSchema], Any]

Parameters:

fieldname (str)
fieldspecs (dict[str, Any])

__getfieldlistfactory(fieldname, fieldspec)

Return type:

Callable[[DataSchema], list[Any]]

Parameters:

fieldname (str)
fieldspec (dict[str, Any])

classmethod __getfieldspecs(key)

Get the specifications for a single DataField or fields in a DataGroup.

Return type:: dict[str, dict[str, Any]]
Parameters:: key (str)

__getgroupfactory(gname, fieldspecs)

Return type:

Callable[[DataSchema], dict[str, Any]]

Parameters:

gname (str)
fieldspecs (dict[str, dict[str, Any]])

classmethod __index()

Creates a flat list of index keys, separating groups and fields with “/”.

Return type:: list[str]

__init__(forcecast=True, ignorecase=True)[source]

Initialises a new DataSchema object.

Parameters:

forcecast (bool)
ignorecase (bool)

classmethod __materialize()

Creates an empty __data store pre-populated with DataGroups/DataFieldLists.

Return type:: dict[str, Union[dict[str, Any], Any]]

static __new__(cls, *args, **kwargs)[source]

Constructs a new DataSchema instance.

Parameters:

args (Any)
kwargs (Any)

classmethod __schematize(schema, schemaname)

Return type:

Union[DataGroup, DataField]

Parameters:

schema (Union[DataGroup, dict[str, Union[DataField, dict[str, Any]]], DataField, dict[str, Any]])
schemaname (str)

__setfieldfactory(fieldname, fieldspec)

Return type:

Callable[[DataSchema, Any], None]

Parameters:

fieldname (str)
fieldspec (dict[str, Any])

__setfieldlistfactory(fieldname, fieldspec)

Return type:

Callable[[DataSchema, list[Any]], None]

Parameters:

fieldname (str)
fieldspec (dict[str, Any])

__setgroupfactory(gname, fieldspecs)

Return type:

Callable[[DataSchema, dict[str, Any]], None]

Parameters:

gname (str)
fieldspecs (dict[str, dict[str, Any]])

_autovalidate(vr, fieldspec, value)[source]

Calls the appropriate validation method for fieldspec and value.

Return type:

ValidationResult

Parameters:

vr (Validator)
fieldspec (dict[str, Any])
value (Any)

_customvalidate(vr, fieldspec, value)[source]

Calls a custom validation method on value and appends result to vr.

Return type:

ValidationResult

Parameters:

vr (Validator)
fieldspec (dict[str, Any])
value (Any)

_getfield(key)[source]

Gets the field or group addressed by key.

Return type:: Union[DataField, DataGroup]
Parameters:: key (str)

_getvalue(key)[source]

Gets the value from the data addressed by key.

Return type:: Any
Parameters:: key (str)

static _isna(value)[source]

Returns True if value is either None or [], False otherwise.

Return type:: bool
Parameters:: value (Any)

_setvalue(key, value)[source]

Sets the value for the data addressed by key (without validation).

Return type:

None

Parameters:

key (str)
value (Any)

data(includemissing=False, onlyrequired=False)[source]

Returns the data of the DataSchema as a schematic dictionary.

Return type:

dict[str, Union[dict[str, Any], Any]]

Parameters:

includemissing (bool)
onlyrequired (bool)

iscomplete(onlyrequired=True)[source]

Checks whether the dataset is complete.

Return type:: bool
Parameters:: onlyrequired (bool)

items(includemissing=False, onlyrequired=False)[source]

Returns a list of key-value pairs for data in the DataSchema.

Return type:

list[tuple[str, Any]]

Parameters:

includemissing (bool)
onlyrequired (bool)

keys(includemissing=False, onlyrequired=False)[source]

Returns a list of keys for the DataSchema.

Return type:

list[str]

Parameters:

includemissing (bool)
onlyrequired (bool)

missing(onlyrequired=True)[source]

Return a list of keys for missing fields.

Return type:: list[str]
Parameters:: onlyrequired (bool)

values(includemissing=False, onlyrequired=False)[source]

Returns a list of values for the data in the DataSchema.

Return type:

list[Any]

Parameters:

includemissing (bool)
onlyrequired (bool)

__data: Type: dict[str, Union[dict[str, Any], Any]]

__keys: Type: list[str]

__schema: Type: dict[str, Union[DataGroup, dict[str, Union[DataField, dict[str, Any]]], DataField, dict[str, Any]]]

__schematized: Type: bool

forcecast: Type: bool

ignorecase: Type: bool

class VField[source]

Bases: DataField

Defines an auto-validated data field for DataSchema classes.

__init__(name, type_, typedesc, constraint, forcecast=None, ignorecase=None, flags=0, required=True)[source]

Instantiates a new VField.

Parameters:

name (str)
type_ (str)
typedesc (str)
constraint (Any)
forcecast (Optional[bool])
ignorecase (Optional[bool])
flags (Union[RegexFlag, int])
required (bool)

_fieldparams = [('name', <class 'str'>, 'The name of the DataField'), ('type_', typing.Any, 'The data type for the DataField'), ('typedesc', <class 'str'>, 'A user-intelligible description of the data type'), ('constraint', typing.Any, 'A DataValidator constraint approprite for type_'), ('forcecast', typing.Optional[bool], 'Whether to force casting of data during validation', None), ('ignorecase', typing.Optional[bool], 'Whether to ignore case for string-type data validation', None), ('flags', typing.Union[re.RegexFlag, int], 'Flags to pass to the regular expression engine if type_ is `str`.', 0), ('required', <class 'bool'>, 'Whether the field is required', True)]: Type: list[Union[tuple[str, Any, str], tuple[str, Any, str, Any]]]

constraint: Type: Any

flags: Type: Union[RegexFlag, int]

forcecast: Type: Optional[bool]

ignorecase: Type: Optional[bool]

class VFieldList[source]

Bases: VField, DataFieldList

Defines a field containing an arbitrary number of VField data.