Skip to content

Rule

Predicate definition holder

Source code in cuallee/__init__.py 
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
@dataclass
class Rule:
    """Predicate definition holder"""

    method: str
    column: Union[str, List[str], Tuple[str, str]]
    value: Optional[Any]
    data_type: CheckDataType
    coverage: float = 1.0
    options: Union[List[Tuple], None] = None
    status: Union[str, None] = None
    violations: int = 0
    pass_rate: float = 0.0
    ordinal: int = 0
    name: str = None

    @property
    def settings(self) -> dict:
        """holds the additional settings for the predicate execution"""
        return dict(self.options)

    @property
    def key(self):
        """blake2s hash of the rule, made of method, column, value, options and coverage"""
        return (
            hashlib.blake2s(
                bytes(
                    f"{self.method}{self.column}{self.value}{self.options}{self.coverage}",
                    "utf-8",
                )
            )
            .hexdigest()
            .upper()
        )

    def __post_init__(self):
        if (self.coverage <= 0) or (self.coverage > 1):
            raise ValueError("Coverage should be between 0 and 1")

        if isinstance(self.column, List):
            self.column = tuple(self.column)

        if isinstance(self.value, List):
            self.value = tuple(self.value)

        if isinstance(self.value, Tuple) & (self.data_type == CheckDataType.AGNOSTIC):
            # All values can only be of one data type in a rule
            if len(Counter(map(type, self.value)).keys()) > 1:
                raise ValueError("Data types in rule values are inconsistent")

        if (
            self.options
            and isinstance(self.options, dict)
            and (rule_name := self.options.get("name"))
        ):
            self.name = rule_name
        else:
            self.name = self.method

    def __repr__(self):
        return f"Rule(method:{self.name}, column:{self.column}, value:{self.value}, data_type:{self.data_type}, coverage:{self.coverage}, ordinal:{self.ordinal}"

    def __rshift__(self, rule_dict: Dict[str, Any]) -> Dict[str, Any]:
        rule_dict[self.key] = self
        return rule_dict

    def evaluate_violations(self, result: Any, rows: int):
        """Calculates the row violations on the rule"""

        if isinstance(result, str):
            if result == "false":
                self.violations = rows
            elif result == "true":
                self.violations = 0
            else:
                self.violations = abs(int(result))
        elif isinstance(result, bool):
            if result is True:
                self.violations = 0
            elif result is False:
                self.violations = rows
        elif isinstance(result, int):
            if result == 0:
                self.violations = rows
            elif result < 0:
                self.violations = abs(result)
            elif (result > 0) and (result < rows):
                self.violations = rows - result

        else:
            self.violations = 0

    def evaluate_pass_rate(self, rows: int):
        """Percentage of successful rows by this rule"""
        if self.violations <= rows:
            try:
                self.pass_rate = 1 - (self.violations / rows)
            except ZeroDivisionError:
                self.pass_rate = 1.0
        else:
            try:
                self.pass_rate = rows / self.violations
            except ZeroDivisionError:
                self.pass_rate = 0.0

    def evaluate_status(self):
        """Overall PASS/FAIL status of the rule"""
        if self.pass_rate >= self.coverage:
            self.status = "PASS"
        else:
            self.status = "FAIL"

    def evaluate(self, result: Any, rows: int):
        """Generic rule evaluation for checks"""
        self.evaluate_violations(result, rows)
        self.evaluate_pass_rate(rows)
        self.evaluate_status()

key property

blake2s hash of the rule, made of method, column, value, options and coverage

settings: dict property

holds the additional settings for the predicate execution

evaluate(result, rows)

Generic rule evaluation for checks

Source code in cuallee/__init__.py 
161
162
163
164
165
def evaluate(self, result: Any, rows: int):
    """Generic rule evaluation for checks"""
    self.evaluate_violations(result, rows)
    self.evaluate_pass_rate(rows)
    self.evaluate_status()

evaluate_pass_rate(rows)

Percentage of successful rows by this rule

Source code in cuallee/__init__.py 
141
142
143
144
145
146
147
148
149
150
151
152
def evaluate_pass_rate(self, rows: int):
    """Percentage of successful rows by this rule"""
    if self.violations <= rows:
        try:
            self.pass_rate = 1 - (self.violations / rows)
        except ZeroDivisionError:
            self.pass_rate = 1.0
    else:
        try:
            self.pass_rate = rows / self.violations
        except ZeroDivisionError:
            self.pass_rate = 0.0

evaluate_status()

Overall PASS/FAIL status of the rule

Source code in cuallee/__init__.py 
154
155
156
157
158
159
def evaluate_status(self):
    """Overall PASS/FAIL status of the rule"""
    if self.pass_rate >= self.coverage:
        self.status = "PASS"
    else:
        self.status = "FAIL"

evaluate_violations(result, rows)

Calculates the row violations on the rule

Source code in cuallee/__init__.py 
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
def evaluate_violations(self, result: Any, rows: int):
    """Calculates the row violations on the rule"""

    if isinstance(result, str):
        if result == "false":
            self.violations = rows
        elif result == "true":
            self.violations = 0
        else:
            self.violations = abs(int(result))
    elif isinstance(result, bool):
        if result is True:
            self.violations = 0
        elif result is False:
            self.violations = rows
    elif isinstance(result, int):
        if result == 0:
            self.violations = rows
        elif result < 0:
            self.violations = abs(result)
        elif (result > 0) and (result < rows):
            self.violations = rows - result

    else:
        self.violations = 0