pipe-audit-core 0.1.3

Universal data validation and compliance audit engine
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
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

// src/validators/column/outlier_sigma.rs

use crate::validators::{ValidationReport, ValidationResult, Validator};
use polars::prelude::*;

pub struct OutlierSigmaValidator {
    pub sigma: f64,
}

impl Validator for OutlierSigmaValidator {
    fn name(&self) -> &'static str {
        "OutlierSigma"
    }

    fn validate(&self, df: &DataFrame, column_name: &str) -> ValidationResult<ValidationReport> {
        let series = df.column(column_name)?;

        let f64_series = match series.strict_cast(&DataType::Float64) {
            Ok(s) => s,
            Err(_) => {
                return Ok(ValidationReport {
                    status: "skipped",
                    details: Some("column could not be cast to a numeric type".to_string()),
                });
            }
        };

        let values = f64_series.f64()?;

        if let (Some(mean), Some(std_dev)) = (values.mean(), values.std(1)) {
            if std_dev == 0.0 {
                return Ok(ValidationReport {
                    status: "pass",
                    details: Some("standard deviation is zero; no outliers possible".to_string()),
                });
            }

            let deviation = (values - mean).into_series();
            let threshold = self.sigma * std_dev;

            let mask = abs(&deviation)?.gt(threshold);

            // Handle the Result from .gt() before calling .sum()
            let outlier_count = mask?.sum().unwrap_or(0);

            if outlier_count > 0 {
                Ok(ValidationReport {
                    status: "fail",
                    details: Some(format!("outliers={}, sigma={}", outlier_count, self.sigma)),
                })
            } else {
                Ok(ValidationReport {
                    status: "pass",
                    details: None,
                })
            }
        } else {
            Ok(ValidationReport {
                status: "skipped",
                details: Some("mean or standard deviation could not be calculated".to_string()),
            })
        }
    }
}
#[cfg(test)]
mod tests {
    use super::*;

    fn make_f64_df(values: &[Option<f64>]) -> DataFrame {
        let s = Series::new("col".into(), values.to_vec());
        DataFrame::new(vec![s.into()]).unwrap()
    }

    fn make_str_df(values: &[Option<&str>]) -> DataFrame {
        let s = Series::new("col".into(), values.to_vec());
        DataFrame::new(vec![s.into()]).unwrap()
    }

    #[test]
    fn passes_when_no_outliers() {
        // mean = 5, std_dev = ~1.41, threshold = 2*1.41 = 2.82
        // values are within 2.82 of mean
        let df = make_f64_df(&[Some(4.0), Some(5.0), Some(6.0)]);
        let validator = OutlierSigmaValidator { sigma: 2.0 };
        let report = validator.validate(&df, "col").unwrap();
        assert_eq!(report.status, "pass");
    }

    #[test]
    fn fails_when_outliers_present() {
        // Eight normals around 10 and one extreme at 100.
        // mean ≈ 20.0, sample std ≈ 30.0, threshold = 2σ ≈ 60.0
        // deviation of 100 from mean ≈ 80.0 > 60.0 → outlier flagged.
        let df = make_f64_df(&[
            Some(10.0),
            Some(10.0),
            Some(10.0),
            Some(10.0),
            Some(10.0),
            Some(10.0),
            Some(10.0),
            Some(10.0),
            Some(100.0),
        ]);
        let validator = OutlierSigmaValidator { sigma: 2.0 };
        let report = validator.validate(&df, "col").unwrap();
        assert_eq!(report.status, "fail");
        assert!(report.details.unwrap().contains("outliers="));
    }

    #[test]
    fn passes_when_std_dev_zero() {
        // all values identical → std_dev = 0
        let df = make_f64_df(&[Some(5.0), Some(5.0), Some(5.0)]);
        let validator = OutlierSigmaValidator { sigma: 2.0 };
        let report = validator.validate(&df, "col").unwrap();
        assert_eq!(report.status, "pass");
        assert!(
            report
                .details
                .unwrap()
                .contains("standard deviation is zero")
        );
    }

    #[test]
    fn skips_when_column_not_numeric() {
        let df = make_str_df(&[Some("a"), Some("b")]);
        let validator = OutlierSigmaValidator { sigma: 2.0 };
        let report = validator.validate(&df, "col").unwrap();
        assert_eq!(report.status, "skipped");
        assert!(report.details.unwrap().contains("could not be cast"));
    }

    #[test]
    fn skips_when_all_nulls() {
        let df = make_f64_df(&[None, None, None]);
        let validator = OutlierSigmaValidator { sigma: 2.0 };
        let report = validator.validate(&df, "col").unwrap();
        assert_eq!(report.status, "skipped");
        assert!(
            report
                .details
                .unwrap()
                .contains("mean or standard deviation")
        );
    }

    #[test]
    fn skips_when_empty_column() {
        let s: Series = Series::new("col".into(), Vec::<Option<f64>>::new());
        let df = DataFrame::new(vec![s.into()]).unwrap();
        let validator = OutlierSigmaValidator { sigma: 2.0 };
        let report = validator.validate(&df, "col").unwrap();
        assert_eq!(report.status, "skipped");
        assert!(
            report
                .details
                .unwrap()
                .contains("mean or standard deviation")
        );
    }
}