datavzrd 2.72.2

A tool to create visual HTML reports from collections of CSV/TSV tables
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

use crate::spec::DatasetSpecs;
use crate::utils::column_type::ColumnType;
use anyhow::Result;
use std::collections::HashMap;

pub(crate) struct ColumnSummary {
    pub(crate) column_type: ColumnType,
    pub(crate) min: f32,
    pub(crate) max: f32,
    pub(crate) value_counts: HashMap<String, usize>,
}

impl ColumnSummary {
    pub(crate) fn unique_count(&self) -> usize {
        self.value_counts.len()
    }
}

pub(crate) struct DatasetSummary {
    pub(crate) headers: Vec<String>,
    columns: Vec<ColumnSummary>,
    positions: HashMap<String, usize>,
}

impl DatasetSummary {
    pub(crate) fn build(dataset: &DatasetSpecs) -> Result<Self> {
        let headers = dataset.reader()?.headers()?.clone();
        let mut columns: Vec<ColumnSummary> = headers
            .iter()
            .map(|_| ColumnSummary {
                column_type: ColumnType::default(),
                min: f32::INFINITY,
                max: f32::NEG_INFINITY,
                value_counts: HashMap::new(),
            })
            .collect();
        for record in dataset.reader()?.records()?.skip(dataset.header_rows - 1) {
            for (summary, (title, value)) in
                columns.iter_mut().zip(headers.iter().zip(record.iter()))
            {
                summary.column_type.update(value, title, true)?;
                if let Ok(number) = value.parse::<f32>() {
                    summary.min = summary.min.min(number);
                    summary.max = summary.max.max(number);
                }
                *summary.value_counts.entry(value.to_string()).or_insert(0) += 1;
            }
        }
        let positions = headers
            .iter()
            .enumerate()
            .map(|(index, header)| (header.clone(), index))
            .collect();
        Ok(Self {
            headers,
            columns,
            positions,
        })
    }

    pub(crate) fn column(&self, title: &str) -> &ColumnSummary {
        &self.columns[self.positions[title]]
    }

    pub(crate) fn column_at(&self, index: usize) -> &ColumnSummary {
        &self.columns[index]
    }
}

#[cfg(test)]
mod tests {
    use super::DatasetSummary;
    use crate::spec::DatasetSpecs;
    use crate::utils::column_type::classify_table;
    use std::collections::HashSet;
    use std::str::FromStr;

    fn dataset(path: &str, header_rows: usize) -> DatasetSpecs {
        DatasetSpecs {
            path: path.to_string().parse().unwrap(),
            separator: char::from_str(",").unwrap(),
            header_rows,
            links: None,
            offer_excel: false,
        }
    }

    fn min_max(dataset: &DatasetSpecs, index: usize) -> (f32, f32) {
        dataset
            .reader()
            .unwrap()
            .records()
            .unwrap()
            .skip(dataset.header_rows - 1)
            .filter_map(|record| record.get(index).unwrap().parse::<f32>().ok())
            .fold((f32::INFINITY, f32::NEG_INFINITY), |(min, max), value| {
                (min.min(value), max.max(value))
            })
    }

    fn unique_count(dataset: &DatasetSpecs, index: usize) -> usize {
        dataset
            .reader()
            .unwrap()
            .records()
            .unwrap()
            .skip(dataset.header_rows - 1)
            .map(|record| record.get(index).unwrap().to_string())
            .collect::<HashSet<_>>()
            .len()
    }

    #[test]
    fn test_summary_matches_standalone_functions() {
        for (path, header_rows) in [
            ("tests/data/uniform_datatypes.csv", 1),
            ("tests/data/non_uniform_datatypes.csv", 1),
            ("tests/data/empty_table.csv", 1),
            ("tests/data/additional_header_rows.csv", 2),
        ] {
            let dataset = dataset(path, header_rows);
            let summary = DatasetSummary::build(&dataset).unwrap();
            let classification = classify_table(&dataset, false).unwrap();

            for (index, header) in summary.headers.iter().enumerate() {
                let column = summary.column_at(index);
                assert_eq!(&column.column_type, classification.get(header).unwrap());
                assert_eq!(column.unique_count(), unique_count(&dataset, index));
                assert_eq!((column.min, column.max), min_max(&dataset, index));
            }
        }
    }
}