tabiew 0.16.0

A lightweight TUI application to view and query tabular data files, such as CSV, TSV, and parquet.
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

use std::{io::Read, path::Path};

use itertools::Itertools;
use polars::{
    frame::DataFrame,
    prelude::{AnyValue, DataType, PlSmallStr, Schema},
};
use rusqlite::Connection;
use tempfile::NamedTempFile;

use crate::{AppResult, args::Args, misc::stdin::stdin, readers::ReaderSource};

use super::{DataFrameReader, NamedFrames};

#[derive(Debug, Default)]
pub struct SqliteToDataFrames {
    key: Option<String>,
}

impl SqliteToDataFrames {
    pub fn from_args(args: &Args) -> Self {
        Self {
            key: args.sqlite_key.clone(),
        }
    }

    pub fn key(self, key: String) -> Self {
        SqliteToDataFrames { key: Some(key) }
    }
}

impl DataFrameReader for SqliteToDataFrames {
    fn read_to_data_frames(&self, input: ReaderSource) -> AppResult<NamedFrames> {
        match input {
            ReaderSource::File(path) => path_to_name_frames(path, self.key.as_deref()),
            ReaderSource::Stdin => {
                let temp_file = NamedTempFile::new()?;
                let mut buf = Vec::new();
                stdin().read_to_end(&mut buf)?;
                std::fs::write(temp_file.path(), buf)?;
                path_to_name_frames(temp_file.path(), self.key.as_deref())
            }
        }
    }
}

fn path_to_name_frames(path: impl AsRef<Path>, key: Option<&str>) -> AppResult<NamedFrames> {
    let conn = Connection::open(path)?;

    if let Some(key) = key {
        conn.pragma_update(None, "key", key)?;
    }
    // Fetch table names
    let names = conn
        .prepare("SELECT name FROM sqlite_master WHERE type='table' AND name NOT LIKE 'sqlite_%';")?
        .query_map([], |row| row.get(0))?
        .collect::<Result<Vec<String>, _>>()?;

    // Iterate over tables and transform them into data_frames
    names
        .into_iter()
        .map(|name| {
            let df = get_data_frame(&conn, &name)?;
            Ok((name, df))
        })
        .collect::<AppResult<Vec<_>>>()
        .map(|vec| vec.into_boxed_slice())
}

fn get_data_frame(conn: &Connection, table_name: &str) -> AppResult<DataFrame> {
    let mut stmt = conn.prepare(&format!(
        "SELECT * FROM \"{}\"",
        table_name.replace('"', "\"\"")
    ))?;
    let names = stmt
        .column_names()
        .into_iter()
        .map(PlSmallStr::from)
        .collect_vec();
    let rows = stmt
        .query_map([], sqlite_to_polars_row)?
        .collect::<Result<Vec<_>, _>>()?;
    let schema = Schema::from_iter(names.into_iter().enumerate().map(|(idx, name)| {
        let dtype = rows
            .iter()
            .filter_map(|row| row.0.get(idx))
            .map(AnyValue::dtype)
            .fold(DataType::Null, merge_dtypes);
        match dtype {
            DataType::Null => (name, DataType::String),
            dtype => (name, dtype),
        }
    }));
    let rows = rows
        .into_iter()
        .map(|row| {
            row.0
                .into_iter()
                .zip(schema.iter_values())
                .map(|(value, dtype)| match (value, dtype) {
                    (AnyValue::Int64(n), DataType::String) => {
                        AnyValue::StringOwned(n.to_string().into())
                    }
                    (AnyValue::Float64(f), DataType::String) => {
                        AnyValue::StringOwned(f.to_string().into())
                    }
                    (AnyValue::BinaryOwned(buf), DataType::String) => {
                        AnyValue::StringOwned(String::from_utf8_lossy(&buf).into_owned().into())
                    }
                    (value, _) => value,
                })
                .collect_vec()
        })
        .map(polars::frame::row::Row)
        .collect_vec();
    Ok(DataFrame::from_rows_and_schema(&rows, &schema)?)
}

fn merge_dtypes(lhs: DataType, rhs: DataType) -> DataType {
    match (lhs, rhs) {
        (DataType::Null, dtype) | (dtype, DataType::Null) => dtype,
        (lhs, rhs) if lhs == rhs => lhs,
        (DataType::Int64, DataType::Float64) | (DataType::Float64, DataType::Int64) => {
            DataType::Float64
        }
        _ => DataType::String,
    }
}

fn sqlite_to_polars_row<'a>(
    row: &rusqlite::Row,
) -> Result<polars::frame::row::Row<'a>, rusqlite::Error> {
    (0..row.as_ref().column_count())
        .map(|idx| row.get_ref(idx))
        .map_ok(|value| match value {
            rusqlite::types::ValueRef::Null => AnyValue::Null,
            rusqlite::types::ValueRef::Integer(n) => AnyValue::Int64(n),
            rusqlite::types::ValueRef::Real(f) => AnyValue::Float64(f),
            rusqlite::types::ValueRef::Text(buf) => {
                AnyValue::StringOwned(String::from_utf8_lossy(buf).into_owned().into())
            }
            rusqlite::types::ValueRef::Blob(buf) => AnyValue::BinaryOwned(buf.to_owned()),
        })
        .collect::<Result<Vec<_>, _>>()
        .map(polars::frame::row::Row)
}