qua 0.36.0

Structure-aware query CLI for the Quarb engine
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

//! `@| ngrams(n)` / `bigrams` / `trigrams`: the adjacent n-windows
//! of the stream, full windows only, each a list in stream order —
//! the tokenization chosen upstream. `@| frequencies[(k)]`: each
//! distinct value with its count, most frequent first.

use quarb::{QueryResult, Value};

fn doc() -> quarb_json::JsonAdapter {
    quarb_json::JsonAdapter::parse(
        r#"{"words": ["the", "cat", "sat", "on", "the", "cat", "the", "end"],
            "short": ["one", "two"],
            "n": [3, 1, 3, 2, null, 1, 3]}"#,
    )
    .unwrap()
}

fn values(q: &str) -> Vec<String> {
    match quarb::run(q, &doc()).unwrap() {
        QueryResult::Values(vs) => vs.iter().map(Value::to_string).collect(),
        _ => panic!("expected values"),
    }
}

#[test]
fn ngrams_are_the_adjacent_full_windows_in_stream_order() {
    assert_eq!(values("/short/* | :: @| bigrams"), ["one, two"]);
    assert_eq!(values("/short/* | :: @| trigrams @| count"), ["0"]);
    assert_eq!(
        values("/words/* | :: @| ngrams(3) @| first"),
        ["the, cat, sat"]
    );
    assert_eq!(values("/words/* | :: @| bigrams @| count"), ["7"]);
    // Repeats stay repeats: the stream is not a set.
    assert_eq!(
        values("/words/* | :: @| bigrams | join(\" \") @| frequencies(1)"),
        ["%(value = \"the cat\"; n = 2)"]
    );
}

#[test]
fn frequencies_count_distinct_values_most_frequent_first() {
    assert_eq!(
        values("/words/* | :: @| frequencies(3)"),
        [
            "%(value = \"the\"; n = 3)",
            "%(value = \"cat\"; n = 2)",
            "%(value = \"end\"; n = 1)",
        ]
    );
    assert_eq!(values("/words/* | :: @| frequencies @| count"), ["5"]);
    // Nulls are left out; equal counts order by value.
    assert_eq!(
        values("/n/* | :: @| frequencies"),
        [
            "%(value = 3; n = 3)",
            "%(value = 1; n = 2)",
            "%(value = 2; n = 1)"
        ]
    );
    assert_eq!(values("/words/* | :: @| frequencies(2) | :n @| sum"), ["5"]);
}

#[test]
fn the_text_function_is_unchanged() {
    assert_eq!(
        values("/words/* | :: @| join(\" \") | ngrams(2) @| first"),
        ["the cat"]
    );
}

#[test]
fn a_rank_range_or_a_negative_rank_selects_rows_of_the_table() {
    // the 3, cat 2, end 1, on 1, sat 1
    assert_eq!(
        values("/words/* | :: @| frequencies(2..3)"),
        ["%(value = \"cat\"; n = 2)", "%(value = \"end\"; n = 1)"]
    );
    assert_eq!(
        values("/words/* | :: @| frequencies(-2)"),
        ["%(value = \"on\"; n = 1)", "%(value = \"sat\"; n = 1)"]
    );
    assert_eq!(
        values("/words/* | :: @| frequencies(-3..-2)"),
        ["%(value = \"end\"; n = 1)", "%(value = \"on\"; n = 1)"]
    );
    assert_eq!(
        values("/words/* | :: @| frequencies(4..) | :value"),
        ["on", "sat"]
    );
    assert_eq!(
        values("/words/* | :: @| frequencies(..1) | :value"),
        ["the"]
    );
    assert_eq!(values("/words/* | :: @| frequencies(9..) @| count"), ["0"]);
}

#[test]
fn first_last_top_and_bottom_take_a_rank_range_too() {
    assert_eq!(values("/words/* | :: @| first(2..3)"), ["cat", "sat"]);
    assert_eq!(values("/words/* | :: @| first(-2)"), ["the", "end"]);
    // `last` counts from the end: ranks 2..3 are the two before
    // the last, in stream order; a negative rank from the start.
    assert_eq!(values("/words/* | :: @| last(2..3)"), ["cat", "the"]);
    assert_eq!(values("/words/* | :: @| last(-2)"), ["the", "cat"]);
    assert_eq!(values("/words/* | :: @| last(2)"), ["the", "end"]);
    assert_eq!(values("/words/* | :: @| first"), ["the"]);
    // 3, 1, 3, 2, null, 1, 3: descending 3 3 3 2 1 1.
    assert_eq!(values("/n/* | :: @| top(2..3; $_)"), ["3", "3"]);
    assert_eq!(values("/n/* | :: @| top(-2; $_)"), ["1", "1"]);
    assert_eq!(values("/n/* | :: @| bottom(3..4; $_)"), ["2", "3"]);
    assert_eq!(values("/n/* | :: @| top(3; $_)"), ["3", "3", "3"]);
}

#[test]
fn jaccard_takes_the_other_set_from_any_list_value() {
    // {the, cat, sat, on, end} against {the, cat, dog}: 2 of 6.
    let third = |vs: Vec<String>| {
        let v: f64 = vs[0].parse().unwrap();
        (v - 1.0 / 3.0).abs() < 1e-9
    };
    assert!(third(values(
        "/words/* | .b(@(\"the\"; \"cat\"; \"dog\")) | :: @| jaccard($.b)"
    )));
    assert!(third(values(
        "/words/* | :: @| jaccard(@(\"the\"; \"cat\"; \"dog\"))"
    )));
    // A sub-pipeline's values: the other text's distinct words.
    assert_eq!(
        values("/words/* | :: @| jaccard((^/short/* | :: @| unique))"),
        ["0"]
    );
    assert_eq!(
        values("/words/* | :: @| unique @| jaccard((^/words/* | :: @| unique))"),
        ["1"]
    );
}