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
164
165
166
use crate::prelude::*;
use crate::utils::suggestions::suggestions;
use nu_engine::WholeStreamCommand;
use nu_errors::ShellError;
use nu_protocol::{ReturnSuccess, Signature, SyntaxShape, Value};
use nu_source::Tagged;
use nu_value_ext::as_string;
pub struct SplitBy;
impl WholeStreamCommand for SplitBy {
fn name(&self) -> &str {
"split-by"
}
fn signature(&self) -> Signature {
Signature::build("split-by").optional(
"column_name",
SyntaxShape::String,
"the name of the column within the nested table to split by",
)
}
fn usage(&self) -> &str {
"Creates a new table with the data from the inner tables split by the column given."
}
fn run_with_actions(&self, args: CommandArgs) -> Result<ActionStream, ShellError> {
split_by(args)
}
}
pub fn split_by(args: CommandArgs) -> Result<ActionStream, ShellError> {
let name = args.call_info.name_tag.clone();
let column_name: Option<Tagged<String>> = args.opt(0)?;
let values: Vec<Value> = args.input.collect();
if values.len() > 1 || values.is_empty() {
return Err(ShellError::labeled_error(
"Expected table from pipeline",
"requires a table input",
name,
));
}
let split = split(&column_name, &values[0], &name)?;
Ok(ActionStream::one(ReturnSuccess::value(split)))
}
enum Grouper {
ByColumn(Option<Tagged<String>>),
}
pub fn split(
column_name: &Option<Tagged<String>>,
values: &Value,
tag: impl Into<Tag>,
) -> Result<Value, ShellError> {
let name = tag.into();
let grouper = if let Some(column_name) = column_name {
Grouper::ByColumn(Some(column_name.clone()))
} else {
Grouper::ByColumn(None)
};
match grouper {
Grouper::ByColumn(Some(column_name)) => {
let block = Box::new(move |_, row: &Value| {
match row.get_data_by_key(column_name.borrow_spanned()) {
Some(group_key) => Ok(as_string(&group_key)?),
None => Err(suggestions(column_name.borrow_tagged(), row)),
}
});
nu_data::utils::split(values, &Some(block), &name)
}
Grouper::ByColumn(None) => {
let block = Box::new(move |_, row: &Value| as_string(row));
nu_data::utils::split(values, &Some(block), &name)
}
}
}
#[cfg(test)]
mod tests {
use super::split;
use super::ShellError;
use nu_data::utils::helpers::committers_grouped_by_date;
use nu_protocol::UntaggedValue;
use nu_source::*;
use nu_test_support::value::{date, int, row, string, table};
#[test]
fn splits_inner_tables_by_key() {
let for_key = Some(String::from("country").tagged_unknown());
assert_eq!(
split(&for_key, &committers_grouped_by_date(), Tag::unknown()).unwrap(),
UntaggedValue::row(indexmap! {
"EC".into() => row(indexmap! {
"2019-07-23".into() => table(&[
row(indexmap!{"name".into() => string("AR"), "country".into() => string("EC"), "date".into() => date("2019-07-23"), "chickens".into() => int(10)})
]),
"2019-09-24".into() => table(&[
row(indexmap!{"name".into() => string("AR"), "country".into() => string("EC"), "date".into() => date("2019-09-24"), "chickens".into() => int(20)})
]),
"2019-10-10".into() => table(&[
row(indexmap!{"name".into() => string("AR"), "country".into() => string("EC"), "date".into() => date("2019-10-10"), "chickens".into() => int(30)})
])
}),
"NZ".into() => row(indexmap! {
"2019-07-23".into() => table(&[
row(indexmap!{"name".into() => string("JT"), "country".into() => string("NZ"), "date".into() => date("2019-07-23"), "chickens".into() => int(5)})
]),
"2019-09-24".into() => table(&[
row(indexmap!{"name".into() => string("JT"), "country".into() => string("NZ"), "date".into() => date("2019-09-24"), "chickens".into() => int(10)})
]),
"2019-10-10".into() => table(&[
row(indexmap!{"name".into() => string("JT"), "country".into() => string("NZ"), "date".into() => date("2019-10-10"), "chickens".into() => int(15)})
])
}),
"US".into() => row(indexmap! {
"2019-07-23".into() => table(&[
row(indexmap!{"name".into() => string("YK"), "country".into() => string("US"), "date".into() => date("2019-07-23"), "chickens".into() => int(2)})
]),
"2019-09-24".into() => table(&[
row(indexmap!{"name".into() => string("YK"), "country".into() => string("US"), "date".into() => date("2019-09-24"), "chickens".into() => int(4)})
]),
"2019-10-10".into() => table(&[
row(indexmap!{"name".into() => string("YK"), "country".into() => string("US"), "date".into() => date("2019-10-10"), "chickens".into() => int(6)})
])
})
}).into_untagged_value()
);
}
#[test]
fn errors_if_key_within_some_inner_table_is_missing() {
let for_key = Some(String::from("country").tagged_unknown());
let nu_releases = row(indexmap! {
"2019-07-23".into() => table(&[
row(indexmap!{"name".into() => string("AR"), "country".into() => string("EC"), "date".into() => string("2019-07-23")})
]),
"2019-09-24".into() => table(&[
row(indexmap!{"name".into() => UntaggedValue::string("JT").into_value(Tag::from(Span::new(5,10))), "date".into() => string("2019-09-24")})
]),
"October 10-2019".into() => table(&[
row(indexmap!{"name".into() => string("YK"), "country".into() => string("US"), "date".into() => string("October 10-2019")})
])
});
assert!(split(&for_key, &nu_releases, Tag::from(Span::new(5, 10))).is_err());
}
#[test]
fn examples_work_as_expected() -> Result<(), ShellError> {
use super::SplitBy;
use crate::examples::test as test_examples;
test_examples(SplitBy {})
}
}