2019-10-21 01:42:07 +02:00
|
|
|
use crate::commands::WholeStreamCommand;
|
|
|
|
use crate::prelude::*;
|
2020-05-18 14:56:01 +02:00
|
|
|
use indexmap::indexmap;
|
Extract core stuff into own crates
This commit extracts five new crates:
- nu-source, which contains the core source-code handling logic in Nu,
including Text, Span, and also the pretty.rs-based debug logic
- nu-parser, which is the parser and expander logic
- nu-protocol, which is the bulk of the types and basic conveniences
used by plugins
- nu-errors, which contains ShellError, ParseError and error handling
conveniences
- nu-textview, which is the textview plugin extracted into a crate
One of the major consequences of this refactor is that it's no longer
possible to `impl X for Spanned<Y>` outside of the `nu-source` crate, so
a lot of types became more concrete (Value became a concrete type
instead of Spanned<Value>, for example).
This also turned a number of inherent methods in the main nu crate into
plain functions (impl Value {} became a bunch of functions in the
`value` namespace in `crate::data::value`).
2019-11-26 03:30:48 +01:00
|
|
|
use nu_errors::ShellError;
|
2020-05-18 14:56:01 +02:00
|
|
|
use nu_protocol::{ReturnSuccess, Signature, SyntaxShape, UntaggedValue, Value};
|
2019-11-21 15:33:14 +01:00
|
|
|
use nu_source::Tagged;
|
2020-06-17 19:33:50 +02:00
|
|
|
use nu_value_ext::as_string;
|
2019-10-21 01:42:07 +02:00
|
|
|
|
|
|
|
pub struct GroupBy;
|
|
|
|
|
|
|
|
#[derive(Deserialize)]
|
|
|
|
pub struct GroupByArgs {
|
2020-05-15 11:16:09 +02:00
|
|
|
column_name: Option<Tagged<String>>,
|
2019-10-21 01:42:07 +02:00
|
|
|
}
|
|
|
|
|
2020-05-29 10:22:52 +02:00
|
|
|
#[async_trait]
|
2019-10-21 01:42:07 +02:00
|
|
|
impl WholeStreamCommand for GroupBy {
|
|
|
|
fn name(&self) -> &str {
|
|
|
|
"group-by"
|
|
|
|
}
|
|
|
|
|
|
|
|
fn signature(&self) -> Signature {
|
2020-05-15 11:16:09 +02:00
|
|
|
Signature::build("group-by").optional(
|
|
|
|
"column_name",
|
|
|
|
SyntaxShape::String,
|
|
|
|
"the name of the column to group by",
|
|
|
|
)
|
2019-10-21 01:42:07 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
fn usage(&self) -> &str {
|
|
|
|
"Creates a new table with the data from the table rows grouped by the column given."
|
|
|
|
}
|
|
|
|
|
2020-05-29 10:22:52 +02:00
|
|
|
async fn run(
|
2019-10-21 01:42:07 +02:00
|
|
|
&self,
|
|
|
|
args: CommandArgs,
|
|
|
|
registry: &CommandRegistry,
|
|
|
|
) -> Result<OutputStream, ShellError> {
|
2020-06-08 06:48:10 +02:00
|
|
|
group_by(args, registry).await
|
2019-10-21 01:42:07 +02:00
|
|
|
}
|
2020-05-12 17:54:29 +02:00
|
|
|
|
2020-05-18 14:56:01 +02:00
|
|
|
fn examples(&self) -> Vec<Example> {
|
|
|
|
vec![
|
|
|
|
Example {
|
|
|
|
description: "Group items by type",
|
|
|
|
example: r#"ls | group-by type"#,
|
|
|
|
result: None,
|
|
|
|
},
|
|
|
|
Example {
|
|
|
|
description: "Group items by their value",
|
|
|
|
example: "echo [1 3 1 3 2 1 1] | group-by",
|
|
|
|
result: Some(vec![UntaggedValue::row(indexmap! {
|
|
|
|
"1".to_string() => UntaggedValue::Table(vec![
|
|
|
|
UntaggedValue::int(1).into(),
|
|
|
|
UntaggedValue::int(1).into(),
|
|
|
|
UntaggedValue::int(1).into(),
|
|
|
|
UntaggedValue::int(1).into(),
|
|
|
|
]).into(),
|
|
|
|
|
|
|
|
"3".to_string() => UntaggedValue::Table(vec![
|
|
|
|
UntaggedValue::int(3).into(),
|
|
|
|
UntaggedValue::int(3).into(),
|
|
|
|
]).into(),
|
|
|
|
|
|
|
|
"2".to_string() => UntaggedValue::Table(vec![
|
|
|
|
UntaggedValue::int(2).into(),
|
|
|
|
]).into(),
|
|
|
|
})
|
|
|
|
.into()]),
|
|
|
|
},
|
|
|
|
]
|
2020-05-12 17:54:29 +02:00
|
|
|
}
|
2019-10-21 01:42:07 +02:00
|
|
|
}
|
|
|
|
|
2020-06-17 19:33:50 +02:00
|
|
|
enum Grouper {
|
|
|
|
ByColumn(Option<Tagged<String>>),
|
|
|
|
}
|
|
|
|
|
2020-06-08 06:48:10 +02:00
|
|
|
pub async fn group_by(
|
|
|
|
args: CommandArgs,
|
|
|
|
registry: &CommandRegistry,
|
|
|
|
) -> Result<OutputStream, ShellError> {
|
2020-05-16 05:18:24 +02:00
|
|
|
let registry = registry.clone();
|
|
|
|
let name = args.call_info.name_tag.clone();
|
2020-06-08 06:48:10 +02:00
|
|
|
let (GroupByArgs { column_name }, input) = args.process(®istry).await?;
|
|
|
|
let values: Vec<Value> = input.collect().await;
|
|
|
|
|
|
|
|
if values.is_empty() {
|
2020-06-17 19:33:50 +02:00
|
|
|
return Err(ShellError::labeled_error(
|
2020-06-08 06:48:10 +02:00
|
|
|
"Expected table from pipeline",
|
|
|
|
"requires a table input",
|
|
|
|
name,
|
2020-06-17 19:33:50 +02:00
|
|
|
));
|
|
|
|
}
|
|
|
|
|
|
|
|
let values = UntaggedValue::table(&values).into_value(&name);
|
|
|
|
|
|
|
|
match group(&column_name, &values, name) {
|
|
|
|
Ok(grouped) => Ok(OutputStream::one(ReturnSuccess::value(grouped))),
|
|
|
|
Err(reason) => Err(reason),
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
pub fn suggestions(tried: Tagged<&str>, for_value: &Value) -> ShellError {
|
|
|
|
let possibilities = for_value.data_descriptors();
|
|
|
|
|
|
|
|
let mut possible_matches: Vec<_> = possibilities
|
|
|
|
.iter()
|
|
|
|
.map(|x| (natural::distance::levenshtein_distance(x, &tried), x))
|
|
|
|
.collect();
|
|
|
|
|
|
|
|
possible_matches.sort();
|
|
|
|
|
|
|
|
if !possible_matches.is_empty() {
|
|
|
|
ShellError::labeled_error(
|
|
|
|
"Unknown column",
|
|
|
|
format!("did you mean '{}'?", possible_matches[0].1),
|
|
|
|
tried.tag(),
|
|
|
|
)
|
2020-06-08 06:48:10 +02:00
|
|
|
} else {
|
2020-06-17 19:33:50 +02:00
|
|
|
ShellError::labeled_error(
|
|
|
|
"Unknown column",
|
|
|
|
"row does not contain this column",
|
|
|
|
tried.tag(),
|
|
|
|
)
|
2020-06-08 06:48:10 +02:00
|
|
|
}
|
2019-10-29 22:04:31 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
pub fn group(
|
2020-06-17 19:33:50 +02:00
|
|
|
column_name: &Option<Tagged<String>>,
|
|
|
|
values: &Value,
|
2019-10-29 22:04:31 +01:00
|
|
|
tag: impl Into<Tag>,
|
2019-11-21 15:33:14 +01:00
|
|
|
) -> Result<Value, ShellError> {
|
2020-06-17 19:33:50 +02:00
|
|
|
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)),
|
|
|
|
}
|
|
|
|
});
|
|
|
|
|
|
|
|
crate::utils::data::group(&values, &Some(block), &name)
|
|
|
|
}
|
|
|
|
Grouper::ByColumn(None) => {
|
|
|
|
let block = Box::new(move |row: &Value| match as_string(row) {
|
|
|
|
Ok(group_key) => Ok(group_key),
|
|
|
|
Err(reason) => Err(reason),
|
|
|
|
});
|
|
|
|
|
|
|
|
crate::utils::data::group(&values, &Some(block), &name)
|
|
|
|
}
|
|
|
|
}
|
2019-10-29 22:04:31 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
#[cfg(test)]
|
|
|
|
mod tests {
|
2020-06-17 19:33:50 +02:00
|
|
|
use super::group;
|
2019-10-29 22:04:31 +01:00
|
|
|
use indexmap::IndexMap;
|
2020-01-02 05:02:46 +01:00
|
|
|
use nu_errors::ShellError;
|
2019-12-04 20:52:31 +01:00
|
|
|
use nu_protocol::{UntaggedValue, Value};
|
2019-11-21 15:33:14 +01:00
|
|
|
use nu_source::*;
|
2019-10-29 22:04:31 +01:00
|
|
|
|
2019-11-21 15:33:14 +01:00
|
|
|
fn string(input: impl Into<String>) -> Value {
|
2019-12-04 20:52:31 +01:00
|
|
|
UntaggedValue::string(input.into()).into_untagged_value()
|
2019-10-29 22:04:31 +01:00
|
|
|
}
|
|
|
|
|
2019-11-21 15:33:14 +01:00
|
|
|
fn row(entries: IndexMap<String, Value>) -> Value {
|
2019-12-04 20:52:31 +01:00
|
|
|
UntaggedValue::row(entries).into_untagged_value()
|
2019-10-29 22:04:31 +01:00
|
|
|
}
|
|
|
|
|
2019-12-31 08:36:08 +01:00
|
|
|
fn table(list: &[Value]) -> Value {
|
2019-12-04 20:52:31 +01:00
|
|
|
UntaggedValue::table(list).into_untagged_value()
|
2019-10-29 22:04:31 +01:00
|
|
|
}
|
|
|
|
|
2020-06-17 19:33:50 +02:00
|
|
|
fn nu_releases_committers() -> Vec<Value> {
|
2019-11-12 08:07:43 +01:00
|
|
|
vec![
|
2019-10-29 22:04:31 +01:00
|
|
|
row(
|
|
|
|
indexmap! {"name".into() => string("AR"), "country".into() => string("EC"), "date".into() => string("August 23-2019")},
|
|
|
|
),
|
|
|
|
row(
|
|
|
|
indexmap! {"name".into() => string("JT"), "country".into() => string("NZ"), "date".into() => string("August 23-2019")},
|
|
|
|
),
|
|
|
|
row(
|
|
|
|
indexmap! {"name".into() => string("YK"), "country".into() => string("US"), "date".into() => string("October 10-2019")},
|
|
|
|
),
|
|
|
|
row(
|
|
|
|
indexmap! {"name".into() => string("AR"), "country".into() => string("EC"), "date".into() => string("Sept 24-2019")},
|
|
|
|
),
|
|
|
|
row(
|
|
|
|
indexmap! {"name".into() => string("JT"), "country".into() => string("NZ"), "date".into() => string("October 10-2019")},
|
|
|
|
),
|
|
|
|
row(
|
|
|
|
indexmap! {"name".into() => string("YK"), "country".into() => string("US"), "date".into() => string("Sept 24-2019")},
|
|
|
|
),
|
|
|
|
row(
|
|
|
|
indexmap! {"name".into() => string("AR"), "country".into() => string("EC"), "date".into() => string("October 10-2019")},
|
|
|
|
),
|
|
|
|
row(
|
|
|
|
indexmap! {"name".into() => string("JT"), "country".into() => string("NZ"), "date".into() => string("Sept 24-2019")},
|
|
|
|
),
|
|
|
|
row(
|
|
|
|
indexmap! {"name".into() => string("YK"), "country".into() => string("US"), "date".into() => string("August 23-2019")},
|
|
|
|
),
|
2019-11-12 08:07:43 +01:00
|
|
|
]
|
|
|
|
}
|
|
|
|
|
|
|
|
#[test]
|
2020-01-02 05:02:46 +01:00
|
|
|
fn groups_table_by_date_column() -> Result<(), ShellError> {
|
2020-06-17 19:33:50 +02:00
|
|
|
let for_key = Some(String::from("date").tagged_unknown());
|
|
|
|
let sample = table(&nu_releases_committers());
|
2019-10-29 22:04:31 +01:00
|
|
|
|
|
|
|
assert_eq!(
|
2020-06-17 19:33:50 +02:00
|
|
|
group(&for_key, &sample, Tag::unknown())?,
|
2019-10-29 22:04:31 +01:00
|
|
|
row(indexmap! {
|
2019-12-31 08:36:08 +01:00
|
|
|
"August 23-2019".into() => table(&[
|
2019-10-29 22:04:31 +01:00
|
|
|
row(indexmap!{"name".into() => string("AR"), "country".into() => string("EC"), "date".into() => string("August 23-2019")}),
|
|
|
|
row(indexmap!{"name".into() => string("JT"), "country".into() => string("NZ"), "date".into() => string("August 23-2019")}),
|
|
|
|
row(indexmap!{"name".into() => string("YK"), "country".into() => string("US"), "date".into() => string("August 23-2019")})
|
|
|
|
]),
|
2019-12-31 08:36:08 +01:00
|
|
|
"October 10-2019".into() => table(&[
|
2019-10-29 22:04:31 +01:00
|
|
|
row(indexmap!{"name".into() => string("YK"), "country".into() => string("US"), "date".into() => string("October 10-2019")}),
|
|
|
|
row(indexmap!{"name".into() => string("JT"), "country".into() => string("NZ"), "date".into() => string("October 10-2019")}),
|
|
|
|
row(indexmap!{"name".into() => string("AR"), "country".into() => string("EC"), "date".into() => string("October 10-2019")})
|
|
|
|
]),
|
2019-12-31 08:36:08 +01:00
|
|
|
"Sept 24-2019".into() => table(&[
|
2019-10-29 22:04:31 +01:00
|
|
|
row(indexmap!{"name".into() => string("AR"), "country".into() => string("EC"), "date".into() => string("Sept 24-2019")}),
|
|
|
|
row(indexmap!{"name".into() => string("YK"), "country".into() => string("US"), "date".into() => string("Sept 24-2019")}),
|
|
|
|
row(indexmap!{"name".into() => string("JT"), "country".into() => string("NZ"), "date".into() => string("Sept 24-2019")})
|
|
|
|
]),
|
|
|
|
})
|
|
|
|
);
|
2020-01-02 05:02:46 +01:00
|
|
|
|
|
|
|
Ok(())
|
2019-10-29 22:04:31 +01:00
|
|
|
}
|
2019-11-12 08:07:43 +01:00
|
|
|
|
|
|
|
#[test]
|
2020-01-02 05:02:46 +01:00
|
|
|
fn groups_table_by_country_column() -> Result<(), ShellError> {
|
2020-06-17 19:33:50 +02:00
|
|
|
let for_key = Some(String::from("country").tagged_unknown());
|
|
|
|
let sample = table(&nu_releases_committers());
|
2019-11-12 08:07:43 +01:00
|
|
|
|
|
|
|
assert_eq!(
|
2020-06-17 19:33:50 +02:00
|
|
|
group(&for_key, &sample, Tag::unknown())?,
|
2019-11-12 08:07:43 +01:00
|
|
|
row(indexmap! {
|
2019-12-31 08:36:08 +01:00
|
|
|
"EC".into() => table(&[
|
2019-11-12 08:07:43 +01:00
|
|
|
row(indexmap!{"name".into() => string("AR"), "country".into() => string("EC"), "date".into() => string("August 23-2019")}),
|
|
|
|
row(indexmap!{"name".into() => string("AR"), "country".into() => string("EC"), "date".into() => string("Sept 24-2019")}),
|
|
|
|
row(indexmap!{"name".into() => string("AR"), "country".into() => string("EC"), "date".into() => string("October 10-2019")})
|
|
|
|
]),
|
2019-12-31 08:36:08 +01:00
|
|
|
"NZ".into() => table(&[
|
2019-11-12 08:07:43 +01:00
|
|
|
row(indexmap!{"name".into() => string("JT"), "country".into() => string("NZ"), "date".into() => string("August 23-2019")}),
|
|
|
|
row(indexmap!{"name".into() => string("JT"), "country".into() => string("NZ"), "date".into() => string("October 10-2019")}),
|
|
|
|
row(indexmap!{"name".into() => string("JT"), "country".into() => string("NZ"), "date".into() => string("Sept 24-2019")})
|
|
|
|
]),
|
2019-12-31 08:36:08 +01:00
|
|
|
"US".into() => table(&[
|
2019-11-12 08:07:43 +01:00
|
|
|
row(indexmap!{"name".into() => string("YK"), "country".into() => string("US"), "date".into() => string("October 10-2019")}),
|
|
|
|
row(indexmap!{"name".into() => string("YK"), "country".into() => string("US"), "date".into() => string("Sept 24-2019")}),
|
|
|
|
row(indexmap!{"name".into() => string("YK"), "country".into() => string("US"), "date".into() => string("August 23-2019")}),
|
|
|
|
]),
|
|
|
|
})
|
|
|
|
);
|
2020-01-02 05:02:46 +01:00
|
|
|
|
|
|
|
Ok(())
|
2019-11-12 08:07:43 +01:00
|
|
|
}
|
2020-05-18 14:56:01 +02:00
|
|
|
|
|
|
|
#[test]
|
|
|
|
fn examples_work_as_expected() {
|
|
|
|
use super::GroupBy;
|
|
|
|
use crate::examples::test as test_examples;
|
|
|
|
|
|
|
|
test_examples(GroupBy {})
|
|
|
|
}
|
2019-10-21 01:42:07 +02:00
|
|
|
}
|