2021-12-21 19:32:09 +01:00
|
|
|
use nu_engine::CallExt;
|
|
|
|
use nu_protocol::{
|
|
|
|
ast::Call,
|
|
|
|
engine::{Command, EngineState, Stack},
|
2022-06-12 21:18:00 +02:00
|
|
|
Category, Example, PipelineData, ShellError, Signature, Span, SyntaxShape, Type, Value,
|
2021-12-21 19:32:09 +01:00
|
|
|
};
|
|
|
|
use polars::prelude::DataType;
|
|
|
|
|
|
|
|
use crate::dataframe::values::Column;
|
|
|
|
|
|
|
|
use super::super::values::NuDataFrame;
|
|
|
|
|
|
|
|
#[derive(Clone)]
|
|
|
|
pub struct TakeDF;
|
|
|
|
|
|
|
|
impl Command for TakeDF {
|
|
|
|
fn name(&self) -> &str {
|
2022-06-12 21:18:00 +02:00
|
|
|
"take"
|
2021-12-21 19:32:09 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
fn usage(&self) -> &str {
|
|
|
|
"Creates new dataframe using the given indices"
|
|
|
|
}
|
|
|
|
|
|
|
|
fn signature(&self) -> Signature {
|
|
|
|
Signature::build(self.name())
|
|
|
|
.required(
|
|
|
|
"indices",
|
|
|
|
SyntaxShape::Any,
|
|
|
|
"list of indices used to take data",
|
|
|
|
)
|
|
|
|
.category(Category::Custom("dataframe".into()))
|
|
|
|
}
|
|
|
|
|
|
|
|
fn examples(&self) -> Vec<Example> {
|
|
|
|
vec![
|
|
|
|
Example {
|
|
|
|
description: "Takes selected rows from dataframe",
|
2022-06-12 21:18:00 +02:00
|
|
|
example: r#"let df = ([[a b]; [4 1] [5 2] [4 3]] | to-df);
|
|
|
|
let indices = ([0 2] | to-df);
|
|
|
|
$df | take $indices"#,
|
2021-12-21 19:32:09 +01:00
|
|
|
result: Some(
|
|
|
|
NuDataFrame::try_from_columns(vec![
|
|
|
|
Column::new(
|
|
|
|
"a".to_string(),
|
|
|
|
vec![Value::test_int(4), Value::test_int(4)],
|
|
|
|
),
|
|
|
|
Column::new(
|
|
|
|
"b".to_string(),
|
|
|
|
vec![Value::test_int(1), Value::test_int(3)],
|
|
|
|
),
|
|
|
|
])
|
|
|
|
.expect("simple df for test should not fail")
|
|
|
|
.into_value(Span::test_data()),
|
|
|
|
),
|
|
|
|
},
|
|
|
|
Example {
|
|
|
|
description: "Takes selected rows from series",
|
2022-06-12 21:18:00 +02:00
|
|
|
example: r#"let series = ([4 1 5 2 4 3] | to-df);
|
|
|
|
let indices = ([0 2] | to-df);
|
|
|
|
$series | take $indices"#,
|
2021-12-21 19:32:09 +01:00
|
|
|
result: Some(
|
|
|
|
NuDataFrame::try_from_columns(vec![Column::new(
|
|
|
|
"0".to_string(),
|
|
|
|
vec![Value::test_int(4), Value::test_int(5)],
|
|
|
|
)])
|
|
|
|
.expect("simple df for test should not fail")
|
|
|
|
.into_value(Span::test_data()),
|
|
|
|
),
|
|
|
|
},
|
|
|
|
]
|
|
|
|
}
|
|
|
|
|
2022-06-12 21:18:00 +02:00
|
|
|
fn input_type(&self) -> Type {
|
|
|
|
Type::Custom("dataframe".into())
|
|
|
|
}
|
|
|
|
|
|
|
|
fn output_type(&self) -> Type {
|
|
|
|
Type::Custom("dataframe".into())
|
|
|
|
}
|
|
|
|
|
2021-12-21 19:32:09 +01:00
|
|
|
fn run(
|
|
|
|
&self,
|
|
|
|
engine_state: &EngineState,
|
|
|
|
stack: &mut Stack,
|
|
|
|
call: &Call,
|
|
|
|
input: PipelineData,
|
|
|
|
) -> Result<PipelineData, ShellError> {
|
|
|
|
command(engine_state, stack, call, input)
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
fn command(
|
|
|
|
engine_state: &EngineState,
|
|
|
|
stack: &mut Stack,
|
|
|
|
call: &Call,
|
|
|
|
input: PipelineData,
|
|
|
|
) -> Result<PipelineData, ShellError> {
|
|
|
|
let index_value: Value = call.req(engine_state, stack, 0)?;
|
|
|
|
let index_span = index_value.span()?;
|
|
|
|
let index = NuDataFrame::try_from_value(index_value)?.as_series(index_span)?;
|
|
|
|
|
|
|
|
let casted = match index.dtype() {
|
|
|
|
DataType::UInt32 | DataType::UInt64 | DataType::Int32 | DataType::Int64 => {
|
|
|
|
index.cast(&DataType::UInt32).map_err(|e| {
|
2022-04-18 14:34:10 +02:00
|
|
|
ShellError::GenericError(
|
2021-12-21 19:32:09 +01:00
|
|
|
"Error casting index list".into(),
|
|
|
|
e.to_string(),
|
2022-04-18 14:34:10 +02:00
|
|
|
Some(index_span),
|
|
|
|
None,
|
|
|
|
Vec::new(),
|
2021-12-21 19:32:09 +01:00
|
|
|
)
|
|
|
|
})
|
|
|
|
}
|
2022-04-18 14:34:10 +02:00
|
|
|
_ => Err(ShellError::GenericError(
|
2021-12-21 19:32:09 +01:00
|
|
|
"Incorrect type".into(),
|
|
|
|
"Series with incorrect type".into(),
|
2022-04-18 14:34:10 +02:00
|
|
|
Some(call.head),
|
|
|
|
Some("Consider using a Series with type int type".into()),
|
|
|
|
Vec::new(),
|
2021-12-21 19:32:09 +01:00
|
|
|
)),
|
|
|
|
}?;
|
|
|
|
|
|
|
|
let indices = casted.u32().map_err(|e| {
|
2022-04-18 14:34:10 +02:00
|
|
|
ShellError::GenericError(
|
2021-12-21 19:32:09 +01:00
|
|
|
"Error casting index list".into(),
|
|
|
|
e.to_string(),
|
2022-04-18 14:34:10 +02:00
|
|
|
Some(index_span),
|
|
|
|
None,
|
|
|
|
Vec::new(),
|
2021-12-21 19:32:09 +01:00
|
|
|
)
|
|
|
|
})?;
|
|
|
|
|
|
|
|
NuDataFrame::try_from_pipeline(input, call.head).and_then(|df| {
|
|
|
|
df.as_ref()
|
|
|
|
.take(indices)
|
|
|
|
.map_err(|e| {
|
2022-04-18 14:34:10 +02:00
|
|
|
ShellError::GenericError(
|
2021-12-21 19:32:09 +01:00
|
|
|
"Error taking values".into(),
|
|
|
|
e.to_string(),
|
2022-04-18 14:34:10 +02:00
|
|
|
Some(call.head),
|
|
|
|
None,
|
|
|
|
Vec::new(),
|
2021-12-21 19:32:09 +01:00
|
|
|
)
|
|
|
|
})
|
|
|
|
.map(|df| PipelineData::Value(NuDataFrame::dataframe_into_value(df, call.head), None))
|
|
|
|
})
|
|
|
|
}
|
|
|
|
|
|
|
|
#[cfg(test)]
|
|
|
|
mod test {
|
|
|
|
use super::super::super::test_dataframe::test_dataframe;
|
|
|
|
use super::*;
|
|
|
|
|
|
|
|
#[test]
|
|
|
|
fn test_examples() {
|
|
|
|
test_dataframe(vec![Box::new(TakeDF {})])
|
|
|
|
}
|
|
|
|
}
|