extern/nushell
1
0
Fork 1
mirror of https://github.com/nushell/nushell.git synced 2024-11-24 17:34:00 +01:00
Code Issues Packages Projects Releases Wiki Activity

Renamed polars concatenate and added expression support. (#13781)

Browse Source 
# Description
In order to be more consistent with the nushell terminology and with
polars expression terminology `polars concatenate` is now `polars
str-join`. `polars str-join` can also be used as expression.

<img width="857" alt="Screenshot 2024-09-04 at 12 41 25"
src="https://github.com/user-attachments/assets/8cc5a0c2-194c-49ec-9fe1-65ec4825414d">


# User-Facing Changes
- `polars concatenate` is now `polars str-join`
- `polars str-join` can be used as an expression
This commit is contained in:
Jack Wright 2024-09-05 09:28:34 -07:00 committed by GitHub
parent e7c5f83460
commit dc53c20628
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
4 changed files with 180 additions and 124 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
crates/nu_plugin_polars/src/dataframe/series/mod.rs
View File

@ -48,7 +48,6 @@ pub(crate) fn series_commands() -> Vec<Box<dyn PluginCommand<Plugin = PolarsPlug
Box::new(ArgUnique),
Box::new(AsDate),
Box::new(AsDateTime),
Box::new(Concatenate),
Box::new(Contains),
Box::new(Cumulative),
Box::new(GetDay),
@ -74,6 +73,7 @@ pub(crate) fn series_commands() -> Vec<Box<dyn PluginCommand<Plugin = PolarsPlug
Box::new(SetSeries),
Box::new(SetWithIndex),
Box::new(Shift),
Box::new(StrJoin),
Box::new(StrLengths),
Box::new(StrSlice),
Box::new(StrFTime),

121
crates/nu_plugin_polars/src/dataframe/series/string/concatenate.rs
View File

@ -1,121 +0,0 @@
use crate::{values::CustomValueSupport, PolarsPlugin};
use super::super::super::values::{Column, NuDataFrame};
use nu_plugin::{EngineInterface, EvaluatedCall, PluginCommand};
use nu_protocol::{
Category, Example, LabeledError, PipelineData, ShellError, Signature, Span, SyntaxShape, Type,
Value,
};
use polars::prelude::{IntoSeries, StringNameSpaceImpl};
#[derive(Clone)]
pub struct Concatenate;
impl PluginCommand for Concatenate {
type Plugin = PolarsPlugin;
fn name(&self) -> &str {
"polars concatenate"
}
fn description(&self) -> &str {
"Concatenates strings with other array."
}
fn signature(&self) -> Signature {
Signature::build(self.name())
.required(
"other",
SyntaxShape::Any,
"Other array with string to be concatenated",
)
.input_output_type(
Type::Custom("dataframe".into()),
Type::Custom("dataframe".into()),
)
.category(Category::Custom("dataframe".into()))
}
fn examples(&self) -> Vec<Example> {
vec![Example {
description: "Concatenate string",
example: r#"let other = ([za xs cd] | polars into-df);
[abc abc abc] | polars into-df | polars concatenate $other"#,
result: Some(
NuDataFrame::try_from_columns(
vec![Column::new(
"0".to_string(),
vec![
Value::test_string("abcza"),
Value::test_string("abcxs"),
Value::test_string("abccd"),
],
)],
None,
)
.expect("simple df for test should not fail")
.into_value(Span::test_data()),
),
}]
}
fn run(
&self,
plugin: &Self::Plugin,
engine: &EngineInterface,
call: &EvaluatedCall,
input: PipelineData,
) -> Result<PipelineData, LabeledError> {
command(plugin, engine, call, input).map_err(LabeledError::from)
}
}
fn command(
plugin: &PolarsPlugin,
engine: &EngineInterface,
call: &EvaluatedCall,
input: PipelineData,
) -> Result<PipelineData, ShellError> {
let df = NuDataFrame::try_from_pipeline_coerce(plugin, input, call.head)?;
let other: Value = call.req(0)?;
let other_span = other.span();
let other_df = NuDataFrame::try_from_value_coerce(plugin, &other, other_span)?;
let other_series = other_df.as_series(other_span)?;
let other_chunked = other_series.str().map_err(|e| ShellError::GenericError {
error: "The concatenate only with string columns".into(),
msg: e.to_string(),
span: Some(other_span),
help: None,
inner: vec![],
})?;
let series = df.as_series(call.head)?;
let chunked = series.str().map_err(|e| ShellError::GenericError {
error: "The concatenate only with string columns".into(),
msg: e.to_string(),
span: Some(call.head),
help: None,
inner: vec![],
})?;
let mut res = chunked.concat(other_chunked);
res.rename(series.name());
let df = NuDataFrame::try_from_series_vec(vec![res.into_series()], call.head)?;
df.to_pipeline_data(plugin, engine, call.head)
}
#[cfg(test)]
mod test {
use super::*;
use crate::test::test_polars_plugin_command;
#[test]
fn test_examples() -> Result<(), ShellError> {
test_polars_plugin_command(&Concatenate)
}
}

4
crates/nu_plugin_polars/src/dataframe/series/string/mod.rs
View File

@ -1,7 +1,7 @@
mod concatenate;
mod contains;
mod replace;
mod replace_all;
mod str_join;
mod str_lengths;
mod str_slice;
mod strftime;
@ -10,10 +10,10 @@ mod to_integer;
mod to_lowercase;
mod to_uppercase;
pub use concatenate::Concatenate;
pub use contains::Contains;
pub use replace::Replace;
pub use replace_all::ReplaceAll;
pub use str_join::StrJoin;
pub use str_lengths::StrLengths;
pub use str_slice::StrSlice;
pub use strftime::StrFTime;

177
crates/nu_plugin_polars/src/dataframe/series/string/str_join.rs Normal file
View File

@ -0,0 +1,177 @@
use crate::{
values::{
cant_convert_err, CustomValueSupport, NuExpression, PolarsPluginObject, PolarsPluginType,
},
PolarsPlugin,
};
use super::super::super::values::{Column, NuDataFrame};
use nu_plugin::{EngineInterface, EvaluatedCall, PluginCommand};
use nu_protocol::{
Category, Example, LabeledError, PipelineData, ShellError, Signature, Span, SyntaxShape, Type,
Value,
};
use polars::{prelude::StringNameSpaceImpl, series::IntoSeries};
#[derive(Clone)]
pub struct StrJoin;
impl PluginCommand for StrJoin {
type Plugin = PolarsPlugin;
fn name(&self) -> &str {
"polars str-join"
}
fn description(&self) -> &str {
"Concatenates strings within a column or dataframes"
}
fn signature(&self) -> Signature {
Signature::build(self.name())
.optional("other", SyntaxShape::Any, "Other dataframe with a single series of strings to be concatenated. Required when used with a dataframe, ignored when used as an expression.")
.named("delimiter", SyntaxShape::String, "Delimiter to join strings within an expression. Other dataframe when used with a dataframe.", Some('d'))
.switch("ignore-nulls", "Ignore null values. Only available when used as an expression.", Some('n'))
.input_output_types(vec![
(
Type::Custom("expression".into()),
Type::Custom("expression".into()),
),
(
Type::Custom("dataframe".into()),
Type::Custom("dataframe".into()),
),
])
.category(Category::Custom("dataframe".into()))
}
fn examples(&self) -> Vec<Example> {
vec![
Example {
description: "Join strings in a column",
example: r#"[[a]; [abc] [abc] [abc]] | polars into-df | polars select (polars col a | polars str-join -d ',') | polars collect"#,
result: Some(
NuDataFrame::try_from_columns(
vec![Column::new(
"a".to_string(),
vec![Value::test_string("abc,abc,abc")],
)],
None,
)
.expect("simple df for test should not fail")
.into_value(Span::test_data()),
),
},
Example {
description: "StrJoin strings across two series",
example: r#"let other = ([za xs cd] | polars into-df);
[abc abc abc] | polars into-df | polars str-join $other"#,
result: Some(
NuDataFrame::try_from_columns(
vec![Column::new(
"0".to_string(),
vec![
Value::test_string("abcza"),
Value::test_string("abcxs"),
Value::test_string("abccd"),
],
)],
None,
)
.expect("simple df for test should not fail")
.into_value(Span::test_data()),
),
},
]
}
fn run(
&self,
plugin: &Self::Plugin,
engine: &EngineInterface,
call: &EvaluatedCall,
input: PipelineData,
) -> Result<PipelineData, LabeledError> {
let value = input.into_value(call.head)?;
match PolarsPluginObject::try_from_value(plugin, &value)? {
PolarsPluginObject::NuDataFrame(df) => command_df(plugin, engine, call, df),
PolarsPluginObject::NuLazyFrame(lazy) => {
command_df(plugin, engine, call, lazy.collect(call.head)?)
}
PolarsPluginObject::NuExpression(expr) => command_expr(plugin, engine, call, expr),
_ => Err(cant_convert_err(&value, &[PolarsPluginType::NuExpression])),
}
.map_err(LabeledError::from)
}
}
fn command_expr(
plugin: &PolarsPlugin,
engine: &EngineInterface,
call: &EvaluatedCall,
expr: NuExpression,
) -> Result<PipelineData, ShellError> {
let delimiter = call
.get_flag::<String>("delimiter")?
.map(|x| x.to_string())
.unwrap_or_else(|| "".to_string());
let ignore_nulls = call.has_flag("ignore-nulls")?;
let res: NuExpression = expr
.into_polars()
.str()
.join(&delimiter, ignore_nulls)
.into();
res.to_pipeline_data(plugin, engine, call.head)
}
fn command_df(
plugin: &PolarsPlugin,
engine: &EngineInterface,
call: &EvaluatedCall,
df: NuDataFrame,
) -> Result<PipelineData, ShellError> {
let other: Value = call.req(0).map_err(|_| ShellError::MissingParameter {
param_name: "other".into(),
span: call.head,
})?;
let other_span = other.span();
let other_df = NuDataFrame::try_from_value_coerce(plugin, &other, other_span)?;
let other_series = other_df.as_series(other_span)?;
let other_chunked = other_series.str().map_err(|e| ShellError::GenericError {
error: "The str-join command only works with string columns".into(),
msg: e.to_string(),
span: Some(other_span),
help: None,
inner: vec![],
})?;
let series = df.as_series(call.head)?;
let chunked = series.str().map_err(|e| ShellError::GenericError {
error: "The str-join command only works only with string columns".into(),
msg: e.to_string(),
span: Some(call.head),
help: None,
inner: vec![],
})?;
let mut res = chunked.concat(other_chunked);
res.rename(series.name());
let df = NuDataFrame::try_from_series_vec(vec![res.into_series()], call.head)?;
df.to_pipeline_data(plugin, engine, call.head)
}
#[cfg(test)]
mod test {
use super::*;
use crate::test::test_polars_plugin_command;
#[test]
fn test_examples() -> Result<(), ShellError> {
test_polars_plugin_command(&StrJoin)
}
}