nushell/crates/nu-command/src/strings/split/row.rs
Stefan Holderbach a52386e837
Box ShellError in Value::Error (#8375)
# Description

Our `ShellError` at the moment has a `std::mem::size_of<ShellError>` of
136 bytes (on AMD64). As a result `Value` directly storing the struct
also required 136 bytes (thanks to alignment requirements).

This change stores the `Value::Error` `ShellError` on the heap.

Pro:
- Value now needs just 80 bytes
- Should be 1 cacheline less (still at least 2 cachelines)

Con:
- More small heap allocations when dealing with `Value::Error`
  - More heap fragmentation
  - Potential for additional required memcopies

# Further code changes

Includes a small refactor of `try` due to a type mismatch in its large
match.

# User-Facing Changes

None for regular users.

Plugin authors may have to update their matches on `Value` if they use
`nu-protocol`

Needs benchmarking to see if there is a benefit in real world workloads.
**Update** small improvements in runtime for workloads with high volume
of values. Significant reduction in maximum resident set size, when many
values are held in memory.

# Tests + Formatting
2023-03-12 09:57:27 +01:00

160 lines
4.6 KiB
Rust

use nu_engine::CallExt;
use nu_protocol::{
ast::Call,
engine::{Command, EngineState, Stack},
Category, Example, PipelineData, ShellError, Signature, Span, Spanned, SyntaxShape, Type,
Value,
};
#[derive(Clone)]
pub struct SubCommand;
impl Command for SubCommand {
fn name(&self) -> &str {
"split row"
}
fn signature(&self) -> Signature {
Signature::build("split row")
.input_output_types(vec![(Type::String, Type::List(Box::new(Type::String)))])
.vectorizes_over_list(true)
.required(
"separator",
SyntaxShape::String,
"the character that denotes what separates rows",
)
.named(
"number",
SyntaxShape::Int,
"Split into maximum number of items",
Some('n'),
)
.category(Category::Strings)
}
fn usage(&self) -> &str {
"Split a string into multiple rows using a separator."
}
fn search_terms(&self) -> Vec<&str> {
vec!["separate", "divide"]
}
fn run(
&self,
engine_state: &EngineState,
stack: &mut Stack,
call: &Call,
input: PipelineData,
) -> Result<PipelineData, ShellError> {
split_row(engine_state, stack, call, input)
}
fn examples(&self) -> Vec<Example> {
vec![
Example {
description: "Split a string into rows of char",
example: "'abc' | split row ''",
result: Some(Value::List {
vals: vec![
Value::test_string(""),
Value::test_string("a"),
Value::test_string("b"),
Value::test_string("c"),
Value::test_string(""),
],
span: Span::test_data(),
}),
},
Example {
description: "Split a string into rows by the specified separator",
example: "'a--b--c' | split row '--'",
result: Some(Value::List {
vals: vec![
Value::test_string("a"),
Value::test_string("b"),
Value::test_string("c"),
],
span: Span::test_data(),
}),
},
Example {
description: "Split a string by '-'",
example: "'-a-b-c-' | split row '-'",
result: Some(Value::List {
vals: vec![
Value::test_string(""),
Value::test_string("a"),
Value::test_string("b"),
Value::test_string("c"),
Value::test_string(""),
],
span: Span::test_data(),
}),
},
]
}
}
fn split_row(
engine_state: &EngineState,
stack: &mut Stack,
call: &Call,
input: PipelineData,
) -> Result<PipelineData, ShellError> {
let name_span = call.head;
let separator: Spanned<String> = call.req(engine_state, stack, 0)?;
let max_split: Option<usize> = call.get_flag(engine_state, stack, "number")?;
input.flat_map(
move |x| split_row_helper(&x, &separator, max_split, name_span),
engine_state.ctrlc.clone(),
)
}
fn split_row_helper(
v: &Value,
separator: &Spanned<String>,
max_split: Option<usize>,
name: Span,
) -> Vec<Value> {
match v.span() {
Ok(v_span) => {
if let Ok(s) = v.as_string() {
match max_split {
Some(max_split) => s
.splitn(max_split, &separator.item)
.map(|s| Value::string(s, v_span))
.collect(),
None => s
.split(&separator.item)
.map(|s| Value::string(s, v_span))
.collect(),
}
} else {
vec![Value::Error {
error: Box::new(ShellError::PipelineMismatch {
exp_input_type: "string".into(),
dst_span: name,
src_span: v_span,
}),
}]
}
}
Err(error) => vec![Value::Error {
error: Box::new(error),
}],
}
}
#[cfg(test)]
mod test {
use super::*;
#[test]
fn test_examples() {
use crate::test_examples;
test_examples(SubCommand {})
}
}