2023-01-11 02:51:25 +01:00
|
|
|
use criterion::{criterion_group, criterion_main, BatchSize, Criterion};
|
|
|
|
use nu_cli::eval_source;
|
|
|
|
use nu_parser::parse;
|
2024-02-25 23:32:50 +01:00
|
|
|
use nu_plugin::{Encoder, EncodingType, PluginCallResponse, PluginOutput};
|
2023-06-17 14:41:29 +02:00
|
|
|
use nu_protocol::{engine::EngineState, PipelineData, Span, Value};
|
2023-01-11 02:51:25 +01:00
|
|
|
use nu_utils::{get_default_config, get_default_env};
|
2023-12-07 15:13:50 +01:00
|
|
|
use std::path::{Path, PathBuf};
|
2023-01-11 02:51:25 +01:00
|
|
|
|
2023-06-17 14:41:29 +02:00
|
|
|
fn load_bench_commands() -> EngineState {
|
|
|
|
nu_command::add_shell_command_context(nu_cmd_lang::create_default_context())
|
|
|
|
}
|
2023-12-07 15:13:50 +01:00
|
|
|
|
|
|
|
fn canonicalize_path(engine_state: &EngineState, path: &Path) -> PathBuf {
|
|
|
|
let cwd = engine_state.current_work_dir();
|
|
|
|
|
|
|
|
if path.exists() {
|
|
|
|
match nu_path::canonicalize_with(path, cwd) {
|
|
|
|
Ok(canon_path) => canon_path,
|
|
|
|
Err(_) => path.to_owned(),
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
path.to_owned()
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
fn get_home_path(engine_state: &EngineState) -> PathBuf {
|
2023-12-16 21:06:42 +01:00
|
|
|
nu_path::home_dir()
|
|
|
|
.map(|path| canonicalize_path(engine_state, &path))
|
|
|
|
.unwrap_or_default()
|
2023-12-07 15:13:50 +01:00
|
|
|
}
|
|
|
|
|
2023-01-11 02:51:25 +01:00
|
|
|
// FIXME: All benchmarks live in this 1 file to speed up build times when benchmarking.
|
|
|
|
// When the *_benchmarks functions were in different files, `cargo bench` would build
|
|
|
|
// an executable for every single one - incredibly slowly. Would be nice to figure out
|
|
|
|
// a way to split things up again.
|
|
|
|
|
|
|
|
fn parser_benchmarks(c: &mut Criterion) {
|
2023-06-17 14:41:29 +02:00
|
|
|
let mut engine_state = load_bench_commands();
|
2023-12-07 15:13:50 +01:00
|
|
|
let home_path = get_home_path(&engine_state);
|
|
|
|
|
|
|
|
// parsing config.nu breaks without PWD set, so set a valid path
|
2023-01-11 02:51:25 +01:00
|
|
|
engine_state.add_env_var(
|
|
|
|
"PWD".into(),
|
2023-12-07 15:13:50 +01:00
|
|
|
Value::string(home_path.to_string_lossy(), Span::test_data()),
|
2023-01-11 02:51:25 +01:00
|
|
|
);
|
|
|
|
|
|
|
|
let default_env = get_default_env().as_bytes();
|
|
|
|
c.bench_function("parse_default_env_file", |b| {
|
|
|
|
b.iter_batched(
|
|
|
|
|| nu_protocol::engine::StateWorkingSet::new(&engine_state),
|
2023-04-07 23:32:44 +02:00
|
|
|
|mut working_set| parse(&mut working_set, None, default_env, false),
|
2023-01-11 02:51:25 +01:00
|
|
|
BatchSize::SmallInput,
|
|
|
|
)
|
|
|
|
});
|
|
|
|
|
|
|
|
let default_config = get_default_config().as_bytes();
|
|
|
|
c.bench_function("parse_default_config_file", |b| {
|
|
|
|
b.iter_batched(
|
|
|
|
|| nu_protocol::engine::StateWorkingSet::new(&engine_state),
|
2023-04-07 23:32:44 +02:00
|
|
|
|mut working_set| parse(&mut working_set, None, default_config, false),
|
2023-01-11 02:51:25 +01:00
|
|
|
BatchSize::SmallInput,
|
|
|
|
)
|
|
|
|
});
|
|
|
|
|
|
|
|
c.bench_function("eval default_env.nu", |b| {
|
|
|
|
b.iter(|| {
|
|
|
|
let mut stack = nu_protocol::engine::Stack::new();
|
|
|
|
eval_source(
|
|
|
|
&mut engine_state,
|
|
|
|
&mut stack,
|
|
|
|
get_default_env().as_bytes(),
|
|
|
|
"default_env.nu",
|
|
|
|
PipelineData::empty(),
|
2023-02-02 23:07:35 +01:00
|
|
|
false,
|
2023-01-11 02:51:25 +01:00
|
|
|
)
|
|
|
|
})
|
|
|
|
});
|
|
|
|
|
|
|
|
c.bench_function("eval default_config.nu", |b| {
|
|
|
|
b.iter(|| {
|
|
|
|
let mut stack = nu_protocol::engine::Stack::new();
|
|
|
|
eval_source(
|
|
|
|
&mut engine_state,
|
|
|
|
&mut stack,
|
|
|
|
get_default_config().as_bytes(),
|
|
|
|
"default_config.nu",
|
|
|
|
PipelineData::empty(),
|
2023-02-02 23:07:35 +01:00
|
|
|
false,
|
2023-01-11 02:51:25 +01:00
|
|
|
)
|
|
|
|
})
|
|
|
|
});
|
|
|
|
}
|
|
|
|
|
|
|
|
fn eval_benchmarks(c: &mut Criterion) {
|
2023-12-07 15:13:50 +01:00
|
|
|
let mut engine_state = load_bench_commands();
|
|
|
|
let home_path = get_home_path(&engine_state);
|
|
|
|
|
|
|
|
// parsing config.nu breaks without PWD set, so set a valid path
|
|
|
|
engine_state.add_env_var(
|
|
|
|
"PWD".into(),
|
|
|
|
Value::string(home_path.to_string_lossy(), Span::test_data()),
|
|
|
|
);
|
|
|
|
|
2023-01-11 02:51:25 +01:00
|
|
|
c.bench_function("eval default_env.nu", |b| {
|
|
|
|
b.iter(|| {
|
|
|
|
let mut stack = nu_protocol::engine::Stack::new();
|
|
|
|
eval_source(
|
|
|
|
&mut engine_state,
|
|
|
|
&mut stack,
|
|
|
|
get_default_env().as_bytes(),
|
|
|
|
"default_env.nu",
|
|
|
|
PipelineData::empty(),
|
2023-02-02 23:07:35 +01:00
|
|
|
false,
|
2023-01-11 02:51:25 +01:00
|
|
|
)
|
|
|
|
})
|
|
|
|
});
|
|
|
|
|
|
|
|
c.bench_function("eval default_config.nu", |b| {
|
|
|
|
b.iter(|| {
|
|
|
|
let mut stack = nu_protocol::engine::Stack::new();
|
|
|
|
eval_source(
|
|
|
|
&mut engine_state,
|
|
|
|
&mut stack,
|
|
|
|
get_default_config().as_bytes(),
|
|
|
|
"default_config.nu",
|
|
|
|
PipelineData::empty(),
|
2023-02-02 23:07:35 +01:00
|
|
|
false,
|
2023-01-11 02:51:25 +01:00
|
|
|
)
|
|
|
|
})
|
|
|
|
});
|
|
|
|
}
|
|
|
|
|
|
|
|
// generate a new table data with `row_cnt` rows, `col_cnt` columns.
|
|
|
|
fn encoding_test_data(row_cnt: usize, col_cnt: usize) -> Value {
|
Create `Record` type (#10103)
# Description
This PR creates a new `Record` type to reduce duplicate code and
possibly bugs as well. (This is an edited version of #9648.)
- `Record` implements `FromIterator` and `IntoIterator` and so can be
iterated over or collected into. For example, this helps with
conversions to and from (hash)maps. (Also, no more
`cols.iter().zip(vals)`!)
- `Record` has a `push(col, val)` function to help insure that the
number of columns is equal to the number of values. I caught a few
potential bugs thanks to this (e.g. in the `ls` command).
- Finally, this PR also adds a `record!` macro that helps simplify
record creation. It is used like so:
```rust
record! {
"key1" => some_value,
"key2" => Value::string("text", span),
"key3" => Value::int(optional_int.unwrap_or(0), span),
"key4" => Value::bool(config.setting, span),
}
```
Since macros hinder formatting, etc., the right hand side values should
be relatively short and sweet like the examples above.
Where possible, prefer `record!` or `.collect()` on an iterator instead
of multiple `Record::push`s, since the first two automatically set the
record capacity and do less work overall.
# User-Facing Changes
Besides the changes in `nu-protocol` the only other breaking changes are
to `nu-table::{ExpandedTable::build_map, JustTable::kv_table}`.
2023-08-24 21:50:29 +02:00
|
|
|
let record = Value::test_record(
|
|
|
|
(0..col_cnt)
|
|
|
|
.map(|x| (format!("col_{x}"), Value::test_int(x as i64)))
|
|
|
|
.collect(),
|
|
|
|
);
|
2023-01-11 02:51:25 +01:00
|
|
|
|
2023-09-03 16:27:29 +02:00
|
|
|
Value::list(vec![record; row_cnt], Span::test_data())
|
2023-01-11 02:51:25 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
fn encoding_benchmarks(c: &mut Criterion) {
|
|
|
|
let mut group = c.benchmark_group("Encoding");
|
2023-06-18 01:15:16 +02:00
|
|
|
let test_cnt_pairs = [(100, 5), (100, 15), (10000, 5), (10000, 15)];
|
2023-01-11 02:51:25 +01:00
|
|
|
for (row_cnt, col_cnt) in test_cnt_pairs.into_iter() {
|
|
|
|
for fmt in ["json", "msgpack"] {
|
|
|
|
group.bench_function(&format!("{fmt} encode {row_cnt} * {col_cnt}"), |b| {
|
|
|
|
let mut res = vec![];
|
2024-02-25 23:32:50 +01:00
|
|
|
let test_data = PluginOutput::CallResponse(
|
|
|
|
0,
|
|
|
|
PluginCallResponse::value(encoding_test_data(row_cnt, col_cnt)),
|
|
|
|
);
|
2023-01-11 02:51:25 +01:00
|
|
|
let encoder = EncodingType::try_from_bytes(fmt.as_bytes()).unwrap();
|
2024-02-25 23:32:50 +01:00
|
|
|
b.iter(|| encoder.encode(&test_data, &mut res))
|
2023-01-11 02:51:25 +01:00
|
|
|
});
|
|
|
|
}
|
|
|
|
}
|
|
|
|
group.finish();
|
|
|
|
}
|
|
|
|
|
|
|
|
fn decoding_benchmarks(c: &mut Criterion) {
|
|
|
|
let mut group = c.benchmark_group("Decoding");
|
2023-06-18 01:15:16 +02:00
|
|
|
let test_cnt_pairs = [(100, 5), (100, 15), (10000, 5), (10000, 15)];
|
2023-01-11 02:51:25 +01:00
|
|
|
for (row_cnt, col_cnt) in test_cnt_pairs.into_iter() {
|
|
|
|
for fmt in ["json", "msgpack"] {
|
|
|
|
group.bench_function(&format!("{fmt} decode for {row_cnt} * {col_cnt}"), |b| {
|
|
|
|
let mut res = vec![];
|
2024-02-25 23:32:50 +01:00
|
|
|
let test_data = PluginOutput::CallResponse(
|
|
|
|
0,
|
|
|
|
PluginCallResponse::value(encoding_test_data(row_cnt, col_cnt)),
|
|
|
|
);
|
2023-01-11 02:51:25 +01:00
|
|
|
let encoder = EncodingType::try_from_bytes(fmt.as_bytes()).unwrap();
|
2024-02-25 23:32:50 +01:00
|
|
|
encoder.encode(&test_data, &mut res).unwrap();
|
2023-01-11 02:51:25 +01:00
|
|
|
let mut binary_data = std::io::Cursor::new(res);
|
2024-02-25 23:32:50 +01:00
|
|
|
b.iter(|| -> Result<Option<PluginOutput>, _> {
|
2023-01-11 02:51:25 +01:00
|
|
|
binary_data.set_position(0);
|
2024-02-25 23:32:50 +01:00
|
|
|
encoder.decode(&mut binary_data)
|
2023-01-11 02:51:25 +01:00
|
|
|
})
|
|
|
|
});
|
|
|
|
}
|
|
|
|
}
|
|
|
|
group.finish();
|
|
|
|
}
|
|
|
|
|
|
|
|
criterion_group!(
|
|
|
|
benches,
|
|
|
|
parser_benchmarks,
|
|
|
|
eval_benchmarks,
|
|
|
|
encoding_benchmarks,
|
|
|
|
decoding_benchmarks
|
|
|
|
);
|
|
|
|
criterion_main!(benches);
|