Replace ExternalStream with new ByteStream type (#12774)

# Description
This PR introduces a `ByteStream` type which is a `Read`-able stream of
bytes. Internally, it has an enum over three different byte stream
sources:
```rust
pub enum ByteStreamSource {
    Read(Box<dyn Read + Send + 'static>),
    File(File),
    Child(ChildProcess),
}
```

This is in comparison to the current `RawStream` type, which is an
`Iterator<Item = Vec<u8>>` and has to allocate for each read chunk.

Currently, `PipelineData::ExternalStream` serves a weird dual role where
it is either external command output or a wrapper around `RawStream`.
`ByteStream` makes this distinction more clear (via `ByteStreamSource`)
and replaces `PipelineData::ExternalStream` in this PR:
```rust
pub enum PipelineData {
    Empty,
    Value(Value, Option<PipelineMetadata>),
    ListStream(ListStream, Option<PipelineMetadata>),
    ByteStream(ByteStream, Option<PipelineMetadata>),
}
```

The PR is relatively large, but a decent amount of it is just repetitive
changes.

This PR fixes #7017, fixes #10763, and fixes #12369.

This PR also improves performance when piping external commands. Nushell
should, in most cases, have competitive pipeline throughput compared to,
e.g., bash.
| Command | Before (MB/s) | After (MB/s) | Bash (MB/s) |
| -------------------------------------------------- | -------------:|
------------:| -----------:|
| `throughput \| rg 'x'` | 3059 | 3744 | 3739 |
| `throughput \| nu --testbin relay o> /dev/null` | 3508 | 8087 | 8136 |

# User-Facing Changes
- This is a breaking change for the plugin communication protocol,
because the `ExternalStreamInfo` was replaced with `ByteStreamInfo`.
Plugins now only have to deal with a single input stream, as opposed to
the previous three streams: stdout, stderr, and exit code.
- The output of `describe` has been changed for external/byte streams.
- Temporary breaking change: `bytes starts-with` no longer works with
byte streams. This is to keep the PR smaller, and `bytes ends-with`
already does not work on byte streams.
- If a process core dumped, then instead of having a `Value::Error` in
the `exit_code` column of the output returned from `complete`, it now is
a `Value::Int` with the negation of the signal number.

# After Submitting
- Update docs and book as necessary
- Release notes (e.g., plugin protocol changes)
- Adapt/convert commands to work with byte streams (high priority is
`str length`, `bytes starts-with`, and maybe `bytes ends-with`).
- Refactor the `tee` code, Devyn has already done some work on this.

---------

Co-authored-by: Devyn Cairns <devyn.cairns@gmail.com>
This commit is contained in:
Ian Manske
2024-05-16 14:11:18 +00:00
committed by GitHub
parent 1b8eb23785
commit 6fd854ed9f
210 changed files with 3955 additions and 4012 deletions

View File

@ -1,7 +1,7 @@
use super::NuSpan;
use anyhow::Result;
use nu_engine::get_columns;
use nu_protocol::{record, ListStream, PipelineData, PipelineMetadata, RawStream, Value};
use nu_protocol::{record, ByteStream, ListStream, PipelineData, PipelineMetadata, Value};
use std::collections::HashMap;
pub fn collect_pipeline(input: PipelineData) -> Result<(Vec<String>, Vec<Vec<Value>>)> {
@ -9,16 +9,7 @@ pub fn collect_pipeline(input: PipelineData) -> Result<(Vec<String>, Vec<Vec<Val
PipelineData::Empty => Ok((vec![], vec![])),
PipelineData::Value(value, ..) => collect_input(value),
PipelineData::ListStream(stream, ..) => Ok(collect_list_stream(stream)),
PipelineData::ExternalStream {
stdout,
stderr,
exit_code,
metadata,
span,
..
} => Ok(collect_external_stream(
stdout, stderr, exit_code, metadata, span,
)),
PipelineData::ByteStream(stream, metadata) => Ok(collect_byte_stream(stream, metadata)),
}
}
@ -42,49 +33,60 @@ fn collect_list_stream(stream: ListStream) -> (Vec<String>, Vec<Vec<Value>>) {
(cols, data)
}
fn collect_external_stream(
stdout: Option<RawStream>,
stderr: Option<RawStream>,
exit_code: Option<ListStream>,
fn collect_byte_stream(
stream: ByteStream,
metadata: Option<PipelineMetadata>,
span: NuSpan,
) -> (Vec<String>, Vec<Vec<Value>>) {
let span = stream.span();
let mut columns = vec![];
let mut data = vec![];
if let Some(stdout) = stdout {
let value = stdout.into_string().map_or_else(
|error| Value::error(error, span),
|string| Value::string(string.item, span),
);
columns.push(String::from("stdout"));
data.push(value);
}
if let Some(stderr) = stderr {
let value = stderr.into_string().map_or_else(
|error| Value::error(error, span),
|string| Value::string(string.item, span),
);
match stream.into_child() {
Ok(child) => match child.wait_with_output() {
Ok(output) => {
let exit_code = output.exit_status.code();
if let Some(stdout) = output.stdout {
columns.push(String::from("stdout"));
data.push(string_or_binary(stdout, span));
}
if let Some(stderr) = output.stderr {
columns.push(String::from("stderr"));
data.push(string_or_binary(stderr, span));
}
columns.push(String::from("exit_code"));
data.push(Value::int(exit_code.into(), span));
}
Err(err) => {
columns.push("".into());
data.push(Value::error(err, span));
}
},
Err(stream) => {
let value = stream
.into_value()
.unwrap_or_else(|err| Value::error(err, span));
columns.push(String::from("stderr"));
data.push(value);
columns.push("".into());
data.push(value);
}
}
if let Some(exit_code) = exit_code {
let list = exit_code.into_iter().collect::<Vec<_>>();
let val = Value::list(list, span);
columns.push(String::from("exit_code"));
data.push(val);
}
if metadata.is_some() {
let val = Value::record(record! { "data_source" => Value::string("ls", span) }, span);
columns.push(String::from("metadata"));
data.push(val);
}
(columns, vec![data])
}
fn string_or_binary(bytes: Vec<u8>, span: NuSpan) -> Value {
match String::from_utf8(bytes) {
Ok(str) => Value::string(str, span),
Err(err) => Value::binary(err.into_bytes(), span),
}
}
/// Try to build column names and a table grid.
pub fn collect_input(value: Value) -> Result<(Vec<String>, Vec<Vec<Value>>)> {
let span = value.span();