Screen reader-friendly errors (#10122)

- Hopefully closes #10120  

# Description

This PR adds a new config item, `error_style`. It will render errors in
a screen reader friendly mode when set to `"simple"`. This is done using
`miette`'s own `NarratableReportHandler`, which seamlessly replaces the
default one when needed.

Before:
```
Error: nu:🐚:external_command

  × External command failed
   ╭─[entry #2:1:1]
 1 │ doesnt exist
   · ───┬──
   ·    ╰── executable was not found
   ╰────
  help: No such file or directory (os error 2)
```

After:
```
Error: External command failed
    Diagnostic severity: error
Begin snippet for entry #4 starting at line 1, column 1

snippet line 1: doesnt exist
    label at line 1, columns 1 to 6: executable was not found
diagnostic help: No such file or directory (os error 2)
diagnostic code: nu:🐚:external_command

```

## Things to be determined

- ~Review naming. `errors.style` is not _that_ consistent with the rest
of the code. Menus use a `style` record, but table rendering mode is set
via `mode`.~ As it's a single config, we're using `error_style` for now.
- Should this kind of setting be toggable with one single parameter?
`accessibility.no_decorations` or similar, which would adjust the style
of both errors and tables accordingly.

# User-Facing Changes

No changes by default, errors will be rendered differently if
`error_style` is set to `simple`.

# Tests + Formatting
<!--
Don't forget to add tests that cover your changes.

Make sure you've run and fixed any issues with these commands:

- `cargo fmt --all -- --check` to check standard code formatting (`cargo
fmt --all` applies these changes)
- `cargo clippy --workspace -- -D warnings -D clippy::unwrap_used` to
check that you're using the standard code style
- `cargo test --workspace` to check that all tests pass (on Windows make
sure to [enable developer
mode](https://learn.microsoft.com/en-us/windows/apps/get-started/developer-mode-features-and-debugging))
- `cargo run -- -c "use std testing; testing run-tests --path
crates/nu-std"` to run the tests for the standard library

> **Note**
> from `nushell` you can also use the `toolkit` as follows
> ```bash
> use toolkit.nu # or use an `env_change` hook to activate it
automatically
> toolkit check pr
> ```
-->

# After Submitting

There's a PR updating the docs over here
https://github.com/nushell/nushell.github.io/pull/1026
This commit is contained in:
Joaquín Triñanes
2023-08-27 13:54:15 +02:00
committed by GitHub
parent 5ac5b90aed
commit cc805f3f01
4 changed files with 93 additions and 11 deletions

View File

@ -49,3 +49,58 @@ fn filesize_format_auto_metric_false() {
let actual = nu!(nu_repl_code(code));
assert_eq!(actual.out, r#"["1.9 MiB", "1.9 GiB", "1.8 TiB"]"#);
}
#[test]
fn fancy_default_errors() {
let actual = nu!(nu_repl_code(&[
r#"def force_error [x] {
let span = (metadata $x).span;
error make {
msg: "oh no!"
label: {
text: "here's the error"
start: $span.start
end: $span.end
}
}
}"#,
r#"force_error "My error""#
]));
assert_eq!(
actual.err,
"Error: \u{1b}[31m×\u{1b}[0m oh no!\n ╭─[\u{1b}[36;1;4mline1\u{1b}[0m:1:1]\n \u{1b}[2m1\u{1b}[0m │ force_error \"My error\"\n · \u{1b}[35;1m ─────┬────\u{1b}[0m\n · \u{1b}[35;1m╰── \u{1b}[35;1mhere's the error\u{1b}[0m\u{1b}[0m\n ╰────\n\n\n"
);
}
#[test]
fn narratable_errors() {
let actual = nu!(nu_repl_code(&[
r#"$env.config = { error_style: "plain" }"#,
r#"def force_error [x] {
let span = (metadata $x).span;
error make {
msg: "oh no!"
label: {
text: "here's the error"
start: $span.start
end: $span.end
}
}
}"#,
r#"force_error "my error""#,
]));
assert_eq!(
actual.err,
r#"Error: oh no!
Diagnostic severity: error
Begin snippet for line2 starting at line 1, column 1
snippet line 1: force_error "my error"
label at line 1, columns 13 to 22: here's the error
"#,
);
}