Commit Graph

7752 Commits

Author SHA1 Message Date
Tilen Gimpelj
bbea7da669
Remove select error if same row/column is provided (#10350)
This PR is in reference to #10215.

This PR changes `select` to work even if multiple equal items were
provided.
This would previously error, but now works
```nushell
let arg = [ 1 a ]
[[a b c]; [1 2 3] [4 5 6] [7 8 9]] 
| select $arg

```
# User-Facing Changes

Nothing too radical, just experience improvements. Users won't need to
pass the values through `unique` beforehand.
2023-09-13 13:49:55 +02:00
Stefan Holderbach
7f39609d9a
Remove python-like string multiplication (#10293)
# Description
Currently we support "multiplication" of strings, resulting in a terse
way to repeat a particular string.
This can have unintended side effects when dealing with mixed data (e.g.
after parsing data that is not all numbers).
Furthermore as we frequently fall-back to strings while parsing source
code, this introduced a runaway edge case in const evaluation (#10212)

Work for #10233

## Details
- Remove python-like string multiplication.
- Workaround for indentation
  - This should probably be addressed with a purpose built command
- Remove special const-eval error test

# User-Facing Changes
**Major breaking change!**
`"string" * 42` will stop working. (This was used for example in the
stdlib)

We should bless a good alternative before landing this

---------

Co-authored-by: JT <547158+jntrnr@users.noreply.github.com>
2023-09-13 19:52:04 +12:00
Stefan Holderbach
a14e9e0a2e
Invert &Options to Option<&T> (#10315)
Elide the reference for `Copy` type (`usize`)
Use the canonical deref where possible.
* `&Box` -> `&`
* `&String` -> `&str`
* `&PathBuf` -> `&Path`

Skips the ctrl-C handler for now.
2023-09-13 07:00:58 +08:00
Stefan Holderbach
3e14dc3eb8
Remove pythonic int * list behavior (#10292)
# Description
The pythonism that multiplying a scalar integer with a list results in a
repeated concatenation of the list, is ambiguous with other possible
interpretations and thus actively harmful to clear semantics in nushell.

Another possible reading of this scalar/vector product would be trying
to perform elementwise multiplication with the scalar.

Before we bless this alternative as a more reasonable design the best
course of action is to remove this pythonism.

Work related to #10233


# User-Facing Changes
Breaking change as this turns `int * list` or `list * int` into hard
errors.

# Tests + Formatting
Remove the associated test
2023-09-13 10:43:49 +12:00
Antoine Stevan
ba6d8ad261
add std repeat command to replace "foo" * 3 (#10339)
related to
- https://github.com/nushell/nushell/issues/10233
- https://github.com/nushell/nushell/pull/10293
- https://github.com/nushell/nushell/pull/10292

inspired by @kubouch 

# Description
this PR adds a `repeat` command to the standard library

# User-Facing Changes
a new `repeat` command in `std`
```nushell
repeat anything a bunch of times, yielding a list of *n* times the input

# Examples
    repeat a string
    > "foo" | std repeat 3 | str join
    "foofoofoo"

Usage:
  > repeat <n>

Flags:
  -h, --help - Display the help message for this command

Parameters:
  n <int>: the number of repetitions, must be positive

Input/output types:
  ╭───┬───────┬───────────╮
  │ # │ input │  output   │
  ├───┼───────┼───────────┤
  │ 0 │ any   │ list<any> │
  ╰───┴───────┴───────────╯
```

# Tests + Formatting
a new test called `repeat_things` in `test_std.nu`

# After Submitting
2023-09-12 21:59:31 +02:00
Darren Schroeder
2a08865851
add a few more columns to linux ps -l output (#10344)
# Description

This PR tried to add a few more columns to the Linux `ps -l` command.
Those columns are:
* start_time
* user_id
* priority
* process_threads

There are a few that I left commented out that could be added but the
screen was beginning to look crowded. So, I left out:
* group_id
* session_id
* tgp_id (which could be helpful for eventual job control)

And there's like 100 more things that could be added that didn't seem
especially useful right now.


![image](https://github.com/nushell/nushell/assets/343840/065c0538-8f7d-4c9f-871f-a1bc98aff9d1)


# User-Facing Changes
<!-- List of all changes that impact the user experience here. This
helps us keep track of breaking changes. -->

# 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
<!-- If your PR had any user-facing changes, update [the
documentation](https://github.com/nushell/nushell.github.io) after the
PR is merged, if necessary. This will help us keep the docs up to date.
-->
2023-09-12 14:50:05 -05:00
Jakub Žádník
0a3bfe7f73
Allow parse-time evaluation of if (#10326) 2023-09-12 21:35:47 +03:00
JT
451a9c64d3
Change echo to print when not redirected (#10338)
# Description

This changes `echo` to work more closely to what users of other shells
would expect:

* when redirected, `echo` works as before and sends values through the
pipeline
* when not redirected, `echo` will print values to the screen/terminal

# User-Facing Changes

A standalone `echo` now will print to the terminal, if not redirected.

The `echo` command is no longer const eval-able, as it will now print to
the terminal in some cases.

# 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
<!-- If your PR had any user-facing changes, update [the
documentation](https://github.com/nushell/nushell.github.io) after the
PR is merged, if necessary. This will help us keep the docs up to date.
-->
2023-09-13 06:35:01 +12:00
Justin Ma
88d79c84cd
Update Nu to v0.84 for release and nightly-build (#10334)
<!--
if this PR closes one or more issues, you can automatically link the PR
with
them by using one of the [*linking
keywords*](https://docs.github.com/en/issues/tracking-your-work-with-issues/linking-a-pull-request-to-an-issue#linking-a-pull-request-to-an-issue-using-a-keyword),
e.g.
- this PR should close #xxxx
- fixes #xxxx

you can also mention related issues, PRs or discussions!
-->

# Description
Update Nu to v0.84 for release and nightly-build
2023-09-12 22:00:58 +08:00
Justin Ma
abcb0877e2
Upgrade softprops/action-gh-release to v0.1.15 for release and nightly build workflow (#10331) 2023-09-12 21:05:55 +08:00
Eugeen Sablin
9e1e2a4320
provide env to commands and try to start provided path (#10302)
fixes #8551

# Description
Use `open::commands` function to get list of command available for
starting given path. run commands directly, providing environment, until
one of them is successful.

example of output if start was not successful:
```
~\code\nushell> start ..\nustart\a.myext                                                                    09/12/2023 01:37:55 PM
Error: nu:🐚:external_command

  × External command failed
   ╭─[entry #1:1:1]
 1 │ start ..\nustart\a.myext
   ·       ─────────┬────────
   ·                ╰── No command found to start with this path
   ╰────
  help: Try different path or install appropriate command

        Command `cmd /c start "" "..\nustart\a.myext"` failed with exit code: 1
```

# User-Facing Changes
`start` command now provides environment to the external command. This
is how it worked in `nu 0.72`, see linked issue.

# Tests + Formatting
`start` command didn't have any tests and this PR does not add any.
Integration-level tests will require setup specific to OS and
potentially change global environment on testing machine. For unit-level
test it is possible to test `try_commands` function. But is still runs
external commands, and robust test will require apriori knowledge which
commands are necessary successful to run and which are not.
2023-09-12 14:03:41 +02:00
Stefan Holderbach
d53b0a99d0
Rename random decimal to random float (#10320)
# Description
Similar to #9979

# User-Facing Changes
`random decimal` will now raise a warning and can be removed in an
upcoming release.

New command is named `random float`

# Tests + Formatting
Tests updated and improved.
2023-09-12 13:03:05 +02:00
Stefan Holderbach
1fb4f9e455
Rename into decimal to into float (#9979)
# Description
We keep "into decimal" for a release and warn through a message that it
will be removed in 0.86.

All tests are updated to use `into float`

# User-Facing Changes
`into decimal` raises a deprecation warning, will be removed soon.
Use `into float` as the new functionally identical command instead.

```
~/nushell> 2 | into decimal
Error:   × Deprecated command
   ╭─[entry #1:1:1]
 1 │ 2 | into decimal
   ·     ──────┬─────
   ·           ╰── `into decimal` is deprecated and will be removed in 0.86.
   ╰────
  help: Use `into float` instead


2
```

# Tests + Formatting
Updated

---------

Co-authored-by: Darren Schroeder <343840+fdncred@users.noreply.github.com>
2023-09-12 13:02:47 +02:00
Stefan Holderbach
6e9b6f22c9
Deref &String arguments to &str where appropriate (#10321)
# Description
This generally makes for nicer APIs, as you are not forced to use an
existing allocation covering the full `String`.

Some exceptions remain where the underlying type requirements favor it.

# User-Facing Changes
None
2023-09-12 14:06:56 +08:00
Stefan Holderbach
e90b099622
Use slices directly instead of &Vec (#10328)
Simplifies the signature, makes it more flexible.
Detected a few unnecessary allocations in the process.
2023-09-12 11:38:20 +08:00
JT
84c10de864
remove profiling from nushell's hot loop (#10325)
# Description

This removes pipeline element profiling. This could be a useful feature,
but pipeline elements are going to be the most sensitive to in terms of
performance, as `eval_block` and how pipelines are built is one of the
hot loops inside of the eval engine.

# User-Facing Changes

Removes pipeline element profiling.

# 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
<!-- If your PR had any user-facing changes, update [the
documentation](https://github.com/nushell/nushell.github.io) after the
PR is merged, if necessary. This will help us keep the docs up to date.
-->
2023-09-12 06:50:03 +12:00
Jakub Žádník
d618b60d9e
Remove leftover const eval file (#10324) 2023-09-11 21:30:38 +03:00
Yuto
c761f7f844
add 'from ndjson' into standard library (#10283)
close #8574
related #10276 
# Description

added below into standard library
```
def "from ndjson" []: string -> any {
    from json --objects
}
```

# User-Facing Changes


Users can use functions like "from ndjson" in standard library, and can
open ndjson files with `open` command.
```
use std formats *
# `from ndjson` is available now
open sample.ndjson
```

# Tests + Formatting

`toolkit check pr`

- 🟢 `toolkit fmt`
- 🟢 `toolkit clippy`
- 🟢 `toolkit test`
- 🟢 `toolkit test stdlib`


# After Submitting

---------

Co-authored-by: Stefan Holderbach <sholderbach@users.noreply.github.com>
2023-09-11 14:59:07 +02:00
Nano
7b89fab327
Keep order for par-each (#10249)
# Description
This PR adds new flag `--keep-order/-k` for the `par_each` filter. This
flag keeps sequence of output same as the order of input.

Output without the flag:
```nu
> 1..6 | par-each {|n| $n * 2 }
╭────╮
│  4 │
│ 10 │
│  2 │
│  8 │
│ 12 │
│  6 │
╰────╯
```

Output with the `--keep-order` flag:
```nu
> 1..6 | par-each --keep-order {|n| $n * 2 }
╭────╮
│  2 │
│  4 │
│  6 │
│  8 │
│ 10 │
│ 12 │
╰────╯
```

I think the presence of this flag is justified, since:
- Much easier to use than `.. | enumerate | par-each {|p| update item
..} | sort-by index | get item`
- Faster, as it uses internally parallel sorting in the same thread pool

A note about naming: it may conflict with `--keep-empty/-k` flag of the
`each` filter if the same feature will be used in `par-each`, so maybe
it needs some other name.
2023-09-11 06:42:09 -05:00
Stefan Holderbach
eddff46155
Bump calamine and fix clippy (#10314)
Upgrade calamine to 0.22
Reduces one potential dependency duplication

Supersedes #10305

Includes fixes for clippy lints as API changed to return owned data.
2023-09-11 12:39:52 +02:00
Stefan Holderbach
baa50ec9b2
Update crates-ci/typos and fix new typos (#10313)
Supersedes #10309
2023-09-11 12:37:06 +02:00
dependabot[bot]
513186c390
Bump bytesize from 1.2.0 to 1.3.0 (#10306) 2023-09-11 09:55:21 +00:00
dependabot[bot]
0c37463bfa
Bump tempfile from 3.7.0 to 3.8.0 (#10307) 2023-09-11 09:54:49 +00:00
dependabot[bot]
94fc33bbee
Bump actions/checkout from 3 to 4 (#10308) 2023-09-11 09:54:22 +00:00
David Matos
ce378a68a6
Fix variables not allowed in ucp (#10304)
<!--
if this PR closes one or more issues, you can automatically link the PR
with
them by using one of the [*linking
keywords*](https://docs.github.com/en/issues/tracking-your-work-with-issues/linking-a-pull-request-to-an-issue#linking-a-pull-request-to-an-issue-using-a-keyword),
e.g.
- this PR should close #xxxx
- fixes #xxxx

you can also mention related issues, PRs or discussions!
-->

# Description
Fixes #10300 , where using variables didnt work with `ucp` as it was
only expecting a `Expr::FilePath`.

Before: (from the issue)
```
❯ ucp -r $var $folder
Error:   × Missing file operand
   ╭─[entry #40:1:1]
 1 │ ucp -r $var $folder
   · ─┬─
   ·  ╰── Missing file operand
   ╰────
  help: Please provide source and destination paths
```
Now:
```
`ucp -r $var $folder`
# success
```

Also added the test to ensure its working:) . Oh, and I tweaked again
slightly the messages on two tests because now the whole `path` is
printed rather than `a`. Say:
```
#before
`cp a a` --> 'a' and 'a' are the same file 
# now
`cp a a` --> /home/current/location/a and /home/current/location/a are the same file
```
<!--
Thank you for improving Nushell. Please, check our [contributing
guide](../CONTRIBUTING.md) and talk to the core team before making major
changes.

Description of your pull request goes here. **Provide examples and/or
screenshots** if your changes affect the user experience.
-->

# User-Facing Changes
<!-- List of all changes that impact the user experience here. This
helps us keep track of breaking changes. -->

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

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

- [X] `cargo fmt --all -- --check` to check standard code formatting
(`cargo fmt --all` applies these changes)
- [X] `cargo clippy --workspace -- -D warnings -D clippy::unwrap_used`
to check that you're using the standard code style
- [X] `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))
- [X] `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
<!-- If your PR had any user-facing changes, update [the
documentation](https://github.com/nushell/nushell.github.io) after the
PR is merged, if necessary. This will help us keep the docs up to date.
-->
2023-09-10 17:54:33 -05:00
Antoine Stevan
fa40740e77
add case-insensitive example to where (#10299)
related to
-
https://discord.com/channels/601130461678272522/614593951969574961/1150395064292495400

# Description
two cool things about the `where` command
- it's versatile enough to allow creating a case-insensitive version of
itself
- it does not require the explicit use of a closure

this PR adds an example showing how to filter with `where` but
case-insensitively and without an explicite closure.

# User-Facing Changes
new example to `where`:
```nushell
  Find case-insensitively files called "readme", without an explicit closure
  > ls | where ($it.name | str downcase) =~ readme
```

# Tests + Formatting
the new example test above.

# After Submitting
2023-09-10 15:24:38 -05:00
Darren Schroeder
762fdb98ac
silence some ucp warnings (#10294)
# Description

This PR fixes some ucp warnings.

# User-Facing Changes
<!-- List of all changes that impact the user experience here. This
helps us keep track of breaking changes. -->

# 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
<!-- If your PR had any user-facing changes, update [the
documentation](https://github.com/nushell/nushell.github.io) after the
PR is merged, if necessary. This will help us keep the docs up to date.
-->
2023-09-09 19:41:53 -05:00
dzorya
5f795b1aec
Mention that perl is required for feature static-link-openssl… (#10291)
this PR should close https://github.com/nushell/nushell/issues/10290

it is a simple comment in Cargo.toml, which explains that perl is
required for feature static-link-openssl to work.
2023-09-09 22:34:07 +02:00
Tilen Gimpelj
6811700b90
update reject to be able to recive arg list (#10216)
This PR is in relation to #10215 

# Description

This PR introduces `reject` to receive list of columns or rows as
argument.
This change is similar to change of `select` and the code used is
similar.

# User-Facing Changes
The user will be able to pass a list as rejection arguments.
```nushell
let arg = [ type size ]
[[name type size]; [ cargo.toml file 20mb ] [ Cargo.lock file 20mb] [src dir 100mb]] | reject $arg
```
2023-09-09 15:01:25 -05:00
Tilen Gimpelj
248aca7a44
reject multiple row args support (#10163)
# Description
This PR fixes `reject` failing when providing row items in ascending
order.


# User-Facing Changes
users can now `reject` multiple rows independently of each other.
```nushell
let foo = [[a b]; [ 1 2] [3 4] [ 5 6]]
# this will work independant of the order
print ($foo | reject 2 1)
print ($foo | reject 1 2)
```

---------

Co-authored-by: Antoine Stevan <44101798+amtoine@users.noreply.github.com>
2023-09-09 13:59:31 -05:00
Antoine Stevan
17abbdf6e0
allow into duration to take an integer amount of ns (#10286)
related to
-
https://discord.com/channels/601130461678272522/615329862395101194/1149717458786197524

# Description
because `1_234 | into datetime` takes an integer number of `ns` and
`1_234 | into filesize` takes an integer amount of bytes, i think `1_234
| into duration` should also be valid and see `1_234` as an integer
amount of `ns` 😋

# User-Facing Changes
## before
either
```nushell
1234 | into string | $in ++ "ns" | into duration
```
```nushell
1234 | $"($in)ns" | into duration
```
or
```nushell
1234 * 1ns
```
and
```nushell
> 1_234 | into duration
Error: nu::parser::input_type_mismatch

  × Command does not support int input.
   ╭─[entry #2:1:1]
 1 │ 1_234 | into duration
   ·         ──────┬──────
   ·               ╰── command doesn't support int input
   ╰────
```

## after
```nushell
> 1_234 | into duration
1µs 234ns
```

# Tests + Formatting
new example test
```rust
Example {
    description: "Convert a number of ns to duration",
    example: "1_234_567 | into duration",
    result: Some(Value::duration(1_234_567, span)),
}
```

# After Submitting
2023-09-09 13:49:08 -05:00
Darren Schroeder
40eca52ed5
allow update to use metadata (#10264)
# Description

This PR is an attempt to fix the `update` command so that it passes
along metadata. I'm not really sure I did this right, so please feel
free to point out where it's wrong.

The point is to be able to do something like this and have it respect
your LS_COLORS.
```
ls | update modified { format date }
```
### Before

![image](https://github.com/nushell/nushell/assets/343840/fc3eb207-4f6f-42b1-b5a4-87a1fe194399)

### After

![image](https://github.com/nushell/nushell/assets/343840/19d58443-7c88-4dd6-9532-1f45f615ac7b)


# User-Facing Changes
<!-- List of all changes that impact the user experience here. This
helps us keep track of breaking changes. -->

# 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
<!-- If your PR had any user-facing changes, update [the
documentation](https://github.com/nushell/nushell.github.io) after the
PR is merged, if necessary. This will help us keep the docs up to date.
-->
2023-09-09 13:47:42 -05:00
nibon7
7907dda8f7
Make cursor_shape optional (#10289)
# Description
There are several cursor shape related issues #7151 #9243 #7271 #8452
#10169, you can't disable the cursor shape feature even if you comment
out the entire `cursor_shape` block in the config.nu, and even worse,
when nushell exits with an error, the cursor shape can't be restored,
that is annoying.

This PR provides an opportunity to disable setting the cursor shape.

# User-Facing Changes
If you use the default config.nu, nothing changes, but if you comment
out `cursor_shape` block or set them to `inherit`, related cursor shape
will not be set.

# 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
<!-- If your PR had any user-facing changes, update [the
documentation](https://github.com/nushell/nushell.github.io) after the
PR is merged, if necessary. This will help us keep the docs up to date.
-->
2023-09-09 13:42:36 -05:00
David Laban
8501024546
signpost 'input list --types [key]' from 'keybindings list' (#10287)
Supercedes https://github.com/nushell/nushell/pull/10196

# Description

After reading
https://github.com/nushell/nushell/pull/10196#issuecomment-1703986359 I
added a signpost from `keybindings listen` to `input listen`

When I initially tried `input listen` it always immediately returned
with:
```
╭───────┬────────╮
│ type  │ focus  │
│ event │ gained │
╰───────┴────────╯
```

I added an example to `input listen --help` to suggest only listening to
key events

Initially I also included a `result` but it prints as:

```
  ╭───────────┬───────────────╮
  │ type      │ key           │
  │ key_type  │ char          │
  │ code      │ c             │
  │ modifiers │ [list 1 item] │
  ╰───────────┴───────────────╯
```

rather than:

```
╭───────────┬───────────────────────────────╮
│ type      │ key                           │
│ key_type  │ char                          │
│ code      │ c                             │
│           │ ╭───┬───────────────────────╮ │
│ modifiers │ │ 0 │ keymodifiers(control) │ │
│           │ ╰───┴───────────────────────╯ │
╰───────────┴───────────────────────────────╯
```
so I removed it.

# User-Facing Changes

<!-- List of all changes that impact the user experience here. This
helps us keep track of breaking changes. -->

* Example describing how to use `input list --types [key]` to listen for
keybindings.
* Signpost pointing at `use std input; input list --types [key]` from
`keybindings list`.

## After merging

It is probably worth:

a) signposting to the keybindings section of the book from both of these
subcommands (like I did in
https://github.com/nushell/nushell/pull/10193),
b) giving an example in the book of how to take the output from `input
listen --types [key]` and format it for including in `config nu`
c) there are not currently any examples in
crates/nu-utils/src/sample_config/default_config.nu for keybindings with
multiple modifiers. Should I add alt+backspace-in-macos-vscode as an
example (gets translated to `{ modifier: control_alt keycode: char_h }`
for historical reasons)?

---------

Co-authored-by: Antoine Stevan <44101798+amtoine@users.noreply.github.com>
2023-09-09 09:42:37 -05:00
Ian Manske
21d30d1e4d
Fix rm on macOS (#10282)
# Description

Removes the signal ignore for `SIGCHLD` added by #10021, since it
interferes with `rm` on macOS.
2023-09-08 19:03:30 -05:00
dependabot[bot]
eeaa65c8af
Bump itertools from 0.10.5 to 0.11.0 (#9524) 2023-09-08 23:12:45 +00:00
Stefan Holderbach
6754b8534e
Remove Cargo.lock in subcrates (#10280)
This should not be there in a workspace crate.

[See cargo
docs](https://doc.rust-lang.org/cargo/reference/workspaces.html)

Causes a spurious dependabot alert and is not easily updated.
2023-09-09 00:45:43 +02:00
Darren Schroeder
2ffff959fc
add 6 more table themes (#10279)
# Description

After looking at a users terminal that didn't support UTF-8, I wanted to
add some themes that may help them. Here's what they look like.

## psql

![image](https://github.com/nushell/nushell/assets/343840/67ac003a-72f1-4e2b-8bb0-244b70385d59)

## markdown

![image](https://github.com/nushell/nushell/assets/343840/a8f4a439-013b-48ee-b9e0-284ec47d1eef)

## dots
please excuse the different theme

![image](https://github.com/nushell/nushell/assets/343840/fb931650-cc64-4f0a-bf3d-ec736e0374ad)

## restructured

![image](https://github.com/nushell/nushell/assets/343840/80595a8e-f2b3-49dc-ad02-81e94bde5253)

## ascii_rounded

![image](https://github.com/nushell/nushell/assets/343840/42f0b8b2-1fd2-4ae5-b28c-477e83ded354)

## basic_compact

![image](https://github.com/nushell/nushell/assets/343840/5888b6b2-b9b8-48bc-963e-5a76ef246adc)


# User-Facing Changes
<!-- List of all changes that impact the user experience here. This
helps us keep track of breaking changes. -->

# 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
<!-- If your PR had any user-facing changes, update [the
documentation](https://github.com/nushell/nushell.github.io) after the
PR is merged, if necessary. This will help us keep the docs up to date.
-->
2023-09-08 16:34:36 -05:00
David Matos
fed4233db4
use uutils/coreutils cp command in place of nushell's cp command (#10097)
<!--
if this PR closes one or more issues, you can automatically link the PR
with
them by using one of the [*linking
keywords*](https://docs.github.com/en/issues/tracking-your-work-with-issues/linking-a-pull-request-to-an-issue#linking-a-pull-request-to-an-issue-using-a-keyword),
e.g.
- this PR should close #xxxx
- fixes #xxxx

you can also mention related issues, PRs or discussions!
-->

# Description
Hi. Basically, this is a continuation of the work that @fdncred started.
Given some nice discussions on #9463 , and [merged uutils
PR](https://github.com/uutils/coreutils/pull/5152) from @tertsdiepraam
we have decided to give the `cp` command the `crawl` stage as it was
named.

> [!NOTE] 
Given that the `uutils` crate has not made the release for the merged
PR, just make sure you checkout latest and put it in the required place
to make this PR work.

The aim of this PR is for is to see how to move forward using `uutils`
crate. In order to getting this started, I have made the current
`nushell cp tests` pass along with some extra ones I copied over from
the `uutils` repo.

With all of that being said, things that would be nice to decide, and
keep working on:

Crawl:
- Handling of certain `named` flags, with their long and short
forms(e.g. --update, --reflink, --preserve, etc), and using default
values. Maybe `-u` can already have a `default_missing_value`.
- Should we maybe just support one single option `switch` flags (see
`--backup` in code) as a contrast to the other named args.
- Complete test coverage from `uutils`. They had > 100 tests, and I
could only port like 12 as they are a bit time consuming given they
cannot be straight up copy pasted. Maybe we do not need all >100, but
maybe the more relevant to what we want.
- Refactor this code

Walk:
- Non fatal errors on `copy` from `utils`. Currently it just sends it to
stdout but errors have no span
- Better integration 

An added possibility is the addition of `SyntaxShape::OneOf()` for
`Named` arguments which was briefly mentioned in the discord server, but
that is still to be decided. This could greatly improve some of the
integration. This would enable something like `cp --preserve [all
timestamp]` or `cp --preserve all` to both work.

I did not want to keep holding on this, and wait till I was happy with
the code because I think its nice if everyone can start up and suggest
refactors, but the main important part now was getting it out the door,
as if I take my sweet time this will take way longer 😛

<!--
Thank you for improving Nushell. Please, check our [contributing
guide](../CONTRIBUTING.md) and talk to the core team before making major
changes.

Description of your pull request goes here. **Provide examples and/or
screenshots** if your changes affect the user experience.
-->

# User-Facing Changes
<!-- List of all changes that impact the user experience here. This
helps us keep track of breaking changes. -->

# Tests + Formatting

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

- [X] cargo fmt --all -- --check` to check standard code formatting
(`cargo fmt --all` applies these changes)
- [X] cargo clippy --workspace -- -D warnings -D clippy::unwrap_used` to
check that you're using the standard code style
- [X] cargo test --workspace` to check that all tests pass
- [X] 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
<!-- If your PR had any user-facing changes, update [the
documentation](https://github.com/nushell/nushell.github.io) after the
PR is merged, if necessary. This will help us keep the docs up to date.
-->

---------

Co-authored-by: Darren Schroeder <343840+fdncred@users.noreply.github.com>
2023-09-08 13:57:38 -05:00
Darren Schroeder
2f47263380
respect a users locale with the right prompt (#10273)
# Description

This PR tries to respect a person's locale setting with date/time in the
default right prompt. This also restores the right prompt so that it
does not load from the US default from reedline.

closes https://github.com/nushell/nushell/issues/9924


Here's the fromat reference if anyone thinks there's a better time
format. I could be talked into `%c` maybe.
https://docs.rs/chrono/latest/chrono/format/strftime/index.html

# User-Facing Changes
<!-- List of all changes that impact the user experience here. This
helps us keep track of breaking changes. -->

# 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
<!-- If your PR had any user-facing changes, update [the
documentation](https://github.com/nushell/nushell.github.io) after the
PR is merged, if necessary. This will help us keep the docs up to date.
-->
2023-09-08 12:24:42 -05:00
Darren Schroeder
4d5386635e
add more ps columns in Windows (#10275)
# Description

This PR adds a few more columns to `ps -l` on Windows. It would be good
to add these changes cross-platform in separate PRs. This PR also fixes
a bug where start time was calculated wrong.

I've added:
start_time
user
user_sid
priority


![image](https://github.com/nushell/nushell/assets/343840/cba16a17-ee70-46b5-9e6d-ef06641b264e)

# User-Facing Changes
<!-- List of all changes that impact the user experience here. This
helps us keep track of breaking changes. -->

# 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
<!-- If your PR had any user-facing changes, update [the
documentation](https://github.com/nushell/nushell.github.io) after the
PR is merged, if necessary. This will help us keep the docs up to date.
-->
2023-09-08 12:24:29 -05:00
Ian Manske
872eb2c3df
Restore initial foreground process group on exit (#10021)
# Description
When launching nushell interactively from another shell, the parent
shell usually gives us own our process group and handles restoring
control to itself. However, other programs that do not support job
control expect us to give control of the terminal back to them. This PR
makes it so that we record the initial foreground process group and
restore it when nushell exits. An "exit" can be from the `exit` command,
a panic, or a `SIGTERM` signal.

The changes in `terminal.rs` mostly follow [fish's
example](0874dd6a96/fish-rust/src/common.rs (L1634)).

# User-Facing Changes
Fixes interactions between nushell and other interactive CLI commands
(e.g., VIFM #10015).
2023-09-08 18:19:01 +02:00
Reilly Wood
e62a77a885
Start removing colour config from explore (#10270)
This PR removes the `explore.try.border_color` config item, and instead
always uses the `separator` colour (the one used for regular table
borders) from the current theme.

The PR also removes some unused `explore.config` bits from the default
config (I missed this in https://github.com/nushell/nushell/pull/10259).

### Future Work

This PR is intentionally small, I want to confirm that I'm on the right
track before I rip out more colour config from `explore`. If all goes
well, expect more PRs like this soon.

### Testing

I confirmed that this works by changing my `separator` colour in
`config.nu`, and also confirmed that nothing breaks if a user still has
`explore.try.border_color` in their config.
2023-09-08 07:34:47 -07:00
Darren Schroeder
9bca63ebef
update format date when using %x %X %r (#10272)
# Description

Apparently some strftime formats are already localized and when you
"double localize" them, they don't work. This PR fixes that so that %x
%X %r %c don't go through the localization step.

Example: %x %X
### Before
```nushell
❯ date now | format date "%x %X %p"
09/08/2023 08 AM
```
### After
```nushell
❯ date now | format date "%x %X %p"
09/08/23 08:09:14 AM
```

I started to make one format_datetime to rule them all but one returns a
string and one returns a value. If we convert to the string, we lose the
nice error messages. If we change to value, more code has to be changed
elsewhere. So, I decided to just leave two functions.

# User-Facing Changes
<!-- List of all changes that impact the user experience here. This
helps us keep track of breaking changes. -->

# 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
<!-- If your PR had any user-facing changes, update [the
documentation](https://github.com/nushell/nushell.github.io) after the
PR is merged, if necessary. This will help us keep the docs up to date.
-->
2023-09-08 08:59:05 -05:00
Stefan Holderbach
ae54dc862c
Move spellcheck config into .github folder (#10267)
Keep the `.typos.toml` out of the repo root for better readability

Also specify a version for the workflow to protect against breakage
2023-09-07 22:46:00 +02:00
Stefan Holderbach
5e951b2be9
Remove codecov.io setup (#10266)
As we currently don't run the coverage in CI and don't upload to
codecov.io remove the config to clean up the repo root.
2023-09-07 22:12:23 +02:00
Marshall Bruner
f78d57a703
feat: Search terms for use, while, and range (#5093) (#10265)
# Description

Add search terms to the commands: `use` ("environment", "import"),
`while` ("loop"), and `range` ("filter", "head", "tail").

# Tests + Formatting
All completed successfully.

---------

Co-authored-by: Stefan Holderbach <sholderbach@users.noreply.github.com>
2023-09-07 22:11:42 +02:00
Reilly Wood
f021be623e
Exit explore on ctrl+c/d/q (#10257)
Currently, `ctrl+z` is the command to exit `explore` no matter where you
are in the UI. IMO this is a bit unintuitive since that's usually used
to suspend a process.

After this change, `ctrl+c`, `ctrl+d`, and `ctrl+q` all work to exit
`explore`.

I think these are all shortcuts that users might try when attempting to
exit `explore`, and I think we might as well handle them all.
2023-09-07 19:47:17 +02:00
Reilly Wood
b6189879e3
explore: remove :config, :show-config, :tweak commands (#10259)
More trimming of underused `explore` functionality.

The `explore` command has subcommands that can be run like `:config` or
`:try` or whatnot. This PR removes the `:config`, `:show-config`, and
`:tweak` commands which are all for viewing+modifying config.

These are interesting commands and they were cool experiments, but
ultimately I don't think they fit with our plans for a simplified
`explore`. They'd need a lot more polish if we want to keep them and I
don't think we do. Happy to discuss if I've missed a good reason to keep
these.

cc @fdncred
2023-09-07 10:34:08 -05:00
Reilly Wood
c7c6445b03
Remove exit_esc and show_banner config from explore (#10258)
Removing 2 underused config options from `explore`.

`show_banner` controls whether `For help type :help"` is shown in the
message area when `explore is first launched. I don't think there's any
good reason not to show it, it's not a modal dialog or anything.

`exit_esc` controls whether to exit `explore` when `esc` is pressed and
we can't "go up" any further (or at least that's what it's supposed to
do, looking at the code I'm not so sure). IMO we don't need to make this
kind of basic interaction configurable unless there's a really good
reason.

## Context

`explore` is complicated and we want to overhaul its design. It will be
easier to make meaningful changes if `explore` is a little slimmer
first, so I'm trying to pare back unused/underused code and config as a
starting point.

I'm gonna be making more PRs like this, I'll try to keep them
small+self-contained.
2023-09-07 14:39:04 +02:00