Commit Graph

9 Commits

Author SHA1 Message Date
Yash Thakur
6d36941e55
Add completions.sort option (#13311) 2024-08-05 20:30:10 -04:00
Stefan Holderbach
e889679d42
Use nightly clippy to kill dead code/fix style (#12334)
- **Remove duplicated imports**
- **Remove unused field in `CompletionOptions`**
- **Remove unused struct in `nu-table`**
- **Clarify generic bounds**
- **Simplify a subtrait bound for `ExactSizeIterator`**
- **Use `unwrap_or_default`**
- **Use `Option` directly instead of empty string**
- **Elide unneeded clone in `to html`**
2024-03-30 09:17:28 +08:00
Ian Manske
c747ec75c9
Add command_prelude module (#12291)
# Description
When implementing a `Command`, one must also import all the types
present in the function signatures for `Command`. This makes it so that
we often import the same set of types in each command implementation
file. E.g., something like this:
```rust
use nu_protocol::ast::Call;
use nu_protocol::engine::{Command, EngineState, Stack};
use nu_protocol::{
    record, Category, Example, IntoInterruptiblePipelineData, IntoPipelineData, PipelineData,
    ShellError, Signature, Span, Type, Value,
};
```

This PR adds the `nu_engine::command_prelude` module which contains the
necessary and commonly used types to implement a `Command`:
```rust
// command_prelude.rs
pub use crate::CallExt;
pub use nu_protocol::{
    ast::{Call, CellPath},
    engine::{Command, EngineState, Stack},
    record, Category, Example, IntoInterruptiblePipelineData, IntoPipelineData, IntoSpanned,
    PipelineData, Record, ShellError, Signature, Span, Spanned, SyntaxShape, Type, Value,
};
```

This should reduce the boilerplate needed to implement a command and
also gives us a place to track the breadth of the `Command` API. I tried
to be conservative with what went into the prelude modules, since it
might be hard/annoying to remove items from the prelude in the future.
Let me know if something should be included or excluded.
2024-03-26 21:17:30 +00:00
Stefan Holderbach
86cd387439
Refactor and fix Config<->Value mechanism (#10896)
# Description
Our config exists both as a `Config` struct for internal consumption and
as a `Value`. The latter is exposed through `$env.config` and can be
both set and read.
Thus we have a complex bug-prone mechanism, that reads a `Value` and
then tries to plug anything where the value is unrepresentable in
`Config` with the correct state from `Config`.

The parsing involves therefore mutation of the `Value` in a nested
`Record` structure. Previously this was wholy done manually, with
indices.
To enable deletion for example, things had to be iterated over from the
back. Also things were indexed in a bunch of places. This was hard to
read and an invitation for bugs.

With #10876 we can now use `Record::retain_mut` to traverse the records,
modify anything that needs fixing, and drop invalid fields.

# Parts:

- Error messages now consistently use the correct spans pointing to the
problematic value and the paths displayed in some messages are also
aligned with the keys used for lookup.
- Reconstruction of values has been fixed for:
	- `table.padding`
	- `buffer_editor`
	- `hooks.command_not_found`
	- `datetime_format` (partial solution)
- Fix validation of `table.padding` input so value is not set (and
underflows `usize` causing `table` to run forever with negative values)
- New proper types for settings. Fully validated enums instead of
strings:
  - `config.edit_mode` -> `EditMode` 
  	- Don't fall back to vi-mode on invalid string
  - `config.table.mode` -> `TableMode`
- there is still a fall back to `rounded` if given an invalid
`TableMode` as argument to the `nu` binary
  - `config.completions.algorithm` -> `CompletionAlgorithm`
  - `config.error_style` -> `ErrorStyle`
    - don't implicitly fall back to `fancy` when given an invalid value.
- This should also shrink the size of `Config` as instead of 4x24 bytes
those fields now need only 4x1 bytes in `Config`
- Completely removed macros relying on the scope of `Value::into_config`
so we can break it up into smaller parts in the future.
- Factored everything into smaller files with the types and helpers for
particular topics.
- `NuCursorShape` now explicitly expresses the `Inherit` setting.
conversion to option only happens at the interface to `reedline`
2023-11-08 20:31:30 +01:00
JT
e36649f74b
Update path completions to handle spaces (#5419) 2022-05-03 12:37:38 +12:00
Richard
9771270b38
Fuzzy completion matching (#5320)
* Implement fuzzy match algorithm for suggestions

* Use MatchingAlgorithm for custom completions
2022-04-24 16:43:18 -05:00
Richard
e6a70f9846
Add MatchAlgorithm for completion suggestions (#5244)
* Pass completion options to each fetch() call

* Add MatchAlgorithm to CompletionOptions

* Add unit test for MatchAlgorithm

* Pass completion options to directory completer
2022-04-23 10:01:19 -05:00
Richard
0de289f6b7
Feature/refactor completion options (#5228)
* Copy completion filter to custom completions

* Remove filter function from completer

This function was a no-op for FileCompletion and CommandCompletion.
Flag- and VariableCompletion just filters with `starts_with` which
happens in both completers anyway and should therefore also be a no-op.
The remaining use case in CustomCompletion was moved into the
CustomCompletion source file.

Filtering should probably happen immediately while fetching completions
to avoid unnecessary memory allocations.

* Add get_sort_by() to Completer trait

* Remove CompletionOptions from Completer::fetch()

* Fix clippy lints

* Apply Completer changes to DotNuCompletion
2022-04-19 13:59:10 -05:00
Herlon Aguiar
13869e7d52
nu-cli: refactor completions (#5102) 2022-04-06 19:58:55 +12:00