* Revert "Revert "Try again: in unix like system, set foreground process while running external command (#6273)" (#6542)"
This reverts commit 2bb367f570.
* Make foreground job control hopefully work correctly
These changes are mostly inspired by the glibc manual.
* Fix typo in external command description
* Only restore tty control to shell when no fg procs are left; reuse pgrp
* Rework terminal acquirement code to be like fish
Co-authored-by: Darren Schroeder <343840+fdncred@users.noreply.github.com>
* Add "space" key to bind in vi normal mode
Implements #6586
No special logic to prevent you from binding it in other modes!
Needs a separate change to reedline to make it available in the default
listing of `keybindings list`.
* Update reedline to report the available `space`
Pulls in nushell/reedline#486
* Add a 'commandline' command for manipulating the current buffer
from `executehostcommand` keybindings. Inspired by fish:
https://fishshell.com/docs/current/cmds/commandline.html
* Update to development reedline
Includes nushell/reedline#472
Co-authored-by: sholderbach <sholderbach@users.noreply.github.com>
* Avoid update_last_command_context "No command run" error
When using `executehostcommand` bindings without having run actual user input commands yet,
update_last_command_context is guaranteed to fail. A function has been added to reedline
that allows checking for this case.
* Update to most recent reedline
Includes bugfixes around the (SQlite) history
Co-authored-by: sholderbach <sholderbach@users.noreply.github.com>
`rstest = 0.12` added support for asynchronous timeouts during testing
thus requiring a larger set of dependencies. Since `rstest = 0.14` this
can be disabled if not used.
Should keep build times for local or CI tests in check.
* Skeleton implementation
Lots and lots of TODOs
* Bootstrap simple CustomValue plugin support test
* Create nu_plugin_custom_value
* Skeleton for nu_plugin_custom_values
* Return a custom value from plugin
* Encode CustomValues from plugin calls as PluginResponse::PluginData
* Add new PluginCall variant CollapseCustomValue
* Handle CollapseCustomValue plugin calls
* Add CallInput::Data variant to CallInfo inputs
* Handle CallInfo with CallInput::Data plugin calls
* Send CallInput::Data if Value is PluginCustomValue from plugin calls
* Remove unnecessary boxing of plugins CallInfo
* Add fields needed to collapse PluginCustomValue to it
* Document PluginCustomValue and its purpose
* Impl collapsing using plugin calls in PluginCustomValue::to_base_value
* Implement proper typetag based deserialization for CoolCustomValue
* Test demonstrating that passing back a custom value to plugin works
* Added a failing test for describing plugin CustomValues
* Support describe for PluginCustomValues
- Add name to PluginResponse::PluginData
- Also turn it into a struct for clarity
- Add name to PluginCustomValue
- Return name field from PluginCustomValue
* Demonstrate that plugins can create and handle multiple CustomValues
* Add bincode to nu-plugin dependencies
This is for demonstration purposes, any schemaless binary seralization
format will work. I picked bincode since it's the most popular for Rust
but there are defintely better options out there for this usecase
* serde_json::Value -> Vec<u8>
* Update capnp schema for new CallInfo.input field
* Move call_input capnp serialization and deserialization into new file
* Deserialize Value's span from Value itself instead of passing call.head
I am not sure if this was correct and I am breaking it or if it was a
bug, I don't fully understand how nu creates and uses Spans. What should
reuse spans and what should recreate new ones?
But yeah it felt weird that the Value's Span was being ignored since in
the json serializer just uses the Value's Span
* Add call_info value round trip test
* Add capnp CallInput::Data serialization and deserialization support
* Add CallInfo::CollapseCustomValue to capnp schema
* Add capnp PluginCall::CollapseCustomValue serialization and deserialization support
* Add PluginResponse::PluginData to capnp schema
* Add capnp PluginResponse::PluginData serialization and deserialization support
* Switch plugins::custom_values tests to capnp
Both json and capnp would work now! Sadly I can't choose both at the
same time :(
* Add missing JsonSerializer round trip tests
* Handle plugin returning PluginData as a response to CollapseCustomValue
* Refactor plugin calling into a reusable function
Many less levels of indentation now!
* Export PluginData from nu_plugin
So plugins can create their very own serve_plugin with whatever
CustomValue behavior they may desire
* Error if CustomValue cannot be handled by Plugin
This PR adds support for an SQLite history via nushell/reedline#401
The SQLite history is enabled by setting history_file_format: "sqlite" in config.nu.
* somewhat working sqlite history
* Hook up history command
* Fix error in SQlitebacked with empty lines
When entering an empty line there previously was the "No command run"
error with `SqliteBackedHistory` during addition of the metadata
May be considered a temporary fix
Co-authored-by: sholderbach <sholderbach@users.noreply.github.com>
Release notes: https://github.com/nushell/reedline/releases/tag/v0.6.0
This release contains several bug fixes and improvements to the vi-emulation and documentation.
- Improvements to the vi-style keybindings (@sadmac7000):
- `w` now correctly moves to the beginning of the word.
- `e` to move to the end of the word.
- Bugfixes:
- Support terminal emulators that erroneously report a size of 0x0 by assuming a default size to avoid panics and draw nevertheless (@DhruvDh)
- Fix `ListMenu` layout calculations. Avoids scrolling bug when wrapping occurs due to the line numbering (@ahkrr)
- Avoid allocating to the total history capacity which can cause the application to go out of memory (@sholderbach)
- Documentation improvements including addition of documentation intended for reedline developers (@petrisch, @sholderbach)
Fixes#5593 (OOM introduced with #5587 when no config was present and an attempt was
made to allocate all memory in advance)
Includes also other changes to reedline:
- Vi word definition fixed and `w` and `e` work as expected
* Try removing debuginfo for ci builds
* oops, wrong inherits
* extra flag
* nextest doesn't support --profile in the same way
* try to allow for a ci-specific target
* Oops, run more tests
* database commands
* db commands
* filesystem opens sqlite file
* clippy error
* corrected error in ci file
* removes matrix flag from ci
* flax matrix for clippy
* add conditional compile for tests
* add conditional compile for tests
* correct order of command
* correct error msg
* correct typo
Changes to reedline since `v0.4.0`:
- vi normal mode `I` for inserting at line beginning
- `InsertNewline` edit command that can be bound to `Alt-Enter` if
desired to have line breaks without relying on the `Validator`
- `ClearScreen` will directly clear the visible screen. `Signal::CtrlL` has been
removed.
- `ClearScrollback` will clear the screen and scrollback. Can be used to
mimic macOS `Cmd-K` screen clearing. Helps with #5089
* Remove unused packages from base Cargo.toml
* Remove unused crossterm_winapi from nu-cli
* Remove unused dependencies from nu-system
* Remove unused dependencies from nu-test-support
* nu-completer with suggestions
* help menu with scrolling
* updates description rows based on space
* configuration for help menu
* update nu-ansi-term
* corrected test for update cells
* changed keybinding
* Refactor usage of is_perf_true to be a parameter passed around
* Move repl loop and command/script execution to nu_cli
* Move config setup out of nu_cli
* Update config_files.rs
* Update main.rs
Co-authored-by: JT <547158+jntrnr@users.noreply.github.com>
* Add profiling build profile and symbol strip
Stripping the symbols for the release build improves the size of the
binary significantly
Adds a custom build profile for performance profiling that includes all
symbols for analysis.
Can be used via
```
cargo build --profile profiling
```
* Retain a minimal backtrace
At the moment `crossterm` apparently has a regression decoding certain important key combinations on Windows.
Thus reedline reverted to the previous version.
Some changes are necessary to remove the need for `crossterm` in the use of `lscolors`.
Introduces two local conversion traits.
Additionally update the `Highlighter` API to support the cursor
position.
This will enable brace/statement match highlighting.
* query command with json, web, xml
* query xml now working
* clippy
* comment out web tests
* Initial work on query web
For now we can query everything except tables
* Support for querying tables
Now we can query multiple tables just like before, now the only thing
missing is the test coverage
* finish off
* comment out web test
Co-authored-by: Luccas Mateus de Medeiros Gomes <luccasmmg@gmail.com>
* Move fetch to extra
* Move byte stream code to a function instead of copying it twice
* Fix formatting issues
* Make fetch a default command
* Fix formatting
* Port fetch to engine-q
* Fix check for path as a string
* Add a timeout flag and fix some span issues
* Add a temporary fetch command that returns byte streams. Got rid of async stuff as we're using the blocking feature of tokio
* More tweaks for the bytestream
* Rewrite fetch using ByteStreams
* buffer read on bytes directly
Co-authored-by: Stefan Stanciulescu <contact@stefanstanciulescu.com>
* MathEval Variance and Stddev
* Fix tests and linting
* Typo
* Deal with streams when they are not tables
* First draft of these commands
* To MD
* To md and to html
* Fixed cargo and to_md
* `into_abbreviated_string` instead of `into_string`
* Changed how inner tables are displayed
* option to replace command same name
* moved order of custom value declarations
* arranged dataframe folders and objects
* sort help commands by name
* added dtypes function for debugging
* corrected name for dataframe commands
* command names using function
* custom value trait
* functions for custom value trait
* custom trait behind flag
* open dataframe command
* command to-df for basic types
* follow path for dataframe
* dataframe operations
* dataframe not default feature
* custom as default feature
* corrected examples in command
* Resolve rebase artifacts
* Remove leftover dependencies on removed feature
* Remove unnecessary 'pub'
* Start taking notes and fooling around
* Split canonicalize to two versions; Add TODOs
One that takes `relative_to` and one that doesn't.
More TODO notes.
* Merge absolutize to and rename resolve_dots
* Add custom absolutize fn and use it in path expand
* Convert a couple of dunce::canonicalize to ours
* Update nu-path description
* Replace all canonicalize with nu-path version
* Remove leftover dunce dependencies
* Fix broken autocd with trailing slash
Trailing slash is preserved *only* in paths that do not contain "." or
"..". This should be fixed in the future to cover all paths but for now
it at least covers basic cases.
* Use dunce::canonicalize for canonicalizing
* Alow cd recovery from non-existent cwd
* Disable removed canonicalize functionality tests
Remove unused import
* Break down nu-path into separate modules
* Remove unused public imports
* Remove abundant cow mapping
* Fix clippy warning
* Reformulate old canonicalize tests to expand_path
They wouldn't work with the new canonicalize.
* Canonicalize also ~ and ndots; Unify path joining
Also, add doc comments in nu_path::expansions.
* Add comment
* Avoid expanding ndots if path is not valid UTF-8
With this change, no lossy path->string conversion should happen in the
nu-path crate.
* Fmt
* Slight expand_tilde refactor; Add doc comments
* Start nu-path integration tests
* Add tests TODO
* Fix docstring typo
* Fix some doc strings
* Add README for nu-path crate
* Add a couple of canonicalize tests
* Add nu-path integration tests
* Add trim trailing slashes tests
* Update nu-path dependency
* Remove unused import
* Regenerate lockfile
On Windows, we used the `is-exeuctable` crate but on Unix, we
duplicated the check that it did, with one difference: We also
looked at whether or not it was a symlink.
The `is-executable` crate uses `std::fs::metadata` which follows
symlinks, so this scenario should never occur here, as it will
return the metadata for the target file.
Using the `is-executable` crate on both Unix and Windows lets us
make it non-optional. This lets us remove the `executable-support`
feature. (It is worth noting that this code didn't compile on
Windows when the `executable-support` feature was not specified.)
Right now, there is an alternate code path for `target_arch` being
`wasm32`. This isn't exactly correct as it should probably handle
something different for when the `target_os` is `wasi`.
Nothing used the `ptree` feature or optional dependency within
`nu-command` except to include it within the `version` output. This
may be related to when `nu-cli` also had a `ptree` feature, but
I'm not sure.
That leaves the code within `nu_plugin_tree` as the sole remaining
user of `ptree`, which is already covered by the feature `tree`
and included in the `version` output.
Seems we do `ctrl` feature checks in `nu-cli` and `nu-command`. We should find a better way to report the enabled features un the `version` command without using the conditionals (or somewhere else)
Added test cases that ensure that special characters in path names are passed
to external commands correctly. These cases have been implemented with rstest
to reuse existing test code.
We've relied on `clap` for building our cli app bootstrapping that figures out the positionals, flags, and other convenient facilities. Nu has been capable of solving this problem for quite some time. Given this and much more reasons (including the build time caused by `clap`) we start here working with our own.
