| Commit message (Collapse) | Author | Age |
|---|
| ... | |
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Remove the expensive and inaccurate `total_history` field from the API
index endpoint. The query `select sum(total) from
total_history_count_user` ran on every request but is no longer
relevant. The underlying table remains for per-user cached counts used
by `/sync/count`.
## Checks
- [x] I am happy for maintainers to push small adjustments to this PR,
to speed up the review cycle
- [x] I have checked that there are no existing pull requests for the
same thing
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Update reqwest from 0.12 to 0.13 and remove the built-in TLS termination
from atuin-server. Users should use a reverse proxy (nginx, caddy,
traefik) for TLS/HTTPS support instead.
This removes:
- axum-server and rustls dependencies
- The [tls] configuration section
- The launch_with_tls function
Also updates metrics-exporter-prometheus from 0.17 to 0.18.
The reverse proxy approach is standard and provides better flexibility
for certificate management. I'd rather keep our server stack as minimal
as possible.
<!-- Thank you for making a PR! Bug fixes are always welcome, but if
you're adding a new feature or changing an existing one, we'd really
appreciate if you open an issue, post on the forum, or drop in on
Discord -->
## Checks
- [ ] I am happy for maintainers to push small adjustments to this PR,
to speed up the review cycle
- [ ] I have checked that there are no existing pull requests for the
same thing
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Here is my case:
```sh
curl -X POST https://example.com/api \
-H "Content-Type: application/json" \
-d '{
"username": "name",
"password": "pass",
"details": {
"age": 30,
"location": "Earth"
}
}'
```
Using current release version i got:
```sh
curl -X POST https://example.com/api \\
-H "Content-Type: application/json" \\
-d '{\
"username": "name",\
"password": "pass",\
"details": {\
"age": 30,\
"location": "Earth"\
}\
}'
```
and i test with `echo \\`
(https://github.com/atuinsh/atuin/pull/100#issuecomment-836477081) it
also works;
there is similar PR https://github.com/atuinsh/atuin/pull/2390, it only
works on multiline with `\` suffix, in my case i got:
```sh
curl -X POST https://example.com/api \
-H "Content-Type: application/json" \
-d '{\
```
## Checks
- [x] I am happy for maintainers to push small adjustments to this PR,
to speed up the review cycle
- [ ] I have checked that there are no existing pull requests for the
same thing
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
resolves #1906
<!-- Thank you for making a PR! Bug fixes are always welcome, but if
you're adding a new feature or changing an existing one, we'd really
appreciate if you open an issue, post on the forum, or drop in on
Discord -->
## Checks
- [ ] I am happy for maintainers to push small adjustments to this PR,
to speed up the review cycle
- [ ] I have checked that there are no existing pull requests for the
same thing
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Configure the interactive search UI appearance. Resolves #998
```toml
[ui]
columns = ["duration", "time", "command"]
```
### `columns`
Default: `["duration", "time", "command"]`
Columns to display in the interactive search, from left to right. The
selection
indicator (`" > "`) is always shown first implicitly.
Each column can be specified as:
- A simple string (uses default width): `"duration"`
- An object with type and optional width/expand: `{ type = "directory",
width = 30 }`
#### Available column types
| Column | Default Width | Description |
| --------- | ------------- |
----------------------------------------------- |
| duration | 5 | Command execution duration (e.g., "123ms") |
| time | 8 | Relative time since execution (e.g., "59m ago") |
| datetime | 16 | Absolute timestamp (e.g., "2025-01-22 14:35") |
| directory | 20 | Working directory (truncated if too long) |
| host | 15 | Hostname |
| user | 10 | Username |
| exit | 3 | Exit code (colored by success/failure) |
| command | * | The command itself (expands by default) |
#### Column options
- **type**: The column type (required when using object format)
- **width**: Custom width in characters (optional, uses default if not
specified)
- **expand**: If `true`, the column fills remaining space. Default is
`true` for `command`, `false` for others. Only one column should have
`expand = true`.
#### Examples
```toml
# Minimal - more space for commands
columns = ["duration", "command"]
# With custom directory width
columns = ["duration", { type = "directory", width = 30 }, "command"]
# Show host for multi-machine sync users
columns = ["duration", "time", "host", "command"]
# Show exit codes prominently
columns = ["exit", "duration", "command"]
# Make directory expand instead of command
columns = ["duration", "time", { type = "directory", expand = true }, { type = "command", expand = false }]
```
|
| | | |
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
<!-- Thank you for making a PR! Bug fixes are always welcome, but if
you're adding a new feature or changing an existing one, we'd really
appreciate if you open an issue, post on the forum, or drop in on
Discord -->
## Checks
- [x] I am happy for maintainers to push small adjustments to this PR,
to speed up the review cycle
- [x] I have checked that there are no existing pull requests for the
same thing
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
<!-- Thank you for making a PR! Bug fixes are always welcome, but if
you're adding a new feature or changing an existing one, we'd really
appreciate if you open an issue, post on the forum, or drop in on
Discord -->
## Checks
- [ ] I am happy for maintainers to push small adjustments to this PR,
to speed up the review cycle
- [ ] I have checked that there are no existing pull requests for the
same thing
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
<!-- Thank you for making a PR! Bug fixes are always welcome, but if
you're adding a new feature or changing an existing one, we'd really
appreciate if you open an issue, post on the forum, or drop in on
Discord -->
Hello. Small PR to add two new columns to the command history, the
`session` and the `uuid` for the history. Having these identifiers
available can help people do more advanced scripting & comprehension of
their history.
## Checks
- [x] I am happy for maintainers to push small adjustments to this PR,
to speed up the review cycle
- [x] I have checked that there are no existing pull requests for the
same thing
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
<!-- Thank you for making a PR! Bug fixes are always welcome, but if
you're adding a new feature or changing an existing one, we'd really
appreciate if you open an issue, post on the forum, or drop in on
Discord -->
when i run `atuin wrapped` currently, my aliases get in the way of
properly calculating my wrapped.
Ex: `ga` -> `git add` -- not added to the `git` count
This PR changes the wrapped calculation to pull the aliases managed by
`atuin dotfiles` and expand aliased commands to calculate them properly.
### Performance
Tested with three shots each
Before change: `0.064` to `0.077` seconds
After change: `0.075` to `0.085` seconds
## Checks
- [x] I am happy for maintainers to push small adjustments to this PR,
to speed up the review cycle
- [x] I have checked that there are no existing pull requests for the
same thing
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This PR does a few things:
- Includes a fix by @sususu98 (added as co-author):
- Fix: #3032
- Issue: https://github.com/PowerShell/PSReadLine/issues/5047
- Adds a couple `catch` blocks for better error safety in
`PSConsoleHostReadLine`.
- Allows reloading the Atuin module on v7+ if it's already loaded, which
could be useful when updating Atuin.
- Refactors the code to clean it up.
I'll be AFK next week, so I preferred to submit this sooner than later.
Closes #3032
## Checks
- [x] I am happy for maintainers to push small adjustments to this PR,
to speed up the review cycle
- [x] I have checked that there are no existing pull requests for the
same thing
---------
Co-authored-by: sususu <suchangshan@foxmail.com>
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This runs `atuin history end` in the background, since it could delay
the next prompt when syncing.
Sorry I didn't realize this command could be slow (I currently don't use
sync).
I went for the .NET [`Process`
class](https://learn.microsoft.com/en-us/dotnet/api/system.diagnostics.process),
which is not pretty, but the other solutions I thought about all had
their own issues (`Start-Process` needs a file to redirect the output so
you can't just ignore it, `Start-Job` creates job objects that will
linger, `&` does the same but doesn't work on PS 5.1, ...). I'm
surprised I couldn't find a nice way to do the equivalent of `command &
>/dev/null` in PowerShell. π
This replaces #3033
## Checks
- [x] I am happy for maintainers to push small adjustments to this PR,
to speed up the review cycle
- [x] I have checked that there are no existing pull requests for the
same thing
|
| | |
| |
| |
| | |
It takes up a very large amount of our query time, despite having been
replaced for >1yr now. We will be dropping support for it on atuin hub
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Support for routing read queries to read replicas for Postgres
We have very high database usage these days, and now run shell history
sync off of [Planetscale](https://planetscale.com/)
This setup gives us 2x read replicas, meaning we can reduce load on the
primary
I doubt this is required for anyone else's setup - lmk if so.
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Do not just pick first from search.filters, consider all options,
filtering based on git-root and workspaces config
Fix: atuinsh/atuin#2700
Might Fix: atuinsh/atuin#2536
Wrote a few simple unit tests, and tried it by running the client
locally. Not sure about which of "[search] filters" vs "workspaces"
should be more respected, so i made it so if workspaces = false,
regardless of whats in "[search] filters" workspace gets filtered out.
First time looking at atuins code, let me know if this needs any
changes.
edit: (from my post here:
https://github.com/atuinsh/atuin/issues/2536#issuecomment-2808053862)
```
workspaces = true
[search]
filters = [ "workspace", "directory", "session", "global" ]
```
| `^R` for the ... time| in a git repo | any other dir|
| ------------- | ------------- | ------------- |
| first | workspace | directory|
| second | directory | session|
| third| session | global|
| fourth | global | directory |
| fifth | workspace | session|
| sixth | directory | global |
<!-- Thank you for making a PR! Bug fixes are always welcome, but if
you're adding a new feature or changing an existing one, we'd really
appreciate if you open an issue, post on the forum, or drop in on
Discord -->
## Checks
- [X] I am happy for maintainers to push small adjustments to this PR,
to speed up the review cycle
- [X] I have checked that there are no existing pull requests for the
same thing
- #1655 might be related, but it seems to also implement code that is
already merged, and has been open for over a year, hope its ok to
suggest this
---------
Co-authored-by: Michelle Tilley <michelle@michelletilley.net>
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Relates to #2520 - Gracefully handle update check failures instead of
crashing the entire interactive search UI.
## Checks
- [x] I am happy for maintainers to push small adjustments to this PR,
to speed up the review cycle
- [x] I have checked that there are no existing pull requests for the
same thing
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
<!-- Thank you for making a PR! Bug fixes are always welcome, but if
you're adding a new feature or changing an existing one, we'd really
appreciate if you open an issue, post on the forum, or drop in on
Discord -->
## Checks
- [x] I am happy for maintainers to push small adjustments to this PR,
to speed up the review cycle
- [x] I have checked that there are no existing pull requests for the
same thing
User can disable colors via `NO_COLOR=1` environment variable if needed
Output of `atuin -h`:
| before | after | `NO_COLOR=1` |
| - | - | - |
| <img width="741" height="768" alt="atuin-before"
src="https://github.com/user-attachments/assets/01921ba0-e94d-4f0a-ac85-b170516e41bc"
/> | <img width="741" height="768" alt="atuin-colors"
src="https://github.com/user-attachments/assets/59d507e1-9a33-4091-bed7-44eef889a72d"
/> | <img width="741" height="768" alt="atuin-no-color"
src="https://github.com/user-attachments/assets/e6b067c6-cf17-4909-bdbf-cc2bca830244"
/> |
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
<!-- Thank you for making a PR! Bug fixes are always welcome, but if
you're adding a new feature or changing an existing one, we'd really
appreciate if you open an issue, post on the forum, or drop in on
Discord -->
The original implementation of the `replxx` importer, from PR #2024,
only supported `.histfile` as a filename for the history file, since
there is no default filename. However, the replxx codebase does use
`replxx_history.txt` as an example history filename (see
[here](https://github.com/AmokHuginnsson/replxx/blob/release-0.0.4/examples/c-api.c#L183)
and
[here](https://github.com/AmokHuginnsson/replxx/blob/release-0.0.4/examples/cxx-api.cxx#L383)),
so this patch adds support for that as an alternative filename.
The implementation was modeled after [the
one](https://github.com/atuinsh/atuin/blob/v18.10.0/crates/atuin-client/src/import/zsh.rs#L29-L44)
currently used for the `zsh` history file importer, which also means the
`replxx` importer now produces a human-friendly error if no history file
is found in the expected path(s).
## Checks
- [x] I am happy for maintainers to push small adjustments to this PR,
to speed up the review cycle
- [x] I have checked that there are no existing pull requests for the
same thing
/cc @amosbird who introduced this importer in #2024.
|
| | | |
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
After `aws-lc-rs 0.15` it seems the illumos issue is resolved, so
reviving the `metrics` et.al. update PR, specifically:
- Update `metrics` to 0.24 and `metrics-exporter-prometheus` to 0.17
- Drop the `ring` feature from `rustls`
- Update `reqwest` to 0.12 (dropping `rustls 0.21` from the lock file)
There still seem to be `ring` dependencies, but not sure if these can be
dropped
---------
Co-authored-by: Helmut K. C. Tessarek <tessarek@evermeet.cx>
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
<!-- Thank you for making a PR! Bug fixes are always welcome, but if
you're adding a new feature or changing an existing one, we'd really
appreciate if you open an issue, post on the forum, or drop in on
Discord -->
## Checks
- [x] I am happy for maintainers to push small adjustments to this PR,
to speed up the review cycle
- [x] I have checked that there are no existing pull requests for the
same thing
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
<!-- Thank you for making a PR! Bug fixes are always welcome, but if
you're adding a new feature or changing an existing one, we'd really
appreciate if you open an issue, post on the forum, or drop in on
Discord -->
Fixed typo in comment.
## Checks
- [x] I am happy for maintainers to push small adjustments to this PR,
to speed up the review cycle
- [x] I have checked that there are no existing pull requests for the
same thing
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
<!-- Thank you for making a PR! Bug fixes are always welcome, but if
you're adding a new feature or changing an existing one, we'd really
appreciate if you open an issue, post on the forum, or drop in on
Discord -->
## Checks
- [x] I am happy for maintainers to push small adjustments to this PR,
to speed up the review cycle
- [x] I have checked that there are no existing pull requests for the
same thing
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
<!-- Thank you for making a PR! Bug fixes are always welcome, but if
you're adding a new feature or changing an existing one, we'd really
appreciate if you open an issue, post on the forum, or drop in on
Discord -->
## Checks
- [ ] I am happy for maintainers to push small adjustments to this PR,
to speed up the review cycle
- [ ] I have checked that there are no existing pull requests for the
same thing
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This adds a workaround for an issue with a Bash-5.1 bug reported by
@MaxenceG2M at
https://github.com/atuinsh/atuin/issues/962#issuecomment-3451132291. See
the added code comments for details.
## Checks
- [x] I am happy for maintainers to push small adjustments to this PR,
to speed up the review cycle
- [x] I have checked that there are no existing pull requests for the
same thing
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This patch tries to port a change for zsh-autosuggestion in #2780 to
suppress the error messages in typing the command. The same issue
happens with ble.sh, so we want to suppress the error message in the
same way as with zsh-autosuggestion.
## Checks
- [x] I am happy for maintainers to push small adjustments to this PR,
to speed up the review cycle
- [x] I have checked that there are no existing pull requests for the
same thing
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
(#2977)
In #2953, I chose `\e[0;<n>A` as special key sequences for
`enter_accept` in Bash, but I realized that the sequence `\e[0;<n>A`
doesn't actually work in the `vi-insert` keymap. I instead suggest using
the key sequences of the form `\C-x\C-_A0\a` (99ddcbf7), which is
unlikely to conflict with the users' keybindings. In addition, for Bash
<= 3.2, I also noticed that we cannot insert `$READLINE_LINE` in the
command line when we are in the `vi-command` keymap. I suggest switching
from `vi-command` to `vi-insert` temporarily (e6c4e9fd).
## Checks
- [x] I am happy for maintainers to push small adjustments to this PR,
to speed up the review cycle
- [x] I have checked that there are no existing pull requests for the
same thing
|
| |/
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
A user mentioned seeing the "UPGRADE" wording, and then expecting `atuin
upgrade` to work, when in fact the correct command is `atuin update`
Easy fix, wording adjusted
<!-- Thank you for making a PR! Bug fixes are always welcome, but if
you're adding a new feature or changing an existing one, we'd really
appreciate if you open an issue, post on the forum, or drop in on
Discord -->
## Checks
- [ ] I am happy for maintainers to push small adjustments to this PR,
to speed up the review cycle
- [ ] I have checked that there are no existing pull requests for the
same thing
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This adds PowerShell support π
I built this script around @lzybkr's
[prototype](https://github.com/atuinsh/atuin/issues/84#issuecomment-1689168533),
so I added him as co-author (I hope that's ok). I wouldn't know where to
start without his contribution.
I'm not a PowerShell expert, so this was a nice opportunity to learn
some stuff. I think it's ok, but I would appreciate if someone more
knowledgeable in the matter could review this though.
It would be nice if other PowerShell users could test this with their
configs and report any issues. I wouldn't be surprised if there are some
remaining bugs or missing features.
Fixes #84
## Installation
If you'd like to test this, you can install the `atuin` from this PR by
running:
```powershell
cargo install --git https://github.com/ltrzesniewski/atuin.git --branch powershell-pr
```
Then, add the following to your PowerShell profile file (whose path is
in `$PROFILE`) and restart the shell:
```powershell
atuin init powershell | Out-String | Invoke-Expression
```
This requires `atuin` to be in the path and the
[PSReadLine](https://github.com/PowerShell/PSReadLine) module to be
installed, which is the case by default.
## Tests
I tested this on the following:
- PowerShell 7.4.6 / PSReadLine 2.3.5 / Windows (the latest one)
- PowerShell 7.5.1 / PSReadLine 2.3.6 / Windows (the latest one)
- PowerShell 5.1.22621.4391 / PSReadLine 2.0.0 / Windows (the one
shipped with Windows)
- PowerShell 7.4.6 / PSReadLine 2.3.5 / Ubuntu WSL (strangely, it didn't
behave exactly like the Windows version)
- PowerShell 7.5.1 / PSReadLine 2.3.6 / Ubuntu WSL
I also tested this with and without my custom [Oh My
Posh](https://ohmyposh.dev/) prompt. It works fine in both cases,
~except that since my OMP config contains `"newline": true`, my prompt
is multiline and shifts downwards by a single line on each `atuin search
-i` invocation. This can be adjusted with the
`$env:ATUIN_POWERSHELL_PROMPT_OFFSET` environment variable (e.g. I set
mine to `-1` to account for the additional prompt line).~ (this variable
is now auto-initialized).
## Checks
- [x] I am happy for maintainers to push small adjustments to this PR,
to speed up the review cycle
- [x] I have checked that there are no existing pull requests for the
same thing
---------
Co-authored-by: Jason Shirk <jasonsh@microsoft.com>
|
| |
|
|
|
|
|
|
|
|
|
|
| |
<!-- Thank you for making a PR! Bug fixes are always welcome, but if
you're adding a new feature or changing an existing one, we'd really
appreciate if you open an issue, post on the forum, or drop in on
Discord -->
## Checks
- [ ] I am happy for maintainers to push small adjustments to this PR,
to speed up the review cycle
- [ ] I have checked that there are no existing pull requests for the
same thing
|
| |
|
|
|
| |
Reverts atuinsh/atuin#2932
Resolves #2960
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
# Summary
Adjust the Nushell script to check the version and enable accept support
when Nushell 0.106.0 or greater is found.
See: https://github.com/nushell/nushell/pull/16193
<!-- Thank you for making a PR! Bug fixes are always welcome, but if
you're adding a new feature or changing an existing one, we'd really
appreciate if you open an issue, post on the forum, or drop in on
Discord -->
## Checks
- [x] I am happy for maintainers to push small adjustments to this PR,
to speed up the review cycle
- [x] I have checked that there are no existing pull requests for the
same thing
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Fix panic when running "atuin stats" with an empty window size:
$ RUST_BACKTRACE=1 atuin stats year -n0
thread 'main' panicked at crates/atuin-history/src/stats.rs:278:14:
window size must be non-zero
stack backtrace:
0: __rustc::rust_begin_unwind
1: core::panicking::panic_fmt
2: core::option::expect_failed
3: atuin_history::stats::compute
4: atuin::command::client::Cmd::run_inner::{{closure}}
5: tokio::runtime::context::scoped::Scoped<T>::set
6: tokio::runtime::scheduler::current_thread::CoreGuard::block_on
7: tokio::runtime::context::runtime::enter_runtime
8: atuin::command::AtuinCmd::run
9: atuin::main
note: Some details are omitted, run with `RUST_BACKTRACE=full` for a
verbose backtrace.
The ngram window size had a default value of 1 (indicating that perhaps
this case was considered), but did not validate the parsed value when a
non-default value was given.
Such validation can either occur:
1. during clap argument parsing
2. after clap parses arguments, but before computing statistics
If we were to do this after clap finishes parsing arguments, this would
be a runtime check in run(), but it seems better to return an error as
soon as possible.
It's possible to do this without a custom parsing function via
clap::value_parser! and specifying a range:
clap::value_parser!(u64).range(1..)
However, this does not appear to support custom error messages so we
would be left with the default message specifying an invalid range.
Therefore, add a custom parser function and return an error when the
value cannot be parsed as a valid ngram value.
<!-- Thank you for making a PR! Bug fixes are always welcome, but if
you're adding a new feature or changing an existing one, we'd really
appreciate if you open an issue, post on the forum, or drop in on
Discord -->
## Checks
- [x] I am happy for maintainers to push small adjustments to this PR,
to speed up the review cycle
- [x] I have checked that there are no existing pull requests for the
same thing
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
### What does this PR do?
Adds simple navigation to the inspector, to explore a session starting
from a single command. This creates a new user flow, where a user can
find a history entry in the interactive view (in, say, Global mode), and
hit Ctrl+o to navigate back and forward through that command's session.
IMAGINED USE-CASE: I remembered that I did a sequence of git steps but I
can't remember the order and forgot to document it. I remember that
`reflog` was involved and want to see the actual sequence, and only
those commands.
IMAGINED USE-CASE: I used a curl command to get my IP address for
greenlisting before I connected to the bastion server `abc.xyz` over SSH
- I could easily find the SSH command with abc.xyz, and go back one step
in the session, but without this change, scrolling through all my curl
commands ever run to find a forgotten URL/domain would be too much work.
Since this gives the inspector tab a broader purpose than viewing
analytics, it needs to function even when there are not enough screen
rows for charts -- hence, this PR also introduces an ultracompact mode
for the inspector that _just_ shows the neighbouring history commands
(as simple scrolling three-entry list, with no panes) if there are fewer
than `auto_hide_height` rows (default: 8). Otherwise, the inspector
behaves as normal, except that Up / Down will change the focused command
by navigating through the session. That means there is no "compact" mode
for the inspector - when the interactive search is compact (but not
ultracompact), the inspector shows its usual chart view.
The UX for this could be improved - to keep this PR as lean as it
realistically can be, I have tried to keep the flow very minimal, but a
follow-up PR could introduce some tooltips, nicer ultracompact
formatting, etc.
A minor QoL improvement that comes with this - since I had to deal with
bold text and would otherwise have need a theming exception, I took the
opportunity to ensure the theme engine sets styles completely (so a
theme can have bold), not just colours. To limit scope creep, I do not
add TOML syntax so (for now) you can only customize colours from config
files, but it means that default-bold text (etc.) can now use the
theming engine if the code-defined default Meaning is bolded.
Key changes:
* introduces a simplified inspector tab, with only previous-current-next
commands as rows, in compact mode
* allows navigation through session history within the inspector (so
compact inspector view is still useful)
It also (see comments below):
* makes `compact` into `compactness`, an enum (to better standardize
across inspector/interactive)
* makes the inspector _only_ change layout for ultracompact mode, which
is still compact+(height<8)
* clippy's complexity limit wanted draw split up a little, so not sure
if this is a reasonable minimal way to do so for now
* adds a `(none)` theme to the theming to enable output testing without
styling
* ~~additional tests, although keen for input on how best to do these~~
one functional test, as a starting point
* ~~documentation~~ [minor doc changes
only](https://github.com/atuinsh/docs/pull/72), as I am not sure there
is much to say
<!-- Thank you for making a PR! Bug fixes are always welcome, but if
you're adding a new feature or changing an existing one, we'd really
appreciate if you open an issue, post on the forum, or drop in on
Discord -->
_Was stacked on #2357, which is now in `main`_
## Checks
- [x] I am happy for maintainers to push small adjustments to this PR,
to speed up the review cycle
- [x] I have checked that there are no existing pull requests for the
same thing
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
<!-- Thank you for making a PR! Bug fixes are always welcome, but if
you're adding a new feature or changing an existing one, we'd really
appreciate if you open an issue, post on the forum, or drop in on
Discord -->
This PR introduces a mechanism to use Readline's `accept-line` to run
the user command properly. The idea is described in [the code
comment](https://github.com/atuinsh/atuin/pull/2953/files#diff-57afeb258339de1b14a8dd3fdc88d1a0e192fd186706e570c44c3ef41f7a8c6dR362-R382)
in the added code. This naturally fixes #2935 because Readline's
`accept-line` also performs the necessary keymap transition.
This PR also fixes the behavior of <kbd>tab</kbd> and <kbd>enter</kbd>
with `enter_accept = false` `in Bash <= 3.2. In the previous
implementation, the selected command was lost in Bash 3.2, but this PR
correctly inserts the selected command into the command line buffer.
This PR adds a utility `atuin-bind` to make it easier to define custom
keybindings. The default bindings are also set up by the new function
`atuin-bind` now. This new function `atuin-bind` arranges all
non-trivial setups to make it possible to call Readline's `accept-line`.
The old mechanism using `__atuin_accept_line` is kept for existing users
who set up custom keybindings (without using the new function
`atuin-bind`).
## Checks
- [x] I am happy for maintainers to push small adjustments to this PR,
to speed up the review cycle
- [x] I have checked that there are no existing pull requests for the
same thing
|
| |
|
|
|
|
|
|
|
|
|
|
| |
<!-- Thank you for making a PR! Bug fixes are always welcome, but if
you're adding a new feature or changing an existing one, we'd really
appreciate if you open an issue, post on the forum, or drop in on
Discord -->
## Checks
- [x] I am happy for maintainers to push small adjustments to this PR,
to speed up the review cycle
- [x] I have checked that there are no existing pull requests for the
same thing
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This adds an `atuin import powershell` command.
Of course, it is related to #2543 but I'm submitting it as a separate PR
since the code is self-contained and simple enough, and the feature
could be useful on its own.
/cc @ajn142 who [requested
it](https://github.com/atuinsh/atuin/issues/84#issuecomment-3091692807).
## Checks
- [x] I am happy for maintainers to push small adjustments to this PR,
to speed up the review cycle
- [x] I have checked that there are no existing pull requests for the
same thing
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This adds the commit SHA to the following:
- `atuin doctor`
- `atuin info` (which already displays the version)
- `atuin --version` (but not `atuin -V` in order not to affect `atuin
--help`)
I'm submitting this because I had issue reports in #2543 which were
already resolved, so being able to easily ask for the commit id would
have been helpful, as the version number isn't meaningful in a PR.
Also, I suppose the info should have been included in `atuin doctor` in
the first place, so that probably fixes an oversight.
## Checks
- [x] I am happy for maintainers to push small adjustments to this PR,
to speed up the review cycle
- [x] I have checked that there are no existing pull requests for the
same thing
|
| |
|
|
|
|
|
|
|
|
|
|
| |
<!-- Thank you for making a PR! Bug fixes are always welcome, but if
you're adding a new feature or changing an existing one, we'd really
appreciate if you open an issue, post on the forum, or drop in on
Discord -->
## Checks
- [ ] I am happy for maintainers to push small adjustments to this PR,
to speed up the review cycle
- [ ] I have checked that there are no existing pull requests for the
same thing
|
| |
|
|
|
|
|
|
|
|
| |
I wanted to use `command_chaining` to write a pipeline (`cmdA | cmdB`)
and was genuinely surprised it didn't work.
## Checks
- [x] I am happy for maintainers to push small adjustments to this PR,
to speed up the review cycle
- [x] I have checked that there are no existing pull requests for the
same thing
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This change introduces (optional) acceptance keys of Backspace and Left
Arrow, when at the start of a line. These two are common muscle memory
actions for users.
The configuration defaults to false so as not to disrupt existing user
patterns.
This also adds a test that exercises the various acceptance modes, which
as it turns out was quite easy to do.
I discussed this on discord where [Ellie suggested I raised an
issue](https://discord.com/channels/954121165239115808/1421180955657244703/1422642337481228400),
but I felt like a PR would be more tangiable. I've tested this locally
and I'm very happy with how these keys work, it fits my needs well.
`exit_past_line_start` and `accept_past_line_start` can technically
co-exist. When this happens `accept_past_line_start` takes precedence.
Is this okay, or should we reconsider the config? Perhaps
`acceptance_keys = []` would be better here? I'm very open to changes
here.
<!-- Thank you for making a PR! Bug fixes are always welcome, but if
you're adding a new feature or changing an existing one, we'd really
appreciate if you open an issue, post on the forum, or drop in on
Discord -->
## Checks
- [x] I am happy for maintainers to push small adjustments to this PR,
to speed up the review cycle
- [x] I have checked that there are no existing pull requests for the
same thing
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
<!-- Thank you for making a PR! Bug fixes are always welcome, but if
you're adding a new feature or changing an existing one, we'd really
appreciate if you open an issue, post on the forum, or drop in on
Discord -->
## Checks
- [X] I am happy for maintainers to push small adjustments to this PR,
to speed up the review cycle
- [X] I have checked that there are no existing pull requests for the
same thing
Hopefully this one is mergeable as is, unlike the 1.89 upgrade.
Fixes: #2915
|
| | |
|
| | |
|
| | |
|
| |
|
|
| |
The command chaining feature can be implemented without the prefix.
|
| | |
|
| | |
|
