| Commit message (Collapse) | Author |
|---|
|
That helps remove duplicated code and rustc/cargo will now also show
dead code correctly.
|
|
## Summary
This PR enables the Atuin CLI to authenticate with Atuin Hub, unifying
authentication across CLI sync and Hub features (AI, runbooks, etc.).
### Key Changes
- **Dual auth support**: New `AuthToken` enum supports both `Bearer`
(Hub) and `Token` (legacy CLI) authentication
- **Smart protocol selection**: New `sync_protocol` setting
(`auto`/`hub`/`legacy`) determines auth method. By default,
`api.atuin.sh` uses Hub auth; custom sync addresses use legacy auth
- **Hub login flow**: `atuin login` now initiates an OAuth-like flow for
Hub users—generates a code, user authorizes in browser, CLI polls for
completion
- **Account linking**: After Hub auth, silently attempts to link
existing CLI sync account to Hub account for seamless migration
- **Graceful fallback**: `sync_auth_token()` prefers Hub token when
available, falls back to CLI session token
### Auth Flow
1. User runs `atuin login` (with default sync address)
2. CLI requests auth code from Hub, displays URL
3. User opens URL, logs in/registers on Hub
4. Hub attaches API token to code
5. CLI polls, receives token, saves as hub session
6. If user had existing CLI sync account, it's automatically linked
### Backward Compatibility
- Existing self-hosted users: unaffected (legacy auth via `Token`
header)
- Existing `api.atuin.sh` users: continue working with CLI session until
they run `atuin login`
- New users: go through Hub flow automatically
## Test Plan
- [ ] New user registration via Hub flow
- [ ] Existing CLI user can still sync without changes
- [ ] `atuin login` links CLI account to Hub account
- [ ] Self-hosted users unaffected by changes
- [ ] AI commands work after Hub auth
---------
Co-authored-by: Ellie Huxtable <ellie@elliehuxtable.com>
|
|
These files have been known to have corruption issues. SQLite will
perform better across filesystems for reads/writes across threads, and
will lock as expected.
I've also put the session file in there, though I'm 50/50 on it - I'll
be replacing it with keyring storage asap anyway.
The key file is _not_ included. It should ~never be changed, and should
be easy for the user to secure + manage themselves
In the future, instead of creating more files, we can just use this as a
kv store
Resolves https://github.com/atuinsh/atuin/issues/2336, resolves
https://github.com/atuinsh/atuin/issues/1650
## 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
|
|
|
|
* fix: typeerror in client sync code
Fixes #2645
This is really weird
1. I have not touched this code in _years_. It has not changed. In
recent rust versions, it has a typeerror (see linked issue)
2. This does not occur when running `cargo build`, in release mode or
otherwise. It only occurs with `cargo install`
3. I can't find any other occurences of this typeerror online - unsure
if it is a compiler regression? The code here is not very complex at
all.
* chore(clippy): remove unused imports
|
|
* chore: upgrade to 2024 edition
* ugh unsafe
* format
* nixxxxxxxxxxx why
|
|
* fix(daemon): do not try to sync if logged out
I've also added Settings::logged_in, as there are a few places where we
switch on login state.
* make session_token a function
|
|
* fix(dotfiles): allow clearing aliases, disable import
At the moment there are far too many edge cases to handle importing
aliases.
1. We need an interactive shell to print aliases. Without it, most
shells won't report much.
2. Many people have their shells print things on startup (graphics,
fortunes, etc). This could be detected as an attempt to set an alias.
Rather than spend the next year finding import edge cases, I'm
disabling it for now. There's probably a better way we can do this?
* clippy
|
|
I'd like to tidy up the root a little, and it's nice to have all the
rust crates in one place
|
|
Imagine the scenario where a sync fails. The function touched here will
never save the sync time.
That means that external to this function, the sync never happened. The
next command will try and start one immediately.
The happy scenario is that this succeeds. The unhappy scenario is that
this fails. Fails, and isn't saved, so we try again...
Should add proper backoff but this is a good start.
|
|
|
|
* clear history id
* fix nu
|
|
|
|
* fix #1236
* lints
|
|
* Add connect timeout and overall timeout
* Make it configurable
* Fix test
* Add docs
|
|
* replace chrono with time
* Fix test chrono usage
---------
Co-authored-by: Ellie Huxtable <ellie@elliehuxtable.com>
|
|
|
|
|
|
* Allow server configured page size
* Backwards compat via semver checks
* Correct header name
|
|
This can occur if history has been added + then deleted on a machine
before it has a chance to be synced to a new one.
|
|
* Delete all instances of a command
Our search command will de-dupe results by default. But... This isn't
great for deleting! You don't want to run it over-and-over-and-over
until all commands are deleted.
Loop the query, and keep on deleting what it returns until they are all
gone.
* Optimize delete upload
It was running a request for every element, on every sync lol
Only push a delete if needed
Future: push all deletes in one request
|
|
* Drop events. I'd still like to do them, but differently
* Start adding delete api stuff
* Set mailmap
* Delete delete delete
* Fix tests
* Make clippy happy
|
|
* Add event data structures
This adds the data structures required to start syncing events, rather
than syncing history directly.
Adjust event
Fix
Add event data structure to client
* Add server event table sql
* Add client event table migration
Adjust migration
* Insert into event table from client
* Add event merge function
Right now this just ensures we have the right amount of events given the
history we have
BUT it will also be used to merge CREATE/DELETE events, resulting in
history being deleted :)
* Make CI happy
* Adjust
* we don't limit history length any more
* Update atuin-client/src/database.rs
Co-authored-by: Conrad Ludgate <conradludgate@gmail.com>
* fix usage
* Fix typo
* New Rust, new clippy stuff
Co-authored-by: Conrad Ludgate <conradludgate@gmail.com>
|
|
|
|
* ignore JB IDEs
* tidy-up imports
* add rustfmt config
|
|
|
|
|
|
The data part of the add history request is actually a string. I don't
want to introduce any structure here, and would rather keep it as "just
a blob". Even if that blob has structure secretly!
My fault for missing this in the last review
|
|
* make everything a cow
* fmt + clippy
|
|
|
|
|
|
* Begin moving to sqlx for local too
* Stupid scanners should just have a nice cup of tea
Random internet shit searching for /.env or whatever
* Remove diesel and rusqlite fully
|
|
Also allow unique listing and more ergonomic cwd usage
|
|
* Switch to Cargo workspaces
Breaking things into "client", "server" and "common" makes managing the
codebase much easier!
client - anything running on a user's machine for adding history
server - handles storing/syncing history and running a HTTP server
common - request/response API definitions, common utils, etc
* Update dockerfile
|
|
* Switch to warp + sql, use async and stable rust
* Update CI to use stable
|
|
* Add encryption
* Add login and register command
* Add count endpoint
* Write initial sync push
* Add single sync command
Confirmed working for one client only
* Automatically sync on a configurable frequency
* Add key command, key arg to login
* Only load session if it exists
* Use sync and history timestamps for download
* Bind other key code
Seems like some systems have this code for up arrow? I'm not sure why,
and it's not an easy one to google.
* Simplify upload
* Try and fix download sync loop
* Change sync order to avoid uploading what we just downloaded
* Multiline import fix
* Fix time parsing
* Fix importing history with no time
* Add hostname to sync
* Use hostname to filter sync
* Fixes
* Add binding
* Stuff from yesterday
* Set cursor modes
* Make clippy happy
* Bump version
|
