Commit Graph
4 Commits
Author SHA1 Message Date
Jinwoo-H c46a06032e feat(cli): name the Orca that owns a workspace 'server'
orca server ... replaces orca environment ... and orca host name; --server and
--parent-server replace --environment and --parent-environment. Old spellings
stay accepted as input only: help, suggestions, errors and printed commands use
the new names, and giving both spellings different values is invalid_argument.
--host is still parsed but leaves help; orca host list/rm are hidden.

Listings name each row's owning server (server: / server=) from data already
returned, mapping a stored owner to a saved-server name only from this
computer's own store, and print server=unknown instead of guessing local. The
ports footer and search hits drop host=. CLI-built JSON uses server keys
(server, address.server, parent.server, unknown-server, servers, knownServers,
status target kind server). create/set print the same parent line as ps.
2026-11-09 02:12:19 -05:00
Neil f36751f79f Revert "feat(cli): create workspaces on environment recipes (#40108)" (#40184)
This reverts commit 862d315575.
2026-11-08 23:10:35 -08:00
NeilandJustas Brazauskas 862d315575 feat(cli): create workspaces on environment recipes (#40108)
* feat(cli): create workspaces on environment recipes

`orca worktree create --recipe <id>` creates the workspace on a new
environment from one of the repo's environment recipes, and
`orca vm recipe list` shows which recipes it accepts.

The desktop app on the provisioning computer runs the same main-side
provisioning the composer uses (write-ahead record, per-step timeouts,
cancellation), registers the project on the environment's Orca server,
then creates (orca-worktree) or adopts (provisioned-root) the
workspace there. Named progress steps ride on long-poll keepalive
frames, so the CLI prints them to stderr; older CLIs ignore them. A
failure after the VM exists destroys it through the shared cleanup and
returns recipe_workspace_create_failed with the failed step, the
recipe failure code and whether the destroy finished. Ctrl+C aborts
the request and destroys a started environment.

Paired devices cannot start a recipe create. Hosts advertise
worktree.recipe-create.v1 only when the desktop wires the recipe
service, so orcad and older hosts are refused before anything runs.

Based on #39228 by jutaz, a port of #28439.

Co-authored-by: Justas Brazauskas <brazauskasjustas@gmail.com>

* fix(cli): report the VM outcome when connecting fails; record no cross-host parent

* fix(vm-recipes): destroy the VM when recording its environment fails; reset exitCode in CLI tests

* perf(types): keep recipe orchestration out of the runtime's type graph

---------

Co-authored-by: Justas Brazauskas <brazauskasjustas@gmail.com>
2026-11-08 22:19:52 -08:00
NeilandFlorin Timbuc 239e17f84f feat(cli): add orca host rm, and clean up after every SSH target removal in main (#38774)
* feat(cli): add orca host rm for saved SSH targets

orca host rm --host ssh:<target-id> calls the runtime RPC ssh.removeTarget,
the same removeSshTarget the Settings UI reaches, so the tombstone keeps the
passive ~/.ssh/config import from adding the alias back.

Also document how the import treats Match blocks and Include files.

* fix(ssh): clear a removed target's live state and rename in main for every removal path

Removing an SSH target through ssh.removeTarget (orca host rm, paired clients)
left open tabs with stale reconnect state and kept the host's rename, because
only the Settings dialogs cleaned up after their own IPC call. The shared
removal mutation now drops the display-label override and tells the renderer,
which clears the target's state once for all callers.

Fixes #37949

* fix(ssh): clean up after a removed target in the teardown every removal ends in

A recipe VM's cleanup removes its SSH target through removeRegisteredSshTarget
directly, so the window kept that target's live state. The broadcast and the
rename drop now run there, after the target is gone.

Fixes #37949

* test(cli): keep host rm's retarget-flag refusal next to its other cases

Its old file was removed on main.

Fixes #37949

* refactor(cli): give the host, managed-server and VM handler groups their own module

Keeps the handler manifest under the file-length limit.

Fixes #37949

---------

Co-authored-by: Florin Timbuc <florin@sowild.design>
2026-11-06 06:23:46 -08:00