mirror of
https://github.com/multica-ai/multica.git
synced 2026-08-12 19:06:06 +02:00
From Emacs's review of #6378. All seven findings reproduced. The two blocking server bugs were the same shape: Mika is one agent per workspace but sessions and ownership are per member. - StartMikaOnboarding required the agent owner, so the member who lost the CreateMikaAgent race — the race that handler's advisory lock exists to survive — got a valid Mika, opened a valid session, and then a 403. Mika is created workspace-visible and workspace-invocable; the session gate and canInvokeAgent were already the checks that matter. - The onboarding session was resolved client-side by listing sessions and creating one on a miss, matched on the localized title. LockWorkspaceForChatSessionCreate is FOR KEY SHARE precisely so concurrent creators do not block, so two tabs each opened their own conversation with its own kickoff, and switching language between a failed attempt and its retry opened another. It is now get-or-create server-side under a per-(workspace, member) advisory lock, keyed on (workspace, creator, agent), returned alongside the agent. Also: - The skipped-runtime welcome dismissed itself silently when provisioning the guide issue failed. The signal is not persisted and onboarding is already complete, so a blip was terminal. It now offers a retry. - The Runtimes recovery card gated on `agents.length === 0`, so creating any ordinary agent hid the only surface that can mint a Mika — the generic endpoint accepts no system_key. Gated on Mika's absence. - CompactRuntimeRow ignored `disabled`; the CLI dialog was already passing it, so the runtime could change mid-submit. It is a real <button> now, which also gets focus and Enter/Space for free. - The rail never went below 15rem while the content pane kept its gutter, leaving ~87px of form at 375px. It is hidden under md, where a compact bar carries the step name and the Back button instead. - Dropped a stray __pycache__ artifact I had committed by accident. Each new test was checked against the bug it covers: reinstating the owner gate, the title-keyed lookup, or the dropped disabled prop makes the corresponding test fail. Co-authored-by: multica-agent <github@multica.ai>
1192 lines
54 KiB
SQL
1192 lines
54 KiB
SQL
-- name: CreateChatSession :one
|
|
INSERT INTO chat_session (workspace_id, agent_id, creator_id, title, runtime_id, is_agent_intro, project_id)
|
|
VALUES ($1, $2, $3, $4, (SELECT runtime_id FROM agent WHERE id = $2), $5, sqlc.narg('project_id'))
|
|
RETURNING *;
|
|
|
|
-- name: ClearChatSessionProjectByProject :exec
|
|
-- Project references are intentionally soft (no database FK). Keep chat
|
|
-- history while removing the context selection when a project is deleted.
|
|
-- Do not touch updated_at: context cleanup is not chat activity.
|
|
UPDATE chat_session
|
|
SET project_id = NULL
|
|
WHERE project_id = $1 AND workspace_id = $2;
|
|
|
|
-- name: GetChatSession :one
|
|
SELECT * FROM chat_session
|
|
WHERE id = $1;
|
|
|
|
-- name: GetChatSessionInWorkspace :one
|
|
SELECT * FROM chat_session
|
|
WHERE id = $1 AND workspace_id = $2;
|
|
|
|
-- name: ListChatSessionsByCreator :many
|
|
-- IM-style list: each active session with its unread *count* (assistant
|
|
-- messages after the read cursor), a preview of the latest message, and
|
|
-- ordered by most-recent activity so a new reply bumps a session to the top.
|
|
SELECT cs.*,
|
|
(SELECT count(*) FROM chat_message m
|
|
WHERE m.chat_session_id = cs.id
|
|
AND m.role = 'assistant'
|
|
AND m.created_at > cs.last_read_at)::int AS unread_count,
|
|
COALESCE(lm.content, '') AS last_message_content,
|
|
COALESCE(lm.role, '') AS last_message_role,
|
|
lm.created_at AS last_message_at,
|
|
lm.failure_reason AS last_message_failure_reason,
|
|
COALESCE(lm.message_kind, '') AS last_message_kind
|
|
FROM chat_session cs
|
|
LEFT JOIN LATERAL (
|
|
SELECT content, role, created_at, failure_reason, message_kind
|
|
FROM chat_message m
|
|
WHERE m.chat_session_id = cs.id
|
|
ORDER BY m.created_at DESC
|
|
LIMIT 1
|
|
) lm ON true
|
|
WHERE cs.workspace_id = $1 AND cs.creator_id = $2 AND cs.status = 'active'
|
|
ORDER BY (cs.pinned_at IS NOT NULL) DESC, cs.pinned_at DESC, COALESCE(lm.created_at, cs.updated_at) DESC;
|
|
|
|
-- name: ListAllChatSessionsByCreator :many
|
|
-- Unlike ListChatSessionsByCreator this returns archived sessions too (for the
|
|
-- "Archived" view), so unread must be forced to 0 for archived rows: archiving
|
|
-- deliberately does NOT advance last_read_at (so unarchive can restore the true
|
|
-- unread state), but an archived session is read-only and hidden from history,
|
|
-- so any residual unread is uncleanable and must not light up any badge. Gating
|
|
-- on status here is the single source of truth for all unread surfaces (FAB,
|
|
-- sidebar Chat tab, chat-window header) — see MUL-4360.
|
|
SELECT cs.*,
|
|
CASE WHEN cs.status = 'archived' THEN 0
|
|
ELSE (SELECT count(*) FROM chat_message m
|
|
WHERE m.chat_session_id = cs.id
|
|
AND m.role = 'assistant'
|
|
AND m.created_at > cs.last_read_at)
|
|
END::int AS unread_count,
|
|
COALESCE(lm.content, '') AS last_message_content,
|
|
COALESCE(lm.role, '') AS last_message_role,
|
|
lm.created_at AS last_message_at,
|
|
lm.failure_reason AS last_message_failure_reason,
|
|
COALESCE(lm.message_kind, '') AS last_message_kind
|
|
FROM chat_session cs
|
|
LEFT JOIN LATERAL (
|
|
SELECT content, role, created_at, failure_reason, message_kind
|
|
FROM chat_message m
|
|
WHERE m.chat_session_id = cs.id
|
|
ORDER BY m.created_at DESC
|
|
LIMIT 1
|
|
) lm ON true
|
|
WHERE cs.workspace_id = $1 AND cs.creator_id = $2
|
|
ORDER BY (cs.pinned_at IS NOT NULL) DESC, cs.pinned_at DESC, COALESCE(lm.created_at, cs.updated_at) DESC;
|
|
|
|
-- name: ListAgentBuilderSessionsByCreator :many
|
|
-- The caller's unfinished agent-creation conversations.
|
|
--
|
|
-- These never appear in ListChatSessionsByCreator: that list is filtered
|
|
-- against ListAllAgents, which is `kind = 'user'` only, so a builder session —
|
|
-- whose agent is the hidden `kind = 'system'` carrier — is invisible to every
|
|
-- chat surface by construction. This statement is the only way back to one,
|
|
-- which is why the studio may stop deleting them on navigation.
|
|
--
|
|
-- `a.runtime_id` is the whole point of the join. The carrier is what
|
|
-- SendDirectChatMessage reads to stamp a chat task's runtime, so it is the only
|
|
-- truthful answer to "where does this conversation run". Deliberately NOT
|
|
-- cs.runtime_id: that is the daemon's resume pointer, left stale on purpose
|
|
-- after a runtime switch (see RebindAgentBuilderRuntime), so resuming from it
|
|
-- would put the picker on a runtime that no longer executes anything — the
|
|
-- exact split MUL-5163 removed.
|
|
--
|
|
-- A conversation qualifies once it holds something the user would miss: a
|
|
-- message, or a saved configuration. Requiring a message alone was wrong — the
|
|
-- form on the right is editable from the moment the session exists and
|
|
-- autosaves, so someone can open the builder, type a name, and leave before the
|
|
-- first turn. That session has real work in it and was unreachable. Requiring
|
|
-- neither is also wrong: a session opened and abandoned untouched is not a
|
|
-- draft, and would put an empty row in front of the user on every accidental
|
|
-- entry into the flow.
|
|
--
|
|
-- The stored draft rides along instead of needing its own fetch: the studio
|
|
-- renders this list beside the conversation it is switching between, so the
|
|
-- configuration for the row the user picks has to be in hand at click time.
|
|
-- LEFT JOIN because a conversation that has only ever been driven by the AI has
|
|
-- no saved draft — the client replays the last <agent_draft> block in that case.
|
|
SELECT cs.id,
|
|
cs.title,
|
|
cs.created_at,
|
|
cs.updated_at,
|
|
a.runtime_id,
|
|
COALESCE(lm.content, '') AS last_message_content,
|
|
COALESCE(lm.role, '') AS last_message_role,
|
|
lm.created_at AS last_message_at,
|
|
d.draft AS stored_draft
|
|
FROM chat_session cs
|
|
JOIN agent a ON a.id = cs.agent_id
|
|
LEFT JOIN agent_builder_draft d ON d.chat_session_id = cs.id
|
|
LEFT JOIN LATERAL (
|
|
SELECT content, role, created_at
|
|
FROM chat_message m
|
|
WHERE m.chat_session_id = cs.id
|
|
ORDER BY m.created_at DESC
|
|
LIMIT 1
|
|
) lm ON true
|
|
WHERE cs.workspace_id = $1
|
|
AND cs.creator_id = $2
|
|
AND cs.status = 'active'
|
|
AND a.kind = 'system'
|
|
AND a.system_key LIKE 'agent_builder:%'
|
|
AND (lm.created_at IS NOT NULL OR d.chat_session_id IS NOT NULL)
|
|
-- A draft-only session has no message to sort by; fall back to when its
|
|
-- configuration was last written so it still lands in activity order.
|
|
ORDER BY COALESCE(lm.created_at, d.updated_at, cs.updated_at) DESC;
|
|
|
|
-- name: UpdateChatSessionTitle :one
|
|
UPDATE chat_session SET title = $2, updated_at = now()
|
|
WHERE id = $1
|
|
RETURNING *;
|
|
|
|
-- name: UpdateChatSessionProject :one
|
|
-- Project context is user-editable session metadata. Do not touch updated_at:
|
|
-- changing context is not conversation activity and must not reorder history.
|
|
UPDATE chat_session
|
|
SET project_id = sqlc.narg('project_id')
|
|
WHERE id = sqlc.arg('id') AND workspace_id = sqlc.arg('workspace_id')
|
|
RETURNING *;
|
|
|
|
-- name: UpdateChatSessionTitleIfCurrent :one
|
|
-- Compare-and-swap the title: only overwrite it when it still equals the
|
|
-- value the caller observed (@expected_title). This is the idempotency /
|
|
-- no-clobber guard behind LLM auto-titling (MUL-4295): the async generator
|
|
-- captures the session's current (default/original) title before calling the
|
|
-- model, and this write lands only if a manual rename or a competing writer
|
|
-- has not changed the title in the meantime. A mismatch returns pgx.ErrNoRows
|
|
-- (zero rows updated), which the caller treats as "someone renamed it — leave
|
|
-- it alone", NOT as an error.
|
|
UPDATE chat_session SET title = @new_title, updated_at = now()
|
|
WHERE id = @id AND title = @expected_title
|
|
RETURNING *;
|
|
|
|
-- name: SetChatSessionPinned :one
|
|
-- Pin/unpin a chat. Deliberately does NOT touch updated_at: pinning is a
|
|
-- list-ordering preference, not activity, so it must not bump the session's
|
|
-- last-activity sort key (which would make an unpinned chat jump the list).
|
|
-- pinned = true stamps pinned_at only when it was NULL, so re-pinning keeps
|
|
-- the original pin order; pinned = false clears it.
|
|
UPDATE chat_session
|
|
SET pinned_at = CASE WHEN @pinned::bool THEN COALESCE(pinned_at, now()) ELSE NULL END
|
|
WHERE id = $1
|
|
RETURNING *;
|
|
|
|
-- name: SetChatSessionArchived :one
|
|
-- Archive/unarchive a chat session by flipping status between 'active' and
|
|
-- 'archived'. Bumps updated_at so the row re-sorts on the receiving list. The
|
|
-- send-message path refuses archived sessions (see SendChatMessage), so the
|
|
-- conversation is effectively read-only until it is unarchived.
|
|
UPDATE chat_session
|
|
SET status = CASE WHEN @archived::bool THEN 'archived' ELSE 'active' END,
|
|
updated_at = now()
|
|
WHERE id = $1
|
|
RETURNING *;
|
|
|
|
-- name: UpdateChatSessionSession :exec
|
|
-- Updates the resume pointer for a chat session. Empty/NULL inputs are
|
|
-- ignored via COALESCE so a task that completes without a session_id (e.g.
|
|
-- the agent crashed before establishing one) cannot wipe out a previously
|
|
-- recorded resume pointer. This makes the chat memory robust against
|
|
-- intermittent agent failures.
|
|
UPDATE chat_session
|
|
SET session_id = COALESCE(sqlc.narg('session_id'), session_id),
|
|
work_dir = COALESCE(sqlc.narg('work_dir'), work_dir),
|
|
runtime_id = COALESCE(sqlc.narg('runtime_id'), runtime_id),
|
|
updated_at = now()
|
|
WHERE id = sqlc.arg('id');
|
|
|
|
-- name: ClearChatSessionSessionIfMatches :exec
|
|
-- Drops the chat session's resume pointer, but only while it still points at
|
|
-- the exact session the caller proved unresumable.
|
|
--
|
|
-- The claim handler reads chat_session.session_id FIRST and only falls back to
|
|
-- GetLastChatTaskSession when it is empty, so a poisoned pointer here bypasses
|
|
-- every filter that query applies. Declining to OVERWRITE the pointer on a
|
|
-- resume-unsafe failure — which is all the fail path used to do — leaves the
|
|
-- dead session in place and the next turn resumes it (GH #6066).
|
|
--
|
|
-- The session_id + runtime_id predicate is what makes this safe to run in the
|
|
-- fail transaction: a concurrent turn that has already written a NEW pointer
|
|
-- does not match, so its healthy session survives instead of being cleared by
|
|
-- a slower sibling's failure. work_dir is deliberately left alone — the
|
|
-- directory is still reusable, only the conversation is not.
|
|
UPDATE chat_session
|
|
SET session_id = NULL,
|
|
runtime_id = NULL,
|
|
updated_at = now()
|
|
WHERE id = sqlc.arg('id')
|
|
AND session_id = sqlc.arg('session_id')
|
|
AND runtime_id = sqlc.arg('runtime_id');
|
|
|
|
-- name: AdvanceCancelledChatSessionPointer :exec
|
|
-- Moves a chat's resume pointer onto the session a CANCELLED task recorded
|
|
-- (GH #6340).
|
|
--
|
|
-- Cancellation is the one terminal state that never reports back: the daemon
|
|
-- discards its result and only sends a cancel-ack, so neither CompleteTask nor
|
|
-- FailTask — the only other writers of chat_session.session_id — ever runs. The
|
|
-- claim handler reads this pointer BEFORE falling back to
|
|
-- GetLastChatTaskSession, so on a chat that already has history a pointer left
|
|
-- on the previous turn shadows the cancelled turn's session no matter what the
|
|
-- fallback would have found.
|
|
--
|
|
-- Two callers, one statement, because both are races the other cannot cover:
|
|
-- the cancel path runs it inside the status-flip transaction (so no follow-up
|
|
-- can observe `cancelled` while the pointer still names the older session), and
|
|
-- the pin path runs it after a mid-flight pin lands on an already-cancelled row
|
|
-- (Codex waits for its rollout, so the pin routinely arrives after the cancel —
|
|
-- at which point the cancel path saw no session to publish).
|
|
--
|
|
-- Everything it decides on is read from the task row inside the statement, so
|
|
-- neither caller can act on a stale in-memory copy. The NOT EXISTS guard is what
|
|
-- makes the late pin safe: a NEWER task on this chat that already recorded a
|
|
-- session owns the pointer, and a straggler must not drag the conversation
|
|
-- backwards onto the turn the user interrupted.
|
|
UPDATE chat_session cs
|
|
SET session_id = t.session_id,
|
|
runtime_id = t.runtime_id,
|
|
work_dir = COALESCE(t.work_dir, cs.work_dir),
|
|
updated_at = now()
|
|
FROM agent_task_queue t
|
|
WHERE t.id = sqlc.arg('task_id')
|
|
AND t.chat_session_id = cs.id
|
|
AND t.status = 'cancelled'
|
|
AND t.session_id IS NOT NULL
|
|
AND t.runtime_id IS NOT NULL
|
|
AND NOT EXISTS (
|
|
SELECT 1 FROM agent_task_queue newer
|
|
WHERE newer.chat_session_id = t.chat_session_id
|
|
AND newer.id <> t.id
|
|
AND newer.session_id IS NOT NULL
|
|
AND newer.created_at > t.created_at
|
|
);
|
|
|
|
-- name: LockChatSessionForDelete :one
|
|
-- Acquires an exclusive (FOR UPDATE) row lock on chat_session(id). Used by
|
|
-- the delete path so that a concurrent SendChatMessage cannot enqueue a new
|
|
-- agent_task_queue row referencing this session between our cancel and
|
|
-- delete steps. The FK from agent_task_queue.chat_session_id takes a
|
|
-- KEY SHARE lock on the parent row during INSERT validation, which
|
|
-- conflicts with FOR UPDATE — concurrent inserts block here and then fail
|
|
-- their FK check after we commit the delete.
|
|
SELECT id FROM chat_session
|
|
WHERE id = $1
|
|
FOR UPDATE;
|
|
|
|
-- name: LockChatSessionForRuntimeBind :one
|
|
-- Acquires an exclusive (FOR UPDATE) row lock on chat_session(id), serialising
|
|
-- "which runtime does this session execute on" against "enqueue the next task".
|
|
--
|
|
-- Both SendDirectChatMessage and the agent-builder runtime switch take this lock
|
|
-- for their whole transaction. Without it the two are a read-then-write race: a
|
|
-- send reads the carrier agent's runtime_id, the switch then passes its
|
|
-- pending-task check and rebinds the carrier, and the send finally inserts a task
|
|
-- still stamped with the pre-switch runtime — so the user is told the switch
|
|
-- succeeded while their message runs on the old runtime (MUL-5163).
|
|
--
|
|
-- The lock alone is not sufficient: the send path must also re-read the agent
|
|
-- INSIDE the locked transaction, because a send blocked at INSERT would otherwise
|
|
-- resume and write the runtime_id it read before blocking.
|
|
--
|
|
-- Same row and same lock mode as LockChatSessionForDelete, and both take it as
|
|
-- their first statement, so the delete path and this one cannot deadlock.
|
|
SELECT id FROM chat_session
|
|
WHERE id = $1
|
|
FOR UPDATE;
|
|
|
|
-- name: LockChatSessionForDraftWrite :one
|
|
-- The autosave half of the agent_builder_draft protocol, and the writer's
|
|
-- answer to LockChatSessionForDelete.
|
|
--
|
|
-- agent_builder_draft carries no chat_session FK (repo rule), so an INSERT into
|
|
-- it takes no lock on the parent row and nothing stops a draft from landing
|
|
-- after its conversation is gone: the save path read the session, the delete
|
|
-- transaction then committed, and the upsert still succeeded — leaving a row
|
|
-- the user explicitly discarded, invisible to the UI and unreachable by every
|
|
-- prune except the workspace teardown. The FK that would have rejected that
|
|
-- INSERT is the one we are not allowed to have, so the lock replaces it.
|
|
--
|
|
-- Returns the whole row, not just the id: the caller must re-check creator and
|
|
-- status INSIDE the transaction, because a save blocked here resumes holding
|
|
-- values it read before blocking (the same reason the runtime-bind path re-reads
|
|
-- the agent under its lock).
|
|
--
|
|
-- Same row and same lock mode as LockChatSessionForDelete and
|
|
-- LockChatSessionForRuntimeBind, taken as the transaction's first statement, so
|
|
-- no ordering against the repo-wide chat_session -> agent_task_queue sequence is
|
|
-- introduced and none of the three can deadlock against each other.
|
|
SELECT * FROM chat_session
|
|
WHERE id = $1
|
|
FOR UPDATE;
|
|
|
|
-- name: DeleteChatSession :exec
|
|
-- Hard delete. chat_message rows cascade via FK ON DELETE CASCADE; the
|
|
-- chat_session_id on agent_task_queue is set NULL by FK so completed/failed
|
|
-- task history survives the session being removed. Callers MUST run inside
|
|
-- the same transaction that holds LockChatSessionForDelete and that has
|
|
-- already cancelled any in-flight tasks (see CancelAgentTasksByChatSession)
|
|
-- so the daemon does not keep running work whose result has nowhere to
|
|
-- land. workspace_id in the WHERE clause is a SQL-layer tenant guard; see
|
|
-- DeleteIssue.
|
|
DELETE FROM chat_session WHERE id = $1 AND workspace_id = $2;
|
|
|
|
-- name: TouchChatSession :exec
|
|
UPDATE chat_session SET updated_at = now()
|
|
WHERE id = $1;
|
|
|
|
-- name: CreateChatMessage :one
|
|
-- message_kind and quick_actions default via COALESCE so every existing caller
|
|
-- (which omits it) keeps writing ordinary messages; the empty-reply path passes
|
|
-- 'no_response' to mark a visible turn with no text output (MUL-4351).
|
|
INSERT INTO chat_message (
|
|
chat_session_id, role, content, task_id, failure_reason, elapsed_ms,
|
|
message_kind, quick_actions, channel_media_pending_until, channel_ingested
|
|
)
|
|
VALUES (
|
|
$1, $2, $3, sqlc.narg(task_id), sqlc.narg(failure_reason), sqlc.narg(elapsed_ms),
|
|
COALESCE(sqlc.narg(message_kind)::text, 'message'),
|
|
COALESCE(sqlc.narg(quick_actions)::jsonb, '[]'::jsonb),
|
|
-- The media deadline is DB-clock time: every consumer compares it against
|
|
-- SQL now() (GetChannelMediaPendingUntil, the deferred promote, the
|
|
-- trailing-message guard), so the writer must use the same clock. The
|
|
-- caller passes a relative budget in seconds; an application-clock
|
|
-- timestamp here would let a skewed app node shrink or stretch the
|
|
-- fallback window.
|
|
CASE WHEN sqlc.narg(channel_media_pending_secs)::float8 IS NULL THEN NULL
|
|
ELSE now() + make_interval(secs => sqlc.narg(channel_media_pending_secs)::float8) END,
|
|
COALESCE(sqlc.narg(channel_ingested)::boolean, FALSE)
|
|
)
|
|
RETURNING *;
|
|
|
|
-- name: TaskHasChannelIngestedMessages :one
|
|
-- Immutable channel provenance for a task's user-message input batch:
|
|
-- channel_ingested is stamped inside the channel append transaction and never
|
|
-- mutated afterwards, so it survives session archiving and installation
|
|
-- rebinds that delete the channel_chat_session_binding row. Callers pass the
|
|
-- batch OWNER id (chat_input_task_id, which auto-retry clones inherit), not
|
|
-- necessarily the task's own id. The cancel restore-delete and the
|
|
-- empty-completion silent-drop both gate on this — a channel sender has no
|
|
-- Multica composer for a restored draft, and the no_response fallback body
|
|
-- must never be pushed to an external channel.
|
|
SELECT EXISTS (
|
|
SELECT 1 FROM chat_message
|
|
WHERE task_id = $1
|
|
AND role = 'user'
|
|
AND channel_ingested
|
|
) AS channel_ingested;
|
|
|
|
-- name: GetChannelMediaPendingUntil :one
|
|
-- The latest unexpired media deadline gates a channel task. Using a durable
|
|
-- task fire_at means a process restart still produces the placeholder fallback.
|
|
-- Only a turn that can join a task's input batch may gate one: channel_command
|
|
-- turns are excluded from the seal below, so an unrelated later message would
|
|
-- otherwise wait out a command's media (or its whole fallback budget, since a
|
|
-- command whose create failed never runs the binder that clears the marker).
|
|
SELECT channel_media_pending_until
|
|
FROM chat_message
|
|
WHERE chat_session_id = $1
|
|
AND role = 'user'
|
|
AND message_kind != 'channel_command'
|
|
AND channel_media_pending_until > now()
|
|
ORDER BY channel_media_pending_until DESC
|
|
LIMIT 1;
|
|
|
|
-- name: ClearChatMessageChannelMediaPending :exec
|
|
UPDATE chat_message
|
|
SET channel_media_pending_until = NULL
|
|
WHERE id = $1 AND chat_session_id = $2;
|
|
|
|
-- name: LinkChatMessageToTask :exec
|
|
UPDATE chat_message
|
|
SET task_id = $2
|
|
WHERE id = $1 AND role = 'user';
|
|
|
|
-- name: LinkUnownedChannelChatMessagesToTask :exec
|
|
-- Seals the trailing channel-message batch to its task. The task row and these
|
|
-- links are committed together, so an older in-flight task cannot absorb a
|
|
-- newer media message and a later assistant row cannot hide that message.
|
|
-- channel_command turns were already handled synchronously by Router; keeping
|
|
-- them visible but unowned prevents both immediate and delayed re-execution.
|
|
UPDATE chat_message AS message
|
|
SET task_id = @task_id
|
|
WHERE message.chat_session_id = @chat_session_id
|
|
AND message.role = 'user'
|
|
AND message.task_id IS NULL
|
|
AND message.message_kind != 'channel_command'
|
|
AND NOT EXISTS (
|
|
SELECT 1
|
|
FROM chat_message AS prior
|
|
WHERE prior.chat_session_id = @chat_session_id
|
|
AND prior.role != 'user'
|
|
AND (prior.created_at, prior.id) > (message.created_at, message.id)
|
|
);
|
|
|
|
-- name: DeferChatTaskForSealedPendingMedia :one
|
|
-- Closes the enqueue-vs-append race: under READ COMMITTED a media message can
|
|
-- commit between GetChannelMediaPendingUntil and the batch seal above, landing
|
|
-- an unexpired media marker inside a task the deadline read decided was
|
|
-- 'queued'. Re-derive the deferral from the sealed batch itself, in the same
|
|
-- transaction, so a task is never claimable while its own input still has an
|
|
-- unexpired marker. No row (ErrNoRows) means no correction was needed.
|
|
UPDATE agent_task_queue AS task
|
|
SET status = 'deferred', fire_at = pending.max_until
|
|
FROM (
|
|
SELECT max(message.channel_media_pending_until) AS max_until
|
|
FROM chat_message AS message
|
|
WHERE message.task_id = @task_id
|
|
AND message.role = 'user'
|
|
AND message.channel_media_pending_until > now()
|
|
) AS pending
|
|
WHERE task.id = @task_id
|
|
AND pending.max_until IS NOT NULL
|
|
AND (task.fire_at IS NULL OR task.fire_at < pending.max_until)
|
|
RETURNING task.*;
|
|
|
|
-- name: DeleteUserChatMessageByTask :one
|
|
DELETE FROM chat_message
|
|
WHERE task_id = $1 AND role = 'user'
|
|
RETURNING *;
|
|
|
|
-- name: ListChatMessages :many
|
|
-- IMPORTANT: the visible-head selector below is also used by
|
|
-- ListChatMessagesForLegacyTask, ListChatMessagesPage,
|
|
-- ReanchorClaimedDirectChatInput, ReanchorNextQueuedDirectChatInput,
|
|
-- ListPendingChatTasksForSession, and CancelQueuedAgentTasksForSession in
|
|
-- agent.sql. Keep the eligible statuses and ordering identical: a claimed task
|
|
-- is current, a deferred retry precedes still-queued work, and queued peers use
|
|
-- claim priority/FIFO order. Background quick-action regeneration is invisible.
|
|
-- This is a presentation/visibility order, not a scheduling guarantee:
|
|
-- deferred rows are not claimable before promotion, so a queued row may be
|
|
-- claimed during the backoff and then becomes the visible claimed head.
|
|
SELECT message.* FROM chat_message AS message
|
|
WHERE message.chat_session_id = $1
|
|
AND NOT (
|
|
message.role = 'user'
|
|
AND EXISTS (
|
|
SELECT 1
|
|
FROM agent_task_queue AS task
|
|
WHERE task.chat_session_id = message.chat_session_id
|
|
AND task.status = 'queued'
|
|
AND task.id = message.task_id
|
|
-- "Queued follow-up" is positional, not the row's transient status:
|
|
-- the first pending task is the current turn even before claim.
|
|
AND task.id <> (
|
|
SELECT head.id
|
|
FROM agent_task_queue AS head
|
|
WHERE head.chat_session_id = $1
|
|
AND head.status IN ('queued', 'dispatched', 'running', 'waiting_local_directory', 'deferred')
|
|
AND head.regenerate_quick_actions_for IS NULL
|
|
ORDER BY
|
|
CASE
|
|
WHEN head.status IN ('dispatched', 'running', 'waiting_local_directory') THEN 0
|
|
WHEN head.status = 'deferred' THEN 1
|
|
ELSE 2
|
|
END,
|
|
head.priority DESC,
|
|
head.created_at ASC,
|
|
head.id ASC
|
|
LIMIT 1
|
|
)
|
|
)
|
|
)
|
|
ORDER BY message.created_at ASC, message.id ASC;
|
|
|
|
-- name: ListChatMessagesForLegacyTask :many
|
|
-- Legacy/reclaimed daemon tasks use trailing history, but must not absorb a
|
|
-- newer queued successor that is already bound to its own user message.
|
|
SELECT message.* FROM chat_message AS message
|
|
WHERE message.chat_session_id = $1
|
|
AND NOT (
|
|
message.role = 'user'
|
|
AND EXISTS (
|
|
SELECT 1
|
|
FROM agent_task_queue AS task
|
|
WHERE task.chat_session_id = message.chat_session_id
|
|
AND task.status = 'queued'
|
|
AND task.id = message.task_id
|
|
AND task.id <> (
|
|
SELECT head.id
|
|
FROM agent_task_queue AS head
|
|
WHERE head.chat_session_id = $1
|
|
AND head.status IN ('queued', 'dispatched', 'running', 'waiting_local_directory', 'deferred')
|
|
AND head.regenerate_quick_actions_for IS NULL
|
|
ORDER BY
|
|
CASE
|
|
WHEN head.status IN ('dispatched', 'running', 'waiting_local_directory') THEN 0
|
|
WHEN head.status = 'deferred' THEN 1
|
|
ELSE 2
|
|
END,
|
|
head.priority DESC,
|
|
head.created_at ASC,
|
|
head.id ASC
|
|
LIMIT 1
|
|
)
|
|
)
|
|
)
|
|
ORDER BY message.created_at ASC, message.id ASC;
|
|
|
|
-- name: ListChatInputMessages :many
|
|
-- Loads the immutable user-message input batch owned by a direct-chat task.
|
|
-- The caller passes the task's chat_input_task_id (itself for an original send,
|
|
-- the root task for an auto-retry child), so a claim reads exactly the messages
|
|
-- the user sent for this turn — and never absorbs a message that arrived after
|
|
-- the batch was sealed, no matter what the assistant wrote or when. Only used
|
|
-- for new task-owned direct-chat tasks; legacy/channel (chat_input_task_id
|
|
-- NULL) tasks keep using ListChatMessagesForLegacyTask + trailingUserMessages.
|
|
SELECT * FROM chat_message
|
|
WHERE task_id = $1 AND role = 'user'
|
|
ORDER BY created_at ASC, id ASC;
|
|
|
|
-- name: ReanchorClaimedDirectChatInput :exec
|
|
-- An idle direct send is visible while it is the positional queue head. A
|
|
-- older/pre-deploy follow-up whose enqueue timestamp puts it before a newer
|
|
-- settled reply can still reach claim; this is the fallback that moves only
|
|
-- that out-of-order user row to the new turn boundary. In-order visible heads
|
|
-- keep their original timestamp, so an idle send already returned to a cursor
|
|
-- client never moves merely because a daemon claimed it.
|
|
--
|
|
-- The task's own created_at remains immutable and continues to drive queue
|
|
-- FIFO, wait duration, and elapsed time. Only the message timestamp changes,
|
|
-- preserving the existing public ordering/cursor contract for older clients.
|
|
-- Add one microsecond when the DB clock ties the latest visible row so UUID
|
|
-- ordering can never put the new user turn before the previous assistant row.
|
|
--
|
|
-- channel_ingested excludes sealed Slack/Lark batches, which can own multiple
|
|
-- user rows and must preserve their original provider order. A direct send
|
|
-- owns exactly one non-channel user row today; if direct batching is added,
|
|
-- this query must assign a stable per-row offset instead of one timestamp.
|
|
-- Retry children are excluded by the caller because their chat_input_task_id
|
|
-- names the root task rather than themselves. Stale dispatched reclaim bypasses
|
|
-- this query, so re-delivery does not move an already-visible turn.
|
|
--
|
|
-- Lock order is task row first (ClaimAgentTask), then this non-FK message-only
|
|
-- update. The visibility subqueries take no row locks, avoiding an inverse
|
|
-- message -> task lock edge with send/completion transactions.
|
|
WITH latest_visible AS (
|
|
SELECT
|
|
claimed_input.id AS claimed_input_id,
|
|
claimed_input.created_at AS input_created_at,
|
|
prior.created_at AS prior_created_at
|
|
FROM chat_message AS claimed_input
|
|
CROSS JOIN LATERAL (
|
|
SELECT prior.created_at
|
|
FROM chat_message AS prior
|
|
WHERE prior.chat_session_id = claimed_input.chat_session_id
|
|
AND prior.id != claimed_input.id
|
|
AND NOT (
|
|
prior.role = 'user'
|
|
AND EXISTS (
|
|
SELECT 1
|
|
FROM agent_task_queue AS queued_task
|
|
WHERE queued_task.chat_session_id = prior.chat_session_id
|
|
AND queued_task.status = 'queued'
|
|
AND queued_task.id = prior.task_id
|
|
AND queued_task.id <> (
|
|
SELECT head.id
|
|
FROM agent_task_queue AS head
|
|
WHERE head.chat_session_id = claimed_input.chat_session_id
|
|
AND head.status IN ('queued', 'dispatched', 'running', 'waiting_local_directory', 'deferred')
|
|
AND head.regenerate_quick_actions_for IS NULL
|
|
ORDER BY
|
|
CASE
|
|
WHEN head.status IN ('dispatched', 'running', 'waiting_local_directory') THEN 0
|
|
WHEN head.status = 'deferred' THEN 1
|
|
ELSE 2
|
|
END,
|
|
head.priority DESC,
|
|
head.created_at ASC,
|
|
head.id ASC
|
|
LIMIT 1
|
|
)
|
|
)
|
|
)
|
|
ORDER BY prior.created_at DESC, prior.id DESC
|
|
LIMIT 1
|
|
) AS prior
|
|
WHERE claimed_input.task_id = @task_id
|
|
AND claimed_input.role = 'user'
|
|
AND NOT claimed_input.channel_ingested
|
|
)
|
|
UPDATE chat_message AS claimed_input
|
|
SET created_at = GREATEST(
|
|
@dispatched_at::timestamptz,
|
|
latest_visible.prior_created_at + interval '1 microsecond'
|
|
)
|
|
FROM latest_visible
|
|
WHERE claimed_input.id = latest_visible.claimed_input_id
|
|
AND latest_visible.prior_created_at >= latest_visible.input_created_at;
|
|
|
|
-- name: ReanchorNextQueuedDirectChatInput :exec
|
|
-- An assistant outcome can make the next queued direct task the positional
|
|
-- head before a daemon claims it (MUL-5750). Its user row was persisted before
|
|
-- this reply, so move that still-hidden single-row input just after the reply.
|
|
-- Callers run this immediately after CreateChatMessage in the same transaction:
|
|
-- readers see either the old active head without the reply, or the settled
|
|
-- reply followed by the newly-visible head — never user B before assistant A.
|
|
-- Channel batches are excluded because their immutable provider ordering may
|
|
-- contain multiple user rows. Direct chat currently owns exactly one row; if
|
|
-- direct batching is added, assign stable per-row offsets here.
|
|
UPDATE chat_message AS queued_input
|
|
SET created_at = sqlc.arg('assistant_created_at')::timestamptz + interval '1 microsecond'
|
|
WHERE queued_input.chat_session_id = $1
|
|
AND queued_input.role = 'user'
|
|
AND NOT queued_input.channel_ingested
|
|
AND queued_input.created_at <= sqlc.arg('assistant_created_at')::timestamptz
|
|
AND EXISTS (
|
|
SELECT 1
|
|
FROM agent_task_queue AS queued_task
|
|
WHERE queued_task.id = queued_input.task_id
|
|
AND queued_task.chat_session_id = queued_input.chat_session_id
|
|
AND queued_task.status = 'queued'
|
|
AND queued_task.chat_input_task_id = queued_task.id
|
|
AND queued_task.regenerate_quick_actions_for IS NULL
|
|
AND queued_task.id = (
|
|
SELECT head.id
|
|
FROM agent_task_queue AS head
|
|
WHERE head.chat_session_id = queued_input.chat_session_id
|
|
AND head.status IN ('queued', 'dispatched', 'running', 'waiting_local_directory', 'deferred')
|
|
AND head.regenerate_quick_actions_for IS NULL
|
|
ORDER BY
|
|
CASE
|
|
WHEN head.status IN ('dispatched', 'running', 'waiting_local_directory') THEN 0
|
|
WHEN head.status = 'deferred' THEN 1
|
|
ELSE 2
|
|
END,
|
|
head.priority DESC,
|
|
head.created_at ASC,
|
|
head.id ASC
|
|
LIMIT 1
|
|
)
|
|
);
|
|
|
|
-- name: ListChatMessagesPage :many
|
|
SELECT message.* FROM chat_message AS message
|
|
WHERE message.chat_session_id = $1
|
|
AND NOT (
|
|
message.role = 'user'
|
|
AND EXISTS (
|
|
SELECT 1
|
|
FROM agent_task_queue AS task
|
|
WHERE task.chat_session_id = message.chat_session_id
|
|
AND task.status = 'queued'
|
|
AND task.id = message.task_id
|
|
AND task.id <> (
|
|
SELECT head.id
|
|
FROM agent_task_queue AS head
|
|
WHERE head.chat_session_id = $1
|
|
AND head.status IN ('queued', 'dispatched', 'running', 'waiting_local_directory', 'deferred')
|
|
AND head.regenerate_quick_actions_for IS NULL
|
|
ORDER BY
|
|
CASE
|
|
WHEN head.status IN ('dispatched', 'running', 'waiting_local_directory') THEN 0
|
|
WHEN head.status = 'deferred' THEN 1
|
|
ELSE 2
|
|
END,
|
|
head.priority DESC,
|
|
head.created_at ASC,
|
|
head.id ASC
|
|
LIMIT 1
|
|
)
|
|
)
|
|
)
|
|
AND (
|
|
sqlc.narg('before_created_at')::timestamptz IS NULL
|
|
OR (message.created_at, message.id) < (sqlc.narg('before_created_at')::timestamptz, sqlc.narg('before_id')::uuid)
|
|
)
|
|
ORDER BY message.created_at DESC, message.id DESC
|
|
LIMIT $2;
|
|
|
|
-- name: GetChatMessage :one
|
|
SELECT * FROM chat_message
|
|
WHERE id = $1;
|
|
|
|
-- name: CreateChatTask :one
|
|
-- The chat sender (initiator) is a direct_human originator and accountable;
|
|
-- attribution provenance is stamped so this path is not a NULL-source enqueue
|
|
-- bypass (MUL-4302 §2).
|
|
INSERT INTO agent_task_queue (
|
|
agent_id, runtime_id, issue_id, status, priority, chat_session_id,
|
|
initiator_user_id, originator_user_id, accountable_user_id, force_fresh_session, runtime_mcp_overlay,
|
|
runtime_connected_apps, originator_source, trigger_evidence_kind, trigger_evidence_ref_id,
|
|
fire_at
|
|
)
|
|
VALUES (
|
|
$1, $2, NULL,
|
|
CASE WHEN sqlc.narg('fire_at')::timestamptz IS NULL THEN 'queued' ELSE 'deferred' END,
|
|
$3, $4, $5,
|
|
sqlc.narg(originator_user_id),
|
|
sqlc.narg(accountable_user_id),
|
|
COALESCE(sqlc.narg('force_fresh_session')::boolean, FALSE),
|
|
sqlc.narg(runtime_mcp_overlay),
|
|
sqlc.narg(runtime_connected_apps),
|
|
sqlc.narg(originator_source),
|
|
sqlc.narg(trigger_evidence_kind),
|
|
sqlc.narg(trigger_evidence_ref_id),
|
|
sqlc.narg('fire_at')::timestamptz
|
|
)
|
|
RETURNING *;
|
|
|
|
-- name: PromoteChannelChatTasksIfMediaReady :many
|
|
-- Media completion may race with the 3s run batcher. Promote every original
|
|
-- channel task waiting for this session only after all unexpired media markers
|
|
-- are gone; retry/escalation/direct-chat deferred tasks are excluded. The
|
|
-- marker scan uses the same population as GetChannelMediaPendingUntil and the
|
|
-- seal: a channel_command turn belongs to no batch, so it must not hold a
|
|
-- deferred task back either.
|
|
UPDATE agent_task_queue AS task
|
|
SET status = 'queued', fire_at = NULL
|
|
WHERE task.chat_session_id = @chat_session_id
|
|
AND task.status = 'deferred'
|
|
AND task.issue_id IS NULL
|
|
AND task.parent_task_id IS NULL
|
|
AND task.escalation_for_task_id IS NULL
|
|
AND NOT EXISTS (
|
|
SELECT 1
|
|
FROM chat_message AS message
|
|
WHERE message.chat_session_id = @chat_session_id
|
|
AND message.role = 'user'
|
|
AND message.message_kind != 'channel_command'
|
|
AND message.channel_media_pending_until > now()
|
|
)
|
|
RETURNING task.*;
|
|
|
|
-- name: SetChatTaskInputOwnerSelf :one
|
|
-- Stamps a freshly-created direct-chat task as the owner of its own input batch
|
|
-- (chat_input_task_id = id), so a later claim loads exactly the user messages
|
|
-- tagged with this task id (ListChatInputMessages) rather than scanning trailing
|
|
-- history. Runs in the same transaction as CreateChatTask + message ownership:
|
|
-- direct-send inserts one owned message, while channel enqueue seals its
|
|
-- trailing unowned batch. Legacy tasks keep chat_input_task_id NULL and retain
|
|
-- the trailing-history fallback during rolling deploys.
|
|
UPDATE agent_task_queue
|
|
SET chat_input_task_id = id
|
|
WHERE id = $1
|
|
RETURNING *;
|
|
|
|
-- name: GetLastChatTaskSession :one
|
|
-- Returns the most recent task in this chat session that managed to record a
|
|
-- session_id. Includes completed, failed AND cancelled tasks: each of them may
|
|
-- have established a real agent session, and we'd rather resume there than
|
|
-- start over and lose conversation memory. Used as a fallback when
|
|
-- chat_session.session_id is NULL. Resume-unsafe failures are excluded because
|
|
-- replaying those sessions deterministically reproduces the same terminal
|
|
-- state. Keep this list in sync with resumeUnsafeFailureReason and
|
|
-- GetLastTaskSession.
|
|
--
|
|
-- 'cancelled' is resumable and its absence was GH #6340: the user stops a turn
|
|
-- the agent had already started answering, and the next message starts from
|
|
-- nothing. A cancelled row only carries a session_id because the daemon pinned
|
|
-- one mid-flight (UpdateAgentTaskSession), which means the provider really did
|
|
-- emit that session — either the resume loaded or it opened a fresh one. The
|
|
-- user interrupted it; the provider did not reject it, so it is no more
|
|
-- suspect than a completed one. Cancellation records no failure_reason/error,
|
|
-- so the poison filters below have nothing to match and cancelled rows pass
|
|
-- them the way completed rows do. The remaining risk — a transcript killed
|
|
-- mid-tool-call that the provider later refuses — is caught downstream by
|
|
-- taskfailure.UnresumableHistory and retires the session on the next turn.
|
|
--
|
|
-- The regex pair mirrors GetLastTaskSession's provider-agnostic guard for an
|
|
-- empty message baked into the conversation history: both must match, and
|
|
-- both track emptyContentRe / historyMessageLocatorRe in
|
|
-- pkg/taskfailure/resume.go (GH #6066).
|
|
--
|
|
-- Selection is per-session, not per-row, and retired sessions are excluded —
|
|
-- both mirroring GetLastTaskSession, which this query had drifted away from.
|
|
-- A plain row-level filter reopens the poisoning wormhole GH #5975 closed on
|
|
-- the issue side: it drops the newest poisoned row for a session and then
|
|
-- happily falls back to an OLDER completed row carrying the same dead
|
|
-- session_id. Judging each session by its LATEST terminal state means a newer
|
|
-- poisoned row invalidates the whole session, while a genuinely different
|
|
-- healthy session stays eligible.
|
|
WITH retired_sessions AS (
|
|
SELECT DISTINCT r.retired_session_id AS session_id
|
|
FROM agent_task_queue r
|
|
WHERE r.chat_session_id = $1
|
|
AND r.retired_session_id IS NOT NULL
|
|
), resume_overflow_at AS (
|
|
-- completed_at alone, where the issue-side twin coalesces four columns:
|
|
-- this query already selects and orders by bare completed_at throughout,
|
|
-- so the cutoff has to be measured on the same clock as the values it is
|
|
-- compared against. Change both halves together if that ever moves.
|
|
SELECT MAX(t.completed_at) AS at
|
|
FROM agent_task_queue t
|
|
WHERE t.chat_session_id = $1
|
|
AND t.status = 'failed'
|
|
AND (
|
|
COALESCE(t.failure_reason, '') = 'codex_resume_oversized'
|
|
OR (COALESCE(t.error, '') ILIKE '%thread/resume failed%' AND COALESCE(t.error, '') ILIKE '%token too long%')
|
|
)
|
|
), latest_per_session AS (
|
|
SELECT DISTINCT ON (t.session_id)
|
|
t.session_id, t.work_dir, t.runtime_id, t.status, t.failure_reason, t.error, t.completed_at
|
|
FROM agent_task_queue t
|
|
WHERE t.chat_session_id = $1
|
|
AND t.session_id IS NOT NULL
|
|
AND t.status IN ('completed', 'failed', 'cancelled')
|
|
ORDER BY t.session_id, t.completed_at DESC
|
|
)
|
|
SELECT session_id, work_dir, runtime_id FROM latest_per_session
|
|
WHERE session_id NOT IN (SELECT session_id FROM retired_sessions)
|
|
AND (
|
|
status IN ('completed', 'cancelled')
|
|
OR (
|
|
status = 'failed'
|
|
AND COALESCE(failure_reason, '') NOT IN ('iteration_limit', 'agent_fallback_message', 'api_invalid_request', 'codex_semantic_inactivity', 'agent_error.context_overflow', 'codex_resume_oversized')
|
|
AND NOT (COALESCE(error, '') ILIKE '%400%' AND COALESCE(error, '') ILIKE '%invalid_request_error%')
|
|
AND NOT (COALESCE(error, '') ~* 'must not be empty|must be non-?empty|must have non-?empty|non-?empty content|cannot be empty|should not be empty'
|
|
AND COALESCE(error, '') ~* 'role[^a-z0-9]{0,2}assistant|assistant message|message at position|messages\.[0-9]|messages\[[0-9]')
|
|
)
|
|
)
|
|
-- MUL-5722, mirroring GetLastTaskSession: an overflowed resume records no
|
|
-- session, so exclude by time instead of by matching the failed row. Note
|
|
-- this only guards the FALLBACK — the claim handler reads
|
|
-- chat_session.session_id first, so a pointer still naming the oversized
|
|
-- thread has to be cleared at fail time (see FailTask) to be covered.
|
|
AND (
|
|
(SELECT at FROM resume_overflow_at) IS NULL
|
|
OR completed_at > (SELECT at FROM resume_overflow_at)
|
|
)
|
|
ORDER BY completed_at DESC
|
|
LIMIT 1;
|
|
|
|
-- name: HasActiveChatTaskForSession :one
|
|
-- True while ANY task — a normal user turn OR a background quick-actions
|
|
-- regenerate — is in flight for the session (contrast GetPendingChatTask, which
|
|
-- hides regenerate passes from the UI). A quick-actions refresh is refused when
|
|
-- this is true: a running turn is about to change the latest reply, so the
|
|
-- target we'd resume is already stale even before its assistant row lands; and a
|
|
-- second concurrent regenerate would double-spend quota on the same turn. Read
|
|
-- inside the same session lock as the enqueue so it cannot race a sibling insert
|
|
-- (MUL-5149 review §1/§2).
|
|
--
|
|
-- 'deferred' is included: an auto-retry armed with a backoff fire_at is inserted
|
|
-- deferred (CreateRetryTask), and provider_network's final chat attempt waits
|
|
-- ~5s that way. During that window the failed turn has written no assistant row,
|
|
-- so the latest-persisted check still points at the OLD turn — omitting deferred
|
|
-- would let a refresh resume a session the retry is about to advance and attach
|
|
-- the new turn's suggestions to the old one (MUL-5149 re-review §1).
|
|
SELECT EXISTS (
|
|
SELECT 1 FROM agent_task_queue
|
|
WHERE chat_session_id = $1
|
|
AND status IN ('queued', 'dispatched', 'running', 'waiting_local_directory', 'deferred')
|
|
) AS has_active;
|
|
|
|
-- name: HasPendingChatTurnForSession :one
|
|
-- Position-only check for a direct send. Unlike GetPendingChatTask this is an
|
|
-- EXISTS query and includes deferred retries: a new user turn must remain a
|
|
-- follow-up while an older retry waits for its backoff, otherwise promotion of
|
|
-- that retry would make the new message disappear from the visible transcript.
|
|
-- Background quick-action regeneration owns no visible turn and is excluded.
|
|
SELECT EXISTS (
|
|
SELECT 1 FROM agent_task_queue
|
|
WHERE chat_session_id = $1
|
|
AND status IN ('queued', 'dispatched', 'running', 'waiting_local_directory', 'deferred')
|
|
AND regenerate_quick_actions_for IS NULL
|
|
) AS has_pending;
|
|
|
|
-- name: GetPendingChatTask :one
|
|
-- Returns the most recent in-flight task for a chat session, if any.
|
|
-- Used by the frontend to recover pending state after refresh / reopen.
|
|
-- created_at is the anchor for the chat StatusPill timer (it computes
|
|
-- elapsed = now - task.created_at), so the pill survives refresh / reopen
|
|
-- without "resetting to 0s".
|
|
SELECT id, status, created_at FROM agent_task_queue
|
|
WHERE chat_session_id = $1 AND status IN ('queued', 'dispatched', 'running', 'waiting_local_directory')
|
|
-- Background quick-actions regeneration passes are invisible to the chat UI:
|
|
-- they own no assistant turn and must not raise the StatusPill or disable the
|
|
-- composer (MUL-5149 refresh follow-up).
|
|
AND regenerate_quick_actions_for IS NULL
|
|
ORDER BY created_at DESC
|
|
LIMIT 1;
|
|
|
|
-- name: ListPendingChatTasksForSession :many
|
|
-- Returns a claimed task first, then a deferred retry, followed by prioritized
|
|
-- then FIFO queued work. See the shared visible-head invariant above
|
|
-- ListChatMessages; changing this order requires changing every selector named
|
|
-- there in the same patch.
|
|
-- The message lateral join reads only the immutable input owned by each task;
|
|
-- it avoids loading the session's complete message history just to render a
|
|
-- one-line queue preview. GetPendingChatTask remains for legacy callers that
|
|
-- only need an existence check.
|
|
SELECT
|
|
task.id,
|
|
task.status,
|
|
task.created_at,
|
|
message.id AS message_id,
|
|
COALESCE(message.content, '')::text AS content
|
|
FROM agent_task_queue AS task
|
|
LEFT JOIN LATERAL (
|
|
SELECT input.id, input.content
|
|
FROM chat_message AS input
|
|
WHERE input.task_id = COALESCE(task.chat_input_task_id, task.id)
|
|
AND input.role = 'user'
|
|
ORDER BY input.created_at ASC, input.id ASC
|
|
LIMIT 1
|
|
) AS message ON TRUE
|
|
WHERE task.chat_session_id = $1
|
|
AND task.status IN ('queued', 'dispatched', 'running', 'waiting_local_directory', 'deferred')
|
|
AND task.regenerate_quick_actions_for IS NULL
|
|
ORDER BY
|
|
CASE
|
|
WHEN task.status IN ('dispatched', 'running', 'waiting_local_directory') THEN 0
|
|
WHEN task.status = 'deferred' THEN 1
|
|
ELSE 2
|
|
END,
|
|
task.priority DESC,
|
|
task.created_at ASC,
|
|
task.id ASC;
|
|
|
|
-- name: PrioritizeQueuedChatTask :one
|
|
WITH target AS MATERIALIZED (
|
|
SELECT candidate.id
|
|
FROM agent_task_queue AS candidate
|
|
WHERE candidate.id = sqlc.arg('id')
|
|
AND candidate.chat_session_id = sqlc.arg('chat_session_id')
|
|
AND candidate.status = 'queued'
|
|
-- "Send now" is valid only while there is a visible claimed task for the
|
|
-- client to cancel. If the visible head is still queued (or deferred), the
|
|
-- selected row would otherwise replace it without any active_task_id.
|
|
AND EXISTS (
|
|
SELECT 1
|
|
FROM agent_task_queue AS active
|
|
WHERE active.chat_session_id = sqlc.arg('chat_session_id')
|
|
AND active.status IN ('dispatched', 'running', 'waiting_local_directory')
|
|
AND active.regenerate_quick_actions_for IS NULL
|
|
)
|
|
FOR UPDATE
|
|
), demoted AS (
|
|
UPDATE agent_task_queue AS queued
|
|
SET priority = 3
|
|
WHERE queued.chat_session_id = sqlc.arg('chat_session_id')
|
|
AND queued.id <> sqlc.arg('id')
|
|
AND queued.status = 'queued'
|
|
AND queued.priority >= 4
|
|
AND EXISTS (SELECT 1 FROM target)
|
|
), prioritized AS (
|
|
UPDATE agent_task_queue AS selected
|
|
SET priority = 4
|
|
FROM target
|
|
WHERE selected.id = target.id
|
|
RETURNING selected.id
|
|
)
|
|
SELECT
|
|
prioritized.id AS task_id,
|
|
(
|
|
SELECT active.id
|
|
FROM agent_task_queue AS active
|
|
WHERE active.chat_session_id = sqlc.arg('chat_session_id')
|
|
AND active.status IN ('dispatched', 'running', 'waiting_local_directory')
|
|
AND active.regenerate_quick_actions_for IS NULL
|
|
ORDER BY active.created_at ASC, active.id ASC
|
|
LIMIT 1
|
|
)::uuid AS active_task_id
|
|
FROM prioritized;
|
|
|
|
-- name: ListPendingChatTasksByCreator :many
|
|
-- Aggregate view of all in-flight chat tasks owned by a given creator in a
|
|
-- workspace. Drives the FAB's "running" indicator when the chat window is
|
|
-- closed and no single session's query is active.
|
|
--
|
|
-- Returns cs.agent_id so the handler can filter tasks belonging to private
|
|
-- agents the caller has lost access to using the already-loaded `allowed`
|
|
-- set — no second ListAllChatSessionsByCreator scan on the hot path.
|
|
--
|
|
-- atq.chat_session_id IS NOT NULL is redundant given the JOIN, but stated
|
|
-- explicitly so the planner can prove the query predicate is a subset of the
|
|
-- idx_agent_task_queue_chat_pending_v3 partial-index predicate and use it.
|
|
SELECT atq.id AS task_id, atq.status, atq.chat_session_id, cs.agent_id
|
|
FROM agent_task_queue atq
|
|
JOIN chat_session cs ON cs.id = atq.chat_session_id
|
|
WHERE atq.chat_session_id IS NOT NULL
|
|
AND atq.status IN ('queued', 'dispatched', 'running', 'waiting_local_directory', 'deferred')
|
|
-- Exclude background quick-actions regeneration passes: they own no assistant
|
|
-- turn and must not surface as "running" chat work (MUL-5149 refresh follow-up).
|
|
AND atq.regenerate_quick_actions_for IS NULL
|
|
AND cs.workspace_id = $1
|
|
AND cs.creator_id = $2
|
|
ORDER BY atq.created_at DESC;
|
|
|
|
-- name: HasPendingChatTasksByCreator :one
|
|
-- Boolean fast-path for the FAB's "running" indicator. Returns a single
|
|
-- EXISTS row instead of the full task list, so the planner can stop at the
|
|
-- first matching in-flight task (LIMIT 1 semantics via EXISTS).
|
|
--
|
|
-- Permission filtering is baked into the query: agent_id = ANY($3) restricts
|
|
-- the result to the agents the caller may currently see, so a member who lost
|
|
-- access to a private agent never gets a true from a task they can no longer
|
|
-- reach. The handler must pass its resolved accessible-agent id set as $3;
|
|
-- an empty array yields false.
|
|
SELECT EXISTS (
|
|
SELECT 1
|
|
FROM agent_task_queue atq
|
|
JOIN chat_session cs ON cs.id = atq.chat_session_id
|
|
WHERE atq.chat_session_id IS NOT NULL
|
|
AND atq.status IN ('queued', 'dispatched', 'running', 'waiting_local_directory', 'deferred')
|
|
-- Background quick-actions regeneration passes own no visible turn and must
|
|
-- never light the FAB "running" indicator (MUL-5149 refresh follow-up).
|
|
AND atq.regenerate_quick_actions_for IS NULL
|
|
AND cs.workspace_id = sqlc.arg(workspace_id)
|
|
AND cs.creator_id = sqlc.arg(creator_id)
|
|
AND cs.agent_id = ANY(sqlc.arg(agent_ids)::uuid[])
|
|
) AS has_pending;
|
|
|
|
-- name: MarkChatSessionRead :exec
|
|
-- Advances the read cursor to now, dropping the session's unread_count to 0.
|
|
UPDATE chat_session SET last_read_at = now()
|
|
WHERE id = $1;
|
|
|
|
-- name: GetMostRecentUserChatMessage :one
|
|
-- Returns the most recent role='user' message in a session. Used by the
|
|
-- Lark `/issue` command parser: when the user types `/issue` with no
|
|
-- title, the spec falls back to "use the previous user message as the
|
|
-- title". Bot replies (role='assistant') are excluded — only human
|
|
-- input qualifies as a fallback title source.
|
|
SELECT * FROM chat_message
|
|
WHERE chat_session_id = $1 AND role = 'user'
|
|
ORDER BY created_at DESC
|
|
LIMIT 1;
|
|
|
|
-- name: ChatSessionHasUserMessage :one
|
|
-- Reports whether a session has any human (role='user') message yet. Used to
|
|
-- scope the is_agent_intro self-introduction prompt to the very first,
|
|
-- server-driven turn: an intro session starts with zero user messages, so the
|
|
-- opening run gets the "introduce yourself" prompt. Once the creator replies,
|
|
-- later turns in the same session must fall back to the normal reply prompt
|
|
-- instead of repeating the introduction every turn (MUL-4259).
|
|
SELECT EXISTS (
|
|
SELECT 1 FROM chat_message
|
|
WHERE chat_session_id = $1 AND role = 'user'
|
|
) AS has_user_message;
|
|
|
|
-- name: CreateChatDraftRestore :one
|
|
-- Persists the deferred-cancellation draft restore (#5219) in the same tx
|
|
-- that deletes the triggering user message: the chat:cancel_finalized
|
|
-- broadcast is best-effort, so an offline client recovers the draft from
|
|
-- this row instead. id is the deleted message's id.
|
|
INSERT INTO chat_draft_restore (id, chat_session_id, task_id, content, attachment_ids)
|
|
VALUES ($1, $2, $3, $4, $5)
|
|
RETURNING *;
|
|
|
|
-- name: ListChatDraftRestoresBySession :many
|
|
SELECT * FROM chat_draft_restore
|
|
WHERE chat_session_id = $1
|
|
ORDER BY created_at ASC;
|
|
|
|
-- name: DeleteChatDraftRestore :execrows
|
|
-- Idempotent consume: deleting an already-consumed restore matches no row.
|
|
DELETE FROM chat_draft_restore
|
|
WHERE id = $1 AND chat_session_id = $2;
|
|
|
|
-- name: DeleteChatDraftRestoresBySession :exec
|
|
-- chat_draft_restore carries no chat_session FK (MUL-3515), so DeleteChatSession
|
|
-- prunes its pending restores in the same tx that deletes the session.
|
|
DELETE FROM chat_draft_restore
|
|
WHERE chat_session_id = $1;
|
|
|
|
-- The chat_session row lock is the mutual-exclusion protocol between the
|
|
-- draft-restore writer (FinalizeDeferredCancelledChat) and every deleter of a
|
|
-- chat_session or one of its cascade parents. Without an FK, an INSERT into
|
|
-- chat_draft_restore takes no lock on its session, so a prune-then-delete would
|
|
-- otherwise miss a restore committed after its snapshot and strand it — with the
|
|
-- user's prompt in it — forever (#5219).
|
|
--
|
|
-- The contract, held by all five paths below:
|
|
-- deleter: lock the sessions FOR UPDATE -> prune restores -> delete the parent
|
|
-- finalizer: lock the session FOR UPDATE -> insert the restore
|
|
-- Whoever locks first wins: a finalizer that got there first commits its row
|
|
-- before the deleter's prune statement takes its snapshot, so the prune sweeps
|
|
-- it; a deleter that got there first leaves no session for the finalizer to
|
|
-- lock, so it never inserts.
|
|
--
|
|
-- Lock order is chat_session -> agent_task_queue everywhere (the finalizer locks
|
|
-- the session before claiming its task) so the deleters' cascade into
|
|
-- agent_task_queue cannot deadlock against it.
|
|
--
|
|
-- The single-session delete path needs no new query: it already holds
|
|
-- LockChatSessionForDelete (same FOR UPDATE row lock) across its prune.
|
|
|
|
-- name: LockChatSessionForTask :one
|
|
-- The finalizer's half of the protocol. No rows means the session is already
|
|
-- gone (its cascade NULLs agent_task_queue.chat_session_id), so there is nothing
|
|
-- to lock and nothing to restore into.
|
|
SELECT cs.id
|
|
FROM agent_task_queue t
|
|
JOIN chat_session cs ON cs.id = t.chat_session_id
|
|
WHERE t.id = $1
|
|
FOR UPDATE OF cs;
|
|
|
|
-- name: LockChatSessionsByWorkspace :many
|
|
-- ORDER BY id: a stable lock order keeps two concurrent deleters from
|
|
-- deadlocking against each other.
|
|
SELECT id FROM chat_session
|
|
WHERE workspace_id = $1
|
|
ORDER BY id
|
|
FOR UPDATE;
|
|
|
|
-- name: LockChatSessionsBySystemRuntimeAgents :many
|
|
SELECT cs.id FROM chat_session cs
|
|
JOIN agent a ON a.id = cs.agent_id
|
|
WHERE a.runtime_id = $1 AND a.kind = 'system'
|
|
ORDER BY cs.id
|
|
FOR UPDATE OF cs;
|
|
|
|
-- name: DeleteChatDraftRestoresBySystemRuntimeAgents :exec
|
|
-- chat_session cascades from agent, so hard-deleting a runtime's system agents
|
|
-- silently drops their sessions — and, without an FK, would strand the pending
|
|
-- restores (which still hold the user's prompt text) forever. Prune them in the
|
|
-- same tx, BEFORE the agent rows go: the join below needs them. Mirrors
|
|
-- DeleteChannelInstallationsBySystemRuntimeAgents.
|
|
--
|
|
-- Only system agents are hard-deleted on runtime teardown since MUL-5559; user
|
|
-- agents (archived or not) are unbound and keep their sessions and restores.
|
|
DELETE FROM chat_draft_restore
|
|
WHERE chat_session_id IN (
|
|
SELECT cs.id FROM chat_session cs
|
|
JOIN agent a ON a.id = cs.agent_id
|
|
WHERE a.runtime_id = $1 AND a.kind = 'system'
|
|
);
|
|
|
|
-- name: GetChatMessageByTaskAssistant :one
|
|
-- The completed turn's assistant outcome row, for the quick-actions
|
|
-- supplement path (daemon suggestion pass finishing after chat:done).
|
|
SELECT * FROM chat_message
|
|
WHERE task_id = $1 AND role = 'assistant'
|
|
ORDER BY created_at DESC
|
|
LIMIT 1;
|
|
|
|
-- name: GetLatestAssistantChatMessageForSession :one
|
|
-- The session's most recent assistant turn, used as the regeneration target
|
|
-- when the user clicks "refresh" on the quick-actions row (MUL-5149). Only rows
|
|
-- with a task_id qualify — the daemon suggest supplement keys off task_id and
|
|
-- a resume needs a real completed turn to resume from.
|
|
SELECT * FROM chat_message
|
|
WHERE chat_session_id = $1 AND role = 'assistant' AND task_id IS NOT NULL
|
|
ORDER BY created_at DESC
|
|
LIMIT 1;
|
|
|
|
-- name: SetChatMessageQuickActionsByTask :one
|
|
UPDATE chat_message
|
|
SET quick_actions = $2
|
|
WHERE id = (
|
|
SELECT inner_msg.id FROM chat_message AS inner_msg
|
|
WHERE inner_msg.task_id = $1 AND inner_msg.role = 'assistant'
|
|
ORDER BY inner_msg.created_at DESC
|
|
LIMIT 1
|
|
)
|
|
RETURNING *;
|
|
|
|
-- name: GetOldestActiveChatSessionForCreatorAgent :one
|
|
-- Identity for "this member's conversation with this agent", independent of
|
|
-- the session title. Mika's onboarding session used to be matched on its
|
|
-- localized title from the client, which made the lookup both racy and
|
|
-- language-dependent. Oldest wins so the answer stays stable once a member
|
|
-- has opened more than one session with the same agent.
|
|
SELECT * FROM chat_session
|
|
WHERE workspace_id = $1
|
|
AND creator_id = $2
|
|
AND agent_id = $3
|
|
AND status = 'active'
|
|
ORDER BY created_at ASC
|
|
LIMIT 1;
|