Files
supabase/apps/studio/components/interfaces/SQLEditor/SQLEditorControllers.tsx
Charis 8b38e0d1ed feat(studio): ClickHouse dialect for logs snippet AI + rewrite to ClickHouse (#48501)
## I have read the
[CONTRIBUTING.md](https://github.com/supabase/supabase/blob/master/CONTRIBUTING.md)
file.

YES

## What kind of change does this PR introduce?

Feature, plus a refactor of the shared logs-rewrite flow.

PR 8 of the SQL editor query-source series. Stacked on #48457 — review
that one first, and merge this after it.

## What is the current behavior?

A `log_sql` snippet runs against the ClickHouse-backed analytics
endpoint, but the SQL editor's AI still writes Postgres: inline edits
get Postgres system prompts, and the result is run through
`sql-formatter`, which mangles ClickHouse backticks and `log_attributes`
map lookups.

Legacy Logs Explorer saved queries open in the editor as `log_sql`
snippets. Those are BigQuery dialect and error against the ClickHouse
endpoint the editor runs them on, with no in-editor way out — only the
Logs Explorer offered a rewrite.

The completion route was also asymmetric. It assembled a
schema/code/instruction message for Postgres but forwarded `prompt`
verbatim for ClickHouse, so a client wanting ClickHouse had to
hand-build the equivalent string.

## What is the new behavior?

**Inline AI speaks ClickHouse for logs snippets.** `sqlSourceToDialect`
maps a snippet's source to `postgres`/`clickhouse` and
`buildCompletionRequestBody` threads it through. For ClickHouse,
`useSqlEditorAi` strips code fences from the response and skips
`formatSql`. Execution and dialect both follow the snippet type, so a
snippet's valid dialect never flips.

**Rewrite to ClickHouse in the editor.** A banner offers the rewrite for
a logs snippet whose text trips `looksLikeLegacyLogsQuery`, and proposes
the result through the editor's existing AI diff view rather than
replacing the snippet, so it's accepted or discarded like any other AI
edit. Gated on `otelLegacyLogs`: on a non-migrated org the BigQuery text
is still correct, so rewriting it would break a working query.

The offer is a state machine (`offered` / `rewriting` / `failed` /
`noRewriteNeeded` / `dismissed`) with a declarative table of valid
transitions, so the states are mutually exclusive by construction and
dismissal is terminal. A failure keeps its message and offers a retry; a
response identical to the input is reported rather than opening an empty
diff.

**One place assembles completion prompts.** The route now uses a single
template for both dialects, branching only the schema section and — for
`intent: 'rewrite'` — the instruction. `lib/ai/clickhouse-logs.ts` is
the single home for ClickHouse-logs prompt content, replacing two
independently maintained descriptions of the same table. Clients carry
no prompt text.

**The rewrite flow is shared with the Logs Explorer.** Both surfaces
previously hand-rolled the same sequence and had drifted: only one
detected a no-op rewrite, they sourced `log_attributes` keys
differently, and the Explorer formatted errors with an `as Error` cast.
Both now use `useLegacyLogsRewrite` and the same state-driven banner, so
the Explorer picks up no-op detection and typed error extraction.

**Attribute keys are fetched on submit, not while typing.** The detected
source would otherwise feed a reactive query key, making every edit that
changed it cost another network call. `useLogsAttributeKeys` is
imperative and goes through `queryClient.fetchQuery`, so a source
already cached — including by the Explorer header and query panel, which
subscribe reactively — is reused. This also closes a gap where inline
edits never received keys at all, unlike full rewrites.

`getErrorMessage` gains an optional typed fallback and no longer
stringifies a bare object into `'[object Object]'`; every existing
caller already hand-rolled a fallback, except `QueueSettings`, which
interpolated the raw result and now passes one.

Nothing here is user-visible until the `sqlEditorLogsSource` flag is
enabled.

Tests: dialect selection and request-body shape, the ClickHouse prompt
content (including that the schema section does not restate the dialect
rules), the reducer's valid and invalid transitions,
`shouldOfferLegacyLogsRewrite`, on-submit key discovery with cache
reuse, and `getErrorMessage`.

## Additional context

<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->

## Summary by CodeRabbit

* **New Features**
* Added an Assistant banner to help rewrite legacy BigQuery-style logs
queries into ClickHouse SQL.
* SQL assistance now adapts to the selected query type, including
relevant log attribute context.
* Rewrite suggestions can be reviewed as editor diffs before being
applied.

* **Bug Fixes**
* Improved rewrite failure handling, retry options, dismissal behavior,
and “no rewrite needed” messaging.
* Error notifications now provide a clearer fallback message when
details are unavailable.

<!-- end of auto-generated comment: release notes by coderabbit.ai -->
2026-08-04 09:02:40 -04:00

279 lines
9.5 KiB
TypeScript

import { type UntrustedSqlFragment } from '@supabase/pg-meta'
import { useParams } from 'common'
import {
createContext,
use,
useCallback,
useEffect,
useEffectEvent,
useMemo,
useState,
type Context,
type Dispatch,
type PropsWithChildren,
type SetStateAction,
} from 'react'
import { useSqlEditorDiff, useSqlEditorPrompt } from './hooks'
import { type QuerySource } from './querySource'
import type { UtilityTab } from './SQLEditor.types'
import { useSQLEditorContext } from './SQLEditorContext'
import { useAddDefinitions } from './useAddDefinitions'
import { useEditorMount } from './useEditorMount'
import { useLogsSqlExecution } from './useLogsSqlExecution'
import { usePrettifyQuery } from './usePrettifyQuery'
import { useRunSource } from './useRunSource'
import { useSnippetIdentity } from './useSnippetIdentity'
import { useSnippetTitleGenerator } from './useSnippetTitleGenerator'
import { useSqlEditorAi } from './useSqlEditorAi'
import { useSqlEditorExecution } from './useSqlEditorExecution'
import { useSqlEditorShortcuts } from './useSqlEditorShortcuts'
import { isValidConnString } from '@/data/fetchers'
import {
untrustedLogSql,
type SafeLogSqlFragment,
type UntrustedLogSqlFragment,
} from '@/data/logs/safe-analytics-sql'
import { useReadReplicasQuery } from '@/data/read-replicas/replicas-query'
import { useSelectedProjectQuery } from '@/hooks/misc/useSelectedProject'
import { useDatabaseSelectorStateSnapshot } from '@/state/database-selector'
import {
getSqlEditorV2StateSnapshot,
useSqlEditorV2StateSnapshot,
} from '@/state/sql-editor/sql-editor-state'
import { createTabId, useTabsStateSnapshot } from '@/state/tabs'
/**
* Reactive SQL-editor controllers, distributed via a handful of focused
* contexts so the visual tree consumes exactly the slice it needs instead of
* prop-drilling whole hook bundles. The stable Monaco refs + imperative helpers
* live in `SQLEditorContext`; this provider owns everything that changes with
* state.
*
* SQL that a user runs is NOT promoted to safety here — the provider only
* exposes the already-safe `executeQuery` pipeline plus
* `readEditorSql` (which returns an `UntrustedSqlFragment`). Each user-action
* site (the toolbar/editor/results components and the shortcut handlers)
* promotes with `acceptUntrustedSql` right where the user acts, so the promotion
* is auditable.
*/
type SnippetContextValue = {
id: string
snippetName: string
isLoading: boolean
}
type AssistantContextValue = {
diff: ReturnType<typeof useSqlEditorDiff>
prompt: ReturnType<typeof useSqlEditorPrompt>
ai: ReturnType<typeof useSqlEditorAi>
}
type SqlEditorExecution = ReturnType<typeof useSqlEditorExecution>
type SqlEditorMount = ReturnType<typeof useEditorMount>
/** Running SQL: the safe execute pipeline, its status, and formatting. */
type RunContextValue = {
executeQuery: SqlEditorExecution['executeQuery']
readEditorSql: () => UntrustedSqlFragment | undefined
isExecuting: boolean
potentialIssues: SqlEditorExecution['potentialIssues']
resetPotentialIssues: () => void
prettifyQuery: () => void
runSource: QuerySource
executeLogsQuery: (sql: SafeLogSqlFragment) => void
readEditorLogsSql: () => UntrustedLogSqlFragment | undefined
}
/** Editor-surface UI state: selection, the active results tab, and mount. */
type UiContextValue = {
hasSelection: boolean
setHasSelection: Dispatch<SetStateAction<boolean>>
activeUtilityTab: UtilityTab
setActiveUtilityTab: Dispatch<SetStateAction<UtilityTab>>
onMount: SqlEditorMount['onMount']
}
const SnippetContext = createContext<SnippetContextValue | null>(null)
const AssistantContext = createContext<AssistantContextValue | null>(null)
const RunContext = createContext<RunContextValue | null>(null)
const UiContext = createContext<UiContextValue | null>(null)
function useGuardedContext<T>(context: Context<T | null>, hookName: string): T {
const value = use(context)
if (!value) {
throw new Error(`${hookName} must be used within a SQLEditorControllersProvider`)
}
return value
}
/** Snippet identity + display name for the current tab. */
export const useSqlEditorSnippet = () => useGuardedContext(SnippetContext, 'useSqlEditorSnippet')
/** The Assistant / diff cluster controllers (diff, prompt, ai). */
export const useSqlEditorAssistant = () =>
useGuardedContext(AssistantContext, 'useSqlEditorAssistant')
/** Run execution, warnings, and query formatting. */
export const useSqlEditorRun = () => useGuardedContext(RunContext, 'useSqlEditorRun')
/** Editor-surface UI state (selection, active results tab, editor mount). */
export const useSqlEditorUi = () => useGuardedContext(UiContext, 'useSqlEditorUi')
export const SQLEditorControllersProvider = ({ children }: PropsWithChildren) => {
const { monacoRef, scrollTopRef, editor } = useSQLEditorContext()
const { ref } = useParams()
const { data: project } = useSelectedProjectQuery()
const tabs = useTabsStateSnapshot()
const snapV2 = useSqlEditorV2StateSnapshot()
const { setSelectedDatabaseId } = useDatabaseSelectorStateSnapshot()
const diff = useSqlEditorDiff()
const { isDiffOpen } = diff
const prompt = useSqlEditorPrompt()
const { promptState, resetPrompt } = prompt
const [hasSelection, setHasSelection] = useState<boolean>(false)
const [activeUtilityTab, setActiveUtilityTab] = useState<UtilityTab>('results')
const { id, urlId, generatedNewSnippetName, isLoading } = useSnippetIdentity()
const { onMount, editorMountCount } = useEditorMount({ id })
const runSource = useRunSource(id)
useAddDefinitions(id, monacoRef.current, { enabled: runSource.type !== 'logs' })
const { data: databases, isSuccess: isSuccessReadReplicas } = useReadReplicasQuery(
{
projectRef: ref,
},
{ enabled: isValidConnString(project?.connectionString) }
)
const { setAiTitle } = useSnippetTitleGenerator()
const prettifyQuery = usePrettifyQuery({ id, isDiffOpen })
// Reads the SQL to run from the editor as an UntrustedSqlFragment. Promotion
// to safety (acceptUntrustedSql) happens at each user-action site, never here.
const readEditorSql = useCallback((): UntrustedSqlFragment | undefined => {
const snippet = getSqlEditorV2StateSnapshot().snippets[id]?.snippet
const fallback = snippet?.type === 'log_sql' ? undefined : snippet?.content?.unchecked_sql
return editor.getSql(fallback)
}, [editor, id])
// Reads the SQL to run from the editor as an UntrustedLogSqlFragment — the logs
// sibling of readEditorSql. Promotion (acceptUntrustedLogsSql) happens at each
// user-action site, never here.
const readEditorLogsSql = useCallback((): UntrustedLogSqlFragment | undefined => {
const snippet = getSqlEditorV2StateSnapshot().snippets[id]?.snippet
const fallback = snippet?.type === 'log_sql' ? snippet.content?.unchecked_sql : undefined
const sql = editor.getSql(fallback)
return sql === undefined ? undefined : untrustedLogSql(sql)
}, [editor, id])
const {
executeQuery,
isExecuting: isExecutingDb,
potentialIssues,
resetPotentialIssues,
} = useSqlEditorExecution({
id,
isDiffOpen,
hasSelection,
setAiTitle,
})
const { executeLogsQuery, isExecuting: isExecutingLogs } = useLogsSqlExecution({ id })
const isExecuting = isExecutingDb || isExecutingLogs
const ai = useSqlEditorAi({ id, editorMountCount, diff, prompt, sqlSource: runSource.type })
const { acceptAiHandler, discardAiHandler } = ai
useSqlEditorShortcuts({
isDiffOpen,
isPromptOpen: promptState.isOpen,
prettifyQuery,
acceptAiHandler,
discardAiHandler,
resetPrompt,
})
const saveScrollPosition = useEffectEvent((snippetId: string) => {
if (ref) {
const tabId = createTabId('sql', { id: snippetId })
tabs.updateTab(tabId, { scrollTop: scrollTopRef.current })
}
})
useEffect(() => {
// Save the departing snippet's scroll position on unmount / snippet switch.
return () => saveScrollPosition(id)
}, [id])
useEffect(() => {
if (isSuccessReadReplicas) {
const primaryDatabase = databases.find((db) => db.identifier === ref)
setSelectedDatabaseId(primaryDatabase?.identifier)
}
}, [isSuccessReadReplicas, databases, ref, setSelectedDatabaseId])
const snippetName =
urlId === 'new'
? generatedNewSnippetName
: (snapV2.snippets[id]?.snippet.name ?? generatedNewSnippetName)
const snippetValue = useMemo<SnippetContextValue>(
() => ({ id, snippetName, isLoading }),
[id, snippetName, isLoading]
)
const assistantValue = useMemo<AssistantContextValue>(
() => ({ diff, prompt, ai }),
[diff, prompt, ai]
)
const runValue = useMemo<RunContextValue>(
() => ({
executeQuery,
readEditorSql,
isExecuting,
potentialIssues,
resetPotentialIssues,
prettifyQuery,
runSource,
executeLogsQuery,
readEditorLogsSql,
}),
[
executeQuery,
readEditorSql,
isExecuting,
potentialIssues,
resetPotentialIssues,
prettifyQuery,
runSource,
executeLogsQuery,
readEditorLogsSql,
]
)
const uiValue = useMemo<UiContextValue>(
() => ({ hasSelection, setHasSelection, activeUtilityTab, setActiveUtilityTab, onMount }),
[hasSelection, activeUtilityTab, onMount]
)
return (
<SnippetContext.Provider value={snippetValue}>
<AssistantContext.Provider value={assistantValue}>
<RunContext.Provider value={runValue}>
<UiContext.Provider value={uiValue}>{children}</UiContext.Provider>
</RunContext.Provider>
</AssistantContext.Provider>
</SnippetContext.Provider>
)
}