mirror of
https://github.com/supabase/supabase.git
synced 2026-10-05 09:25:06 +03:00
bot/decrease-eslint-ratchet-baselines
100
Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
 Charis
|
84e7316d8d |
disable coderabbit high-level summary (#51180)
<!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Chores** * No user-facing changes are included in this update. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
0daafca2ca |
feat(studio): status page banner (incident / maintenance / upcoming) (#51044)
## Summary * Adds a new global status banner (`StatusBanner`) driven by the [incident.io](<http://incident.io>) status page data, showing at most one of: an active incident, in-progress maintenance, or upcoming maintenance, in that priority order. * All three types are independently dismissible (persisted to a new localStorage key, `status-banner-dismissed-keys`); dismissal hides the banner for items still active, and a new incident reappears even if a related item was previously dismissed. * Only shows to users who are actually affected (based on their projects' regions) or when region data is incomplete (fails open), and is bypassed entirely by the existing emergency incident override. * Behind the existing `incidentIoStatusPage` ConfigCat flag — `AppBannerWrapper` renders this new banner instead of the legacy `StatusPageBanner` only when the flag is on; default behavior is unchanged. * This is PR 5b in a stacked series for Linear [FE-4057](https://linear.app/supabase/issue/FE-4057) — see that issue for full design context. ## Test plan * New unit tests (`StatusBanner.utils.test.ts`) covering banner-selection priority, dismissal-key handling, and copy generation * New MSW component test (`StatusBanner.test.tsx`) covering loading state, dismiss-and-persist, and the emergency-override path * `pnpm --filter studio run typecheck`, `lint:ratchet`, `pnpm knip --workspace apps/studio`, `pnpm test:prettier`, and relevant vitest suites all pass 🤖 Generated with [Claude Code](<https://claude.com/claude-code>) Co-Authored-By: Claude [noreply@anthropic.com](<mailto:noreply@anthropic.com>) <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Added status banners for relevant incidents and scheduled maintenance, including incident details, maintenance timing, and links to the status page. * Banners can be dismissed, and dismissed items stay hidden while new incidents or maintenance updates can still appear. * Upcoming maintenance banners appear within the relevant lead time, and maintenance timing is shown when available. * Emergency overrides display a warning banner without a dismiss option. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
d22702e907 |
refactor(studio): extract user project regions + emergency override hooks (#51033)
## Summary - Extracts `useUserProjectRegions` (org + project region aggregation, fail-open on fetch errors) and `useEmergencyIncidentOverride` (the `ongoingIncident` flag / env var check) out of `useStatusPageBannerVisibility`, so the upcoming incident.io status-page banner can reuse both without duplicating the org/project fan-out logic. - No behavior change for the legacy banner except one intentional fix: `StatusPageBanner.utils.ts` compared the incident's affected regions (lowercased) against the user's regions (not normalized), so a mixed-case region on either side could silently fail to match. Both sides now go through the same `normalizeRegion` helper. - Part of the Linear FE-4057 stack (PR 5a of 6). Base branch is PR 4 (`charis/fe-4057-pr4-project-creation-status-admonition`), not master. Linear: FE-4057 ## Test plan - [x] `pnpm --filter studio run typecheck` - [x] `pnpm --filter studio run lint:ratchet` - [x] `pnpm knip --workspace apps/studio` - [x] `pnpm test:prettier` - [x] `pnpm --filter studio exec vitest run` on the touched test files and the `hooks/misc/` and `components/layouts/AppLayout/` directories (all passing, no regressions) <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Bug Fixes** * Improved status banner targeting by matching incidents against normalized regions across the user’s projects. * Updated banner visibility checks to handle incomplete project or region data, including when project information fails to load. * Improved loading behavior so the banner can be evaluated based on user-region data rather than waiting for separate organization and project requests. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
53b20cad5e |
feat(studio): project creation status admonition (#51029)
## Summary - Adds a new project-creation-form incident warning path behind the `incidentIoStatusPage` ConfigCat flag (default off — no visible behavior change until the flag is enabled) - When the flag is on, reads the new `/api/status-page` endpoint (added earlier in this stack) instead of the legacy `/api/incident-status` endpoint, and only fetches the legacy endpoint when the flag is off - Extracts region-matching logic into `RegionSelector.utils.ts` (`regionMatches`, `getItemsAffectingProjectCreation`) with unit test coverage See Linear FE-4057 for full context and design. ## Test plan - [x] `pnpm --filter studio run typecheck` - [x] `pnpm --filter studio run lint:ratchet` - [x] `pnpm knip --workspace apps/studio` - [x] `pnpm test:prettier` - [x] `pnpm --filter studio exec vitest run components/interfaces/ProjectCreation/RegionSelector.utils.test.ts` (35/35 passing) <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Project creation displays a warning and links to the status page when an incident or maintenance event may affect the selected region. * Status notices match exact regions and broader smart-region groupings; global project-creation notices are also shown. * The AI assistant can report visible, active incidents and in-progress maintenance, including affected components and update details. When there are no active events, it reports that status. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
13de8189c9 |
feat(studio): assistant get_active_incidents on status page endpoint (#50994)
## Summary * Adds a second branch to the assistant's `get_active_incidents` tool (`apps/studio/lib/ai/tools/incident-tools.ts`) that reads `/api/status-page` when the global ConfigCat flag `incidentIoStatusPage` is on, instead of the legacy `/api/incident-status` endpoint. * Filters on `visible`, ignores `show_banner` (only the global banner respects it), and concatenates ongoing incidents + in-progress maintenances (not scheduled maintenances). * Adds `isServerFlagEnabled` to `lib/server/configcat.ts` for reading global, non-user-targeted ConfigCat flags server-side, and threads the flag through `getTools` from `pages/api/ai/sql/generate-v4.ts`. This is PR 3 of 6 in the [FE-4057](https://linear.app/supabase/issue/FE-4057/frontend-bannerbot-reconfigured) stack — stacked on `charis/fe-4057-pr2-support-form`. Behind the `incidentIoStatusPage` flag (default off), so this ships no user-visible change on its own. See Linear [FE-4057](https://linear.app/supabase/issue/FE-4057/frontend-bannerbot-reconfigured) for full context. ## Test plan - [X] `pnpm --filter studio run typecheck` - [X] `pnpm --filter studio run lint:ratchet` - [X] `pnpm knip --workspace apps/studio` - [X] `pnpm test:prettier` - [X] `pnpm --filter studio exec vitest run lib/ai/tools/incident-tools.test.ts` (19/19 passing, including 7 new tests for the status-page branch) |
||
|
Charis
|
5573d0dfbc |
fix(studio): disable network restrictions for v3 (AWS_K8S) projects (#50996)
## Summary - Network restrictions aren't supported on v3 (AWS_K8S) projects, but the Network Restrictions settings section previously only disabled itself for High Availability projects, leaving a fully working (but non-functional) UI for non-HA v3 projects. - Adds `useIsAwsK8sCloudProvider()` to the section's disabled check, alongside a v3-specific disabled notice mirroring the existing High Availability disabled notice pattern. Resolves [FE-4482](https://linear.app/supabase/issue/FE-4482/disable-network-restrictions-for-v3-aws-k8s-projects). ## Test plan - [x] `tsc --noEmit` passes for the changed file - [x] `eslint` passes for the changed file - [x] `prettier --check` passes for the changed file - [x] Mirrors the existing, already-shipped High Availability disabled-section pattern in the same component 🤖 Generated with [Claude Code](https://claude.com/claude-code) <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Behavior Changes** * Network restrictions are unavailable for AWS Kubernetes and High Availability projects, and for users without update permission. AWS Kubernetes projects display a dedicated explanation; restriction details and management controls are hidden. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
8b8569bd1a |
feat(studio): status page client data layer + support form (#50984)
## Summary * Adds the client query layer for `/api/status-page` (`statusPageQueryOptions` in `data/platform/status-page-query.ts`) and a shared normalization module (`lib/status-page/status-page.utils.ts`) that maps the endpoint response into region- and project-creation-aware `StatusItem`s — later PRs in this stack (assistant tool, project-creation admonition, global banner) build on this same module. * Wires the support form's status pill and incident admonition to the new data behind the `incidentIoStatusPage` ConfigCat flag. `useSupportStatus` branches between the new endpoint and the existing `useIncidentStatusQuery`/`processIncidentData` path, both mapped to the same `SupportStatus` shape. * `IncidentAdmonition` becomes purely presentational; the status link now points at the `pageUrl` returned by the endpoint instead of a hardcoded URL. Part of [FE-4057](https://linear.app/supabase/issue/FE-4057/frontend-bannerbot-reconfigured) — see Linear for full design context. ## Test plan - [X] `pnpm --filter studio run typecheck` - [X] `pnpm --filter studio run lint:ratchet` - [X] `pnpm knip --workspace apps/studio` - [X] `pnpm test:prettier` - [X] `pnpm --filter studio exec vitest run` (touched files) — 70 tests passing, including a flag-on case for `SupportFormPage` <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Support pages now display current incident and maintenance information, including relevant status descriptions and links to the status page. * Status details can reflect items affecting a user’s region or project-creation services. Upcoming maintenance is excluded from active alerts. * **Bug Fixes** * Status labels and alerts now account for loading, errors, incidents, and maintenance consistently across support pages. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
CharisandClaude Code
|
b2cf3693dd |
feat(studio): /api/status-page endpoint backed by incident.io Widget API (#50931)
## Summary * Adds `/api/status-page` (Next route + TanStack wrapper), backed by the [incident.io](<http://incident.io>) Widget API, annotating each item with `visible`, `show_banner`, and (for scheduled maintenances) `banner_lead_days`. * Deployment-mode visibility is driven by a new `status_page:visibility_field_ids` custom-content key. * Widget array parsing is fault-tolerant: a malformed item in one array is dropped and logged rather than failing the whole response, so one bad item can't hide a real ongoing incident. * 429s from [incident.io](<http://incident.io>) are retried with equal-jitter exponential backoff, respecting `Retry-After`, up to 2 retries. * Nothing consumes this endpoint yet — it replaces no existing behavior and changes nothing user-visible. Later PRs (this is PR 1 of a stack) wire up consumers behind the `incidentIoStatusPage` ConfigCat flag. Part of [FE-4057](https://linear.app/supabase/issue/FE-4057/frontend-bannerbot-reconfigured) — see Linear for full design context. ## Test plan - [X] `pnpm --filter studio run typecheck` - [X] `pnpm --filter studio run lint:ratchet` - [X] `pnpm knip --workspace apps/studio` - [X] `pnpm test:prettier` - [X] `pnpm --filter studio exec vitest run status-page` — 44 tests passing, including a regression test built from a real production [incident.io](<http://incident.io>) payload that initially failed to parse, and a compile-time type-safety regression test for the array-parsing helper Co-authored-by: Claude Code [charis@supabase.io](<mailto:charis@supabase.io>) <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Added a status page that displays ongoing incidents and maintenance, with visibility and banner settings based on linked incident details. * Status page data is available through a new API endpoint, with caching for successful responses and degraded results. * **Bug Fixes** * Status page data can still display when some linked incident details are unavailable; affected results are marked as degraded. * Improved handling of invalid widget entries so they don’t prevent valid items from being processed. <!-- end of auto-generated comment: release notes by coderabbit.ai --> --------- Co-authored-by: Claude Code <charis@supabase.io> |
||
|
Charis
|
5b50701c11 |
fix(studio): surface link to Explorer start-page preference (#50670)
## Summary * Added a persistent "Preferences" link (with Settings icon) pinned to the bottom of the Explorer sidebar's root navigation * Links to `/account/me#dashboard`, the existing Account settings anchor that contains the Explorer startup preference dropdown * Restructured sidebar layout to keep the footer link pinned while content scrolls, addressing [FE-4437](https://linear.app/supabase/issue/FE-4437/give-higher-visibility-to-changing-explorer-start-page-preference) * The Home tab disappears when "SQL query" is selected as the start page, so this visible link ensures users can always find and change their preference ## Test plan - [X] Typecheck passes on `apps/studio/components/layouts/ExplorerLayout/ExplorerNavHome.tsx` - [X] ESLint passes on the changed file - [X] Manual verification of sidebar behavior and link functionality Fixes [FE-4437](https://linear.app/supabase/issue/FE-4437/give-higher-visibility-to-changing-explorer-start-page-preference): "Give higher visibility to changing Explorer start page preference" <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit - **New Features** - Added a fixed **Preferences** link to the Explorer navigation footer. - **UI Improvements** - Updated Explorer navigation layout so resource links and recently updated items scroll independently from the footer. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
ea203f70df |
fix(studio): use configured gp3 max-IOPS ceiling (#50269)
## Summary * GP3 IOPS calculation was hardcoded to 16,000 instead of reading from DISK_LIMITS config * AWS updated the real ceiling to 80,000, making the hardcoded constant stale * Users with large multi-TB GP3 disks were incorrectly blocked from provisioning above 16,000 IOPS ## Test plan - [X] Updated unit tests for `calculateMaxIopsAllowedForDiskSizeWithGp3` now assert correct behavior: scaling linearly (100 GB → 50,000 IOPS) and capping at new 80,000 ceiling (1000 GB → 80,000 IOPS) - [X] All 26 tests in DiskManagement.test.ts pass locally - [X] Manually verify Infrastructure Settings > Disk IOPS field no longer blocks GP3 disks above 16,000 IOPS up to 80,000 Closes [FE-4379](https://linear.app/supabase/issue/FE-4379/update-iops-limits-for-new-aws-disk-capacity) <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit - **Bug Fixes** - Updated GP3 disk performance calculations to support up to 80,000 IOPS. - Disk sizes below the minimum threshold continue to receive the correct 3,000 IOPS floor. - Larger disks now scale linearly until reaching the updated maximum IOPS limit. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
CharisandJoshen Lim
|
82d7d347c4 |
fix(studio): eliminate per-row query duplication on org team page (#50238)
## Summary * Fixes extreme slowness (browser-crashing on filter) on `/org/[slug]/team` for orgs with 200+ members. * Root cause: `MemberRow`/`MemberActions` each independently subscribed to org-wide React Query data (roles, projects, permissions, feature flags) and rendered a hidden `UpdateRolesPanel` per row. Filtering caused hundreds of duplicate query observers to mount/unmount on every keystroke, each scheduling its own stale-timeout bookkeeping and blocking the main thread for multiple seconds. * Hoisted all org-wide data fetching (`useOrganizationRolesV2Query`, `useOrgProjectsInfiniteQuery`, `usePermissionsQuery`, `useSelectedOrganizationQuery`, `useIsFeatureEnabled`) to `MembersView` and passed the results down as props. * Replaced the per-row `useAsyncCheckPermissions` hook calls in `MemberActions` with the underlying pure `doPermissionsCheck` function memoized locally, removing their internal query subscriptions. * Simplified `useGetRolesManagementPermissions` to stop calling a query-fetching fallback hook that was unreachable given all current call sites already pass `permissions`/`orgSlug` directly. * Replaced 200 hidden per-row `UpdateRolesPanel` instances with a single shared instance owned by `MembersView`, opened via an `onManageAccess` callback. * Cached the regex built by `doPermissionsCheck`'s `toRegexpString` instead of rebuilding it on every permission check. Diagnosed from two Chrome performance traces of the team page while typing in the filter box (multi-second main-thread blocking tasks traced to React Query `QueryObserver` mount/unmount storms). ## Test plan - [X] `tsc --noEmit` clean (only one pre-existing, unrelated error in `packages/ui-patterns`) - [X] `eslint` clean on all changed files (only pre-existing warnings) - [X] `vitest run tests/components/Organization/TeamSettings` — 51 tests pass - [X] Manually verify filtering is smooth on an org with 200+ members <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Team Settings provides centralized member access and role management. * Members can update roles through the access-management panel. * **Improvements** * Permission checks now more accurately handle organization and project scopes, including wildcard patterns. * Member search is debounced for smoother filtering while typing. * Access-management actions use current organization members, roles, permissions, and feature settings. <!-- end of auto-generated comment: release notes by coderabbit.ai --> --------- Co-authored-by: Joshen Lim <joshenlimek@gmail.com> |
||
|
Charis
|
0bebe50a76 |
fix(studio): serialize SQL folder deletion IDs (#50223)
## Summary * serialize SQL folder IDs as the comma-delimited API query value * add regression coverage for the folder deletion request ## Testing * `pnpm --filter studio exec vitest run data/content/sql-folders-delete-mutation.test.ts` * `pnpm exec prettier --check apps/studio/data/content/sql-folders-delete-mutation.ts apps/studio/data/content/sql-folders-delete-mutation.test.ts` * `pnpm --filter studio exec eslint data/content/sql-folders-delete-mutation.ts data/content/sql-folders-delete-mutation.test.ts` * `pnpm --filter studio exec tsc --noEmit --pretty false` <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Bug Fixes** * Fixed SQL snippet folder deletion requests so multiple selected folders are processed correctly. * Improved request handling by formatting folder identifiers consistently when submitting bulk deletions. * **Tests** * Added coverage to verify that deleting multiple SQL snippet folders sends the expected folder identifiers. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
CharisandJoshen Lim
|
2ac052752a |
fix(pg-meta): fix invalid SQL when setting column default to NULL (#50097)
## Summary
- Fixes invalid SQL generation when setting a column's default value to
NULL through Studio's Table Editor
- The bug occurred because raw `null` values bypassed literal
serialization in the SQL builder
- Added special case handling in both `create()` and `update()`
functions to use `literal(null)` regardless of format
## Details
Setting a column's default to NULL via the "Set as NULL" suggestion in
Studio generated invalid SQL: `ALTER TABLE ... ALTER COLUMN ... SET
DEFAULT ;` (missing the NULL keyword). This was caused by
`default_value` being JS `null` and bypassing the literal()
serialization function when `default_value_format === 'expression'`.
The fix ensures that when `default_value === null`, we always use
`literal(null)` to produce the correct `SET DEFAULT NULL` clause.
## Test plan
- New regression test in `packages/pg-meta/test/columns.test.ts`
("update column default value to NULL via expression format") verifies
the fix by:
- Creating a text column with a literal default
- Updating it using `default_value_format: 'expression', default_value:
null` (exact payload from Studio)
- Asserting the column's default value is null
- All 58 tests in `packages/pg-meta/test/columns.test.ts` pass
- TypeScript check (`tsc --noEmit`) passes
Fixes FE-4343
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
- **Bug Fixes**
- Fixed column creation and updates so explicitly setting a default
value to `NULL` correctly preserves the `DEFAULT NULL` clause.
- Ensured setting a column default to `NULL` through expression-based
editing is reflected accurately when column details are retrieved.
- **Style**
- Improved the column editor’s suggestion menu layout and spacing for
clearer readability.
- **Tests**
- Added coverage for updating a column default to `NULL`.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
---------
Co-authored-by: Joshen Lim <joshenlimek@gmail.com>
|
||
|
Charis
|
1308a1d0d0 |
fix(studio): close popover menu when dragging block (FE-4301) (#49912)
## Summary * Closes the block options popover (grip dropdown menu) when a drag operation starts on that block * Fixes the issue where the menu would remain visible during the drag if it was already open <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Bug Fixes** * Section action menus now remain closed when dragging begins, preventing delayed reopening and keeping the editing interface clear and focused. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
CharisandJoshen Lim
|
2ee0108928 |
studio: increase gp3 max size to new aws limit (#45611)
<!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Increased GP3 disk limits to support up to 64 TB of storage, 80,000 IOPS, and 2,000 MB/s throughput. * **Documentation** * Updated the disk comparison table to reflect the expanded GP3 limits. * **Bug Fixes** * Improved disk-size validation and recommendation messaging for clearer wording. <!-- end of auto-generated comment: release notes by coderabbit.ai --> --------- Co-authored-by: Joshen Lim <joshenlimek@gmail.com> |
||
|
Charis
|
a4d925c230 |
fix(studio): write debug prompt into current chat when assistant is already open (#49911)
## Summary * Query blocks embedded inside an active assistant conversation (`AssistantQueryCell`) reused the same "Debug with Assistant" handler as standalone query blocks (Explorer Query tab, notebook cells), which always opens a brand-new chat and navigates away. * Clicking Debug on a block that's already part of the open conversation silently abandoned it for an unrelated new chat, which read as the button doing nothing. * Added an optional `onDebug` override threaded through `QueryEditor` → `QueryResultRenderer` → `QueryResultError`; `AssistantQueryCell` now uses it to write the debug prompt into the currently active chat's composer (`ai-assistant-state`'s new `setInitialInput`) instead of creating a new chat. Standalone query blocks keep the existing "open a new chat" behavior since no `onDebug` override is passed there. * `ExplorerChatTab` now wires `composerContext` into `AssistantChat` (it wasn't before), so the pre-filled prompt actually reaches the visible textarea on the Explorer chat route. Fixes [FE-4319](https://linear.app/supabase/issue/FE-4319/debug-with-ai-assistant-does-seemingly-nothing-if-query-is-already). ## Test plan - [X] `pnpm vitest run` on `QueryResultError.test.tsx` / `QueryResultError.selfhosted.test.tsx` / `ExplorerChatTab.test.tsx` / `AssistantQueryCell.utils.test.ts` — all pass, including new test asserting `onDebug` is called instead of `createChat`. - [X] `pnpm exec eslint` on touched files — clean (only pre-existing unrelated warnings). - [X] Manual check: run a query inside an assistant chat that errors, click "Debug with Assistant" on that block, confirm the debug prompt appears in the current chat's composer rather than opening a new chat. <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Added a “Debug with Assistant” workflow that sends SQL error details to the AI Assistant as its initial input. * Preserved the existing behavior of opening a new debug chat when the Assistant panel is unavailable. * **Tests** * Added coverage confirming that debugging invokes the Assistant callback without creating an additional chat. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
357d6cd3eb |
add github discussion link to explorer preview (#49918)
Adds the GitHub discussion link for Explorer feature preview. Resolves FE-4256 <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Bug Fixes** * Updated the Explorer & Notebooks feature preview with a working link to its GitHub discussion. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
e7d91dbd06 |
fix(studio): display diff for view-only notebook cell edits (#49901)
## Summary Fixed a bug in the AI Assistant notebook-update proposal preview where a `replace_cell` operation that only changed a cell's view (table ↔ chart) or chart parameters (type, x/y columns, cumulative, scale, labels) would show as a "Replaced" row but the expanded diff would appear empty. **Root cause:** The diff editor only compared the cell's SQL text; view and chart configuration were never considered, so changes to those aspects showed no diff. **Solution:** * Refactored `getCellMetadata` to return structured `NotebookCellFields` with separate `source` (database/time range) and `view` (table/chart) fields instead of a single concatenated string * Added `formatChartConfig` and `formatCellView` helpers to describe chart cells * Updated `getEntryMetadata` to diff source and view independently, showing only the fields that actually changed (e.g., "Table → Chart (bar, ...)" when only the view changed, with the unchanged database omitted) * If neither field changed, metadata is hidden entirely ## Test plan * Added test cases for: chart-view cells reporting a `view` field, view-only changes surfacing without the unchanged database, chart-parameter-only changes surfacing without the unchanged database, database-only changes surfacing without the unchanged view, and fully-unchanged replacements hiding metadata entirely * All 43 tests in the touched test file pass * `tsc --noEmit` on apps/studio shows no new type errors ## Summary by CodeRabbit * **Enhancements** * Improved AI Assistant notebook previews with clearer cell details, including source content and table or chart views. * Chart previews now show key configuration details, such as chart type and selected dimensions * Replacement previews highlight only the fields that changed and hide entries with no visible changes. <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Notebook previews now distinguish cell content from its view, including table and chart details. * Chart previews display relevant configuration, such as chart type and axes. * Log previews include their formatted time range. * Replacement previews now show only the fields that changed. * **Bug Fixes** * Unchanged replacements are now hidden instead of displaying misleading content. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
9aaa753306 |
feat(studio): add telemetry for explorer/sql editor temporary switch buttons (#49898)
## Summary Add PostHog event tracking for the two new buttons introduced in PR supabase/supabase#49698 that allow users to temporarily switch between the Explorer and SQL Editor: * **Explorer button**: "Back to SQL Editor" button in the Explorer sidebar title bar now fires `explorer_temp_access_sql_editor_clicked` event * **SQL Editor button**: "Back to Explorer" button in the SQL Editor title bar (shown during temporary visits) now fires `sql_editor_back_explorer_clicked` event Both event interfaces follow the repo's telemetry-standards conventions, carrying only `groups: TelemetryGroups` property with no additional custom properties. ## Test plan - [X] Verify `explorer_temp_access_sql_editor_clicked` event fires in PostHog when clicking "Back to SQL Editor" button in Explorer - [X] Verify `sql_editor_back_explorer_clicked` event fires in PostHog when clicking "Back to Explorer" button in SQL Editor - [X] Run typecheck: `pnpm typecheck` passes without errors - [X] Run lint: `pnpm lint --filter=studio` passes ## Issue Resolves [FE-4213](https://linear.app/supabase/issue/FE-4213/explorer-set-up-telemetry-for-metrics-where-appropriate) ## Summary by CodeRabbit * **Analytics** * Added tracking for navigation from the Explorer to the SQL Editor. * Added tracking for returning from the SQL Editor to the Explorer. |
||
|
Charis
|
17d15ee1e2 |
fix(studio): confirm destructive notebook queries (#49658)
## What - add a second notebook-level confirmation for destructive SQL before forced batch execution - reuse the shared SQL safety detector for stored and live cell SQL - cover destructive confirmation, cancellation, non-destructive mutations, and live SQL ## Testing - pnpm --filter studio exec vitest run components/interfaces/Explorer/ExplorerNotebookTab.utils.test.ts components/interfaces/Explorer/__tests__/ExplorerNotebookTab.test.tsx --coverage.enabled=false - pnpm --filter studio exec eslint components/interfaces/Explorer/ExplorerNotebookTab.tsx components/interfaces/Explorer/ExplorerNotebookTab.utils.ts apps/studio/components/interfaces/Explorer/ExplorerNotebookTab.utils.test.ts apps/studio/components/interfaces/Explorer/__tests__/ExplorerNotebookTab.test.tsx Closes FE-4284 <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Improvements** * Simplified notebook execution into a single confirmation step for mutating queries. * Destructive queries, including operations such as `DROP` or `TRUNCATE`, are clearly marked with a **Destructive** badge. * The confirmation dialog lists affected queries and lets you proceed or cancel. * Detection uses the latest SQL from the editor and ignores destructive keywords in comments. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
8db7368ccf |
feat(studio): clarify quick query as temporary space (#49695)
## 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 ## What is the current behavior? Standalone queries in the Explorer "Quick Query" surface have editable titles (defaulting to "Untitled query") without clear indication that they are temporary and not saved, similar to the old SQL Editor snippet model. ## What is the new behavior? The Quick Query tab now clearly indicates that standalone queries are temporary. The title is no longer editable and displays "Temporary space for one-off queries" as static muted italic text. The default query name has been changed from "Untitled query" to "Quick query" for clarity. ## Additional context Resolves FE-4297. Notebook cells (variant === 'embedded') remain unchanged with editable titles. <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **UI Improvements** * Renamed new one-off queries from **“Quick query”** to **“Run SQL”** when no title is provided. * Updated viewport query editor toolbars to display the fixed **“Run SQL”** label. * Preserved editable titles for embedded query editors. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
4b1f93bb99 |
feat(explorer): add path back to SQL Editor for snippet access (#49698)
## 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 ## What is the current behavior? Users who have opted into the Explorer feature preview have no way back to the SQL Editor from within Explorer, so they can't easily check their old snippets. ## What is the new behavior? - The Explorer sidebar title bar now has a button (using the same icon as the SQL Editor/Explorer nav entry) that links to the SQL Editor, with a tooltip explaining it's a temporary switch to access snippets. - Clicking it marks the visit as temporary in localStorage, which surfaces a matching "Back to Explorer" button in the SQL Editor title bar. Clicking that button clears the temporary flag and returns to Explorer. - Fixed the product menu title bar badge slot to sit flush right instead of directly next to the title text. ## Additional context <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Added a quick switch from the Explorer to the project’s SQL Editor. * Added a “Back to Explorer” option in the SQL Editor when opened from Explorer. * Added tooltips to clarify these navigation actions. * Navigation state is preserved per project for a smoother return experience. * **UI Improvements** * Improved product menu spacing and title truncation for better layout handling. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
b278b1ec8a |
fix(studio): Debug with Assistant and Copy prompt work (#49690)
## Summary
* Resolved hanging buttons in Explorer's QueryResultError panel that
were wired to stub no-ops (`buildPrompt={() => ''}`,
`onOpenAssistant={() => {}}`).
* "Debug with Assistant" now opens a new chat seeded with a real prompt
combining the SQL query and error context using existing
`buildDebugPromptText` util and `useCreateChat` hook.
* "Copy prompt" now copies the same real debug prompt text to clipboard.
* Threaded `sql` and query `source` props down through `QueryEditor` →
`QueryResultRenderer` → `QueryResultError` while keeping them optional
for backward compatibility with other callers like
`AssistantNotebookPreviewCell`.
## Test plan
- [X] Run `pnpm typecheck` — passes
- [X] Run `pnpm lint --filter=studio` — passes
- [X] Run `pnpm test:studio --run
apps/studio/components/interfaces/Explorer/QueryEditor` — Explorer
vitest suite (99 tests across 13 files) passes with no regressions
- [X] Manually verify in Explorer: trigger an ad-hoc SQL query that
fails, confirm "Debug with Assistant" opens a new chat with the error
prompt seeded, and "Copy prompt" copies the prompt to clipboard
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
- **New Features**
- Added “Debug with Assistant” to query errors using the submitted SQL
and error details.
- Added an option to copy the debugging prompt for easier
troubleshooting.
- Assistant actions are hidden when query details are unavailable or
restricted.
- **Bug Fixes**
- Ensured query errors reference the SQL that produced them, rather than
later editor changes.
- **Tests**
- Added coverage for assistant debugging, prompt copying, conditional
visibility, and self-hosted behavior.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
|
||
|
Charis
|
86c813ec03 |
fix(notebooks): reset insert offset when anchor cell moves (#49694)
## 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? Bug fix ## What is the current behavior? When a cell gets moved via `move_cell` operation in `deriveNotebookDiff`, the `insertedAfter` offset map is not cleared for that anchor cell. This causes later `insert_cell` operations anchored on the same (now-moved) cell to apply the stale offset on top of the correct current-position lookup, resulting in the new cell landing after the wrong position. ## What is the new behavior? The offset for an anchor cell is now cleared from `insertedAfter` when it gets moved, since cells previously inserted after it stay behind at its old location and should not affect subsequent inserts at its new position. A regression test has been added that reproduces the exact ticket scenario (insert after cell-1, move cell-1 after cell-3, insert after cell-1 again) and verifies the correct final cell order. ## Additional context Fixes: https://linear.app/supabase/issue/FE-4308/insert-anchored-to-a-previously-moved-cell-lands-after-the-wrong-cell <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Bug Fixes** * Fixed notebook cell insertions after moving an anchor cell, ensuring new inserts appear relative to the anchor’s updated position. * Preserved the placement of inserts made before the anchor cell was moved. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
b366496c8c |
[FE-4246] fix(studio): resolve notebook database labels (#49554)
## 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? Bug fix for Studio notebook previews. ## What is the current behavior? Notebook diffs display the raw database_identifier value. This exposes opaque database IDs and does not communicate whether the cell targets the primary database or a read replica. Related issue: [https://linear.app/supabase/issue/FE-4246](<https://linear.app/supabase/issue/FE-4246>) ## What is the new behavior? Notebook database metadata is resolved independently from existing selector formatting: * an omitted database identifier is Primary * an identifier matching the project ref is Primary * other identifiers show a loading state while databases load * a loaded non-primary match is Replica * an unmatched identifier is Unknown * database lookup failures use a neutral unavailable state Labels are compact: Database: Primary, Database: Replica, and Database: Unknown. The databases query is enabled only when a preview contains an explicit non-primary identifier. ## Additional context Validation: * 39 focused notebook preview tests * Studio typecheck * targeted ESLint * Prettier check * git diff --check <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Notebook previews now identify database targets as primary, read replica, or unknown. * Database metadata is resolved automatically when needed. * Loading states display a clear “Loading database…” indicator. * Database metadata now handles unavailable, hidden, and error states more clearly. * Notebook entries reflect updated database information before and after replacement. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
4dee589735 |
fix(studio): stop assistant fabricating database_identifier for notebooks (#49558)
## Summary * Fixes [FE-4275](https://linear.app/supabase/issue/FE-4275/assistant-always-creates-notebooks-with-wrong-identifier-first-try): the assistant always created database notebook cells with a fabricated `database_identifier` (`"primary"`, later observed as `""` / `"_primary"` under different prompt wording) instead of omitting the key for the project's primary database, which tripped the tool's reject-and-retry validation on the very first attempt. * Prompt wording alone wasn't reliable — live eval runs against the real model kept substituting a new placeholder every time the prompt was tightened further. * Normalizes an empty-string `database_identifier` to absent at the schema level (`databaseIdentifierSchema` in `notebook-schema.ts`), which is inherited by every schema built from it — the AI SDK's `inputSchema` for `create_notebook`/`update_notebook`, and the write-boundary `writableNotebookSchema` used right before the PUT to the backend. * Adds an eval case (`evals/dataset.ts`) reproducing the original bug, plus unit tests covering schema-level and write-boundary normalization. ## Test plan - [X] `pnpm --filter studio exec tsc --noEmit` passes - [X] `pnpm exec prettier --check` passes on touched files - [X] Unit tests pass: `notebook-schema.test.ts`, `notebook-upsert-mutation.test.ts`, `notebook-tools.test.ts` (104 tests) - [X] Ran the new eval case against the real model 3x before the code fix (0% correctness, fabricated `""`/`"_primary"`) and 3x after (100% correctness) ## Summary by CodeRabbit * **Bug Fixes** * Improved notebook handling of empty database identifiers by treating them as absent. * Ensured notebook requests omit unused database identifier fields. * Added validation guidance for read-replica database identifiers. |
||
|
Charis
|
8f124cd2e5 |
chore: gitignore personal skills in agents (#49544)
<!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Chores** * Updated ignore rules to exclude local agent skill files from version control. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
632e099487 |
feat(studio): track assistant notebook divergence (#49539)
## Summary - record server divergence when an assistant changes a notebook with local unsaved edits - clear the session-only marker after a successful save or notebook removal - cover update, delete, saved eviction, and lifecycle behavior ## Verification - pnpm --dir apps/studio exec vitest run state/notebooks/notebooks-state.test.ts lib/ai/notebook-cache-invalidation.test.ts - pnpm --dir apps/studio exec eslint state/notebooks/notebooks-state.ts state/notebooks/notebooks-state.test.ts lib/ai/notebook-cache-invalidation.ts lib/ai/notebook-cache-invalidation.test.ts - pnpm --dir apps/studio typecheck Stacked on the approval-warning PR for FE-4255. <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Bug Fixes** * Improved handling of server changes to notebooks with unsaved local edits. * Server updates and deletions are now tracked as divergences instead of being silently skipped. * Divergence indicators are cleared when changes are saved or notebooks are removed. * Unrelated notebook changes no longer create false conflicts. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
b686a72d89 |
feat(studio): warn on dirty notebook assistant proposals (#49536)
## Summary * Warn before approving assistant update/delete proposals when the notebook has unsaved local changes. * Keep approval available; this is an informed-choice warning. * Add coverage for dirty, saved, and absent local notebook state. Towards [FE-4255](https://linear.app/supabase/issue/FE-4255/assistant-update-notebook-can-silently-overwrite-unsaved-local) <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Added warnings when approving notebook update or deletion proposals that could discard unsaved local changes. * Warning messages now distinguish between update and delete actions. * Update proposal approval remains available after the warning is displayed. * **Bug Fixes** * Prevented unnecessary warnings when notebooks have no saved local changes or are unavailable. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
CharisandJoshen Lim
|
40d9a2e929 |
fix(studio): invalidate notebook cache after assistant delete (#49496)
Stacked on supabase/supabase#49415. ## Summary - collect completed assistant delete_notebook tool calls as deleted cache effects - evict deleted notebooks from the React Query cache and notebook state - cover collector, cache eviction, and the open-tab Notebook not found state ## Verification - pnpm test:studio -- notebook-cache-invalidation ExplorerNotebookTab.assistant-cache-invalidation - pnpm --filter studio typecheck - touched-file ESLint and Prettier checks <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Bug Fixes** * Improved notebook deletion handling so removed notebooks no longer remain visible after deletion. * Navigation and cached notebook data now update promptly when a notebook is deleted. * Attempting to access a deleted notebook now displays a clear “Notebook not found” message. * **Tests** * Added coverage for notebook deletion, cache invalidation, navigation updates, and the resulting not-found state. <!-- end of auto-generated comment: release notes by coderabbit.ai --> --------- Co-authored-by: Joshen Lim <joshenlimek@gmail.com> |
||
|
Charis
|
de3a8799d6 |
fix(studio): invalidate notebook caches after assistant create/update (#49415)
## Summary - The assistant's `create_notebook`/`update_notebook` tools run entirely server-side, so an open notebook tab's React Query cache and Valtio store never learn a write happened — the tab keeps showing stale content until a manual reload. - Adds `collectNotebookCacheEffects`/`applyNotebookCacheEffects` (`apps/studio/lib/ai/notebook-cache-invalidation.ts`), which scan finished assistant messages for completed `create_notebook`/`update_notebook` tool calls and evict the affected notebook via `evictNotebookFromCaches` (`apps/studio/data/content/notebooks/notebook-cache.ts`), plus invalidate the nav list. - Wired into `createChatInstance`'s `onFinish` in `state/ai-assistant-state.tsx`, with per-chat dedupe so replayed history isn't reprocessed. - Removes the cache entry outright rather than invalidating it, since a remounting `useNotebookQuery` would otherwise read the stale cached value synchronously before its refetch lands. - Explicitly skips eviction when the open tab has unsaved local edits, so an assistant write can't silently discard them. Related: [FE-4235](https://linear.app/supabase/issue/FE-4235) **Out of scope:** this only protects the client-side cache/store from being clobbered after the fact. Preventing the assistant's `update_notebook` tool call itself from overwriting a user's unsaved edits (a data-layer conflict, not a cache-freshness one) is tracked separately in [FE-4255](https://linear.app/supabase/issue/FE-4255). ## Test plan - [x] `pnpm test:studio -- notebook-cache notebook-cache-invalidation ai-assistant-state.notebook-cache-invalidation ExplorerNotebookTab.assistant-cache-invalidation ExplorerNotebookTabCoordinator` — all passing - [x] Reproduction-first component test (`ExplorerNotebookTab.assistant-cache-invalidation.test.tsx`) — verified it fails without the fix (stale content persists) and passes with it - [x] Regression test for the dirty-notebook guard (an edited, unsaved notebook is left untouched by an assistant write) - [x] `pnpm typecheck --filter=studio` / `pnpm lint --filter=studio` clean <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Bug Fixes** * Notebook changes made through the AI assistant now appear correctly in open notebook tabs and after reopening them. * Saved notebook caches are refreshed after completed create or update actions, preventing stale content from being displayed. * Unsaved notebook changes are preserved during cache cleanup. * Closing a notebook tab now consistently removes its cached content. * **Tests** * Added coverage for assistant-driven updates, remounts, duplicate actions, project context changes, and cache behavior. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
CharisandJoshen Lim
|
89b4f1aca4 |
feat(studio): add delete_notebook tool to AI assistant (#49413)
## Summary * Adds a `delete_notebook` AI assistant tool (`needsApproval: true`) that lets the assistant delete a notebook with explicit user approval, mirroring the existing `create_notebook`/`update_notebook` tools. * Wires up a destructive-styled approval card in the AI Assistant Panel (fetches the notebook to show its name, warns the deletion is permanent) using the same `Confirm`/tool-approval plumbing as the other notebook tools. * Updates `tool-filter.ts` opt-in gating, the assistant system prompt, the eval-harness mock tools, and the eval dataset with `delete_notebook` coverage. * Adds test coverage in `notebook-tools.test.ts`, `mock-tools.test.ts`, and `NotebookProposalRenderer.test.tsx`. Closes [FE-4242](https://linear.app/supabase/issue/FE-4242/assistant-delete-notebook-tool). ## Test plan - [X] `pnpm typecheck --filter=studio` passes - [X] `pnpm --filter studio exec vitest run` for the touched files (notebook-tools, mock-tools, NotebookProposalRenderer, [Message.Parts](<http://Message.Parts>), and existing consumers of `content-delete-mutation`) — all passing - [X] `eslint` and `prettier --check` clean on all touched files - [X] Manual verification of the approval UI in a running Studio instance (not done in this session) ## Summary by CodeRabbit * **New Features** * Added AI-assisted notebook deletion with explicit confirmation and irreversible-action warnings. * Added safeguards to distinguish deleting an entire notebook from removing individual panels. * Completed deletions now display the deleted notebook’s name without an option to reopen it. * **Bug Fixes** * Improved handling of missing notebooks and invalid deletion requests. * **Tests** * Added coverage for deletion approval, denial, errors, and successful completion. <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Added AI-assisted notebook deletion with explicit approval and irreversible-action warnings. * Added confirmation, loading, error, and completion states for notebook deletion. * Prevented accidental full-notebook deletion when only a panel or section should be removed. * Improved notebook update results by showing applied changes when available. * **Bug Fixes** * Notebook deletion now uses the required API version. * Improved handling and validation of missing notebooks during deletion. <!-- end of auto-generated comment: release notes by coderabbit.ai --> --------- Co-authored-by: Joshen Lim <joshenlimek@gmail.com> |
||
|
CharisandJoshen Lim
|
48cc37f0d2 |
refactor(studio): extract notebook cache eviction helper (#49414)
## Summary Part 1 of the FE-4247 stack ([FE-4247](https://linear.app/supabase/issue/FE-4247/assistant-invalidate-cache-after-notebook-editdeletion)). Pure refactor, no behavior change — extracts the notebook cache eviction logic that `ExplorerNotebookTabCoordinator` had open-coded into a shared helper, so the upcoming assistant create/update/delete cache invalidation (PR 2/3 in the stack) can reuse it instead of duplicating the two-cache-layer eviction dance. - New `evictNotebookFromCaches({ queryClient, projectRef, id, mode })` in `apps/studio/data/content/notebooks/notebook-cache.ts`. `mode: 'refresh' | 'remove'` selects `invalidateQueries` vs `removeQueries` on `contentKeys.resource`. Drops the notebook from `notebooksState` only when its status is `'saved'`, matching the original open-coded guard exactly. Returns whether it evicted, so callers can branch. - `ExplorerNotebookTabCoordinator` now calls the helper with `mode: 'remove'` instead of inlining the logic. ## Test plan - [x] `pnpm test:studio -- notebook-cache ExplorerNotebookTabCoordinator` — new helper tests (refresh/remove/dirty-guard/unknown-id) and existing coordinator tests all pass - [x] `pnpm typecheck --filter=studio` - [x] `pnpm lint --filter=studio` — 0 errors, no new warnings <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit - **Bug Fixes** - Improved detection of unsaved notebook changes for tab indicators and close confirmations. - Empty, never-saved notebooks are no longer included in discard prompts. - Improved cache cleanup when closing saved notebooks while preserving unsaved work. - Added safeguards for missing notebook records. - **Tests** - Added coverage for notebook cache refresh, removal, preservation, and no-op scenarios. <!-- end of auto-generated comment: release notes by coderabbit.ai --> --------- Co-authored-by: Joshen Lim <joshenlimek@gmail.com> |
||
|
Charis
|
233cbdc8e5 |
Restore notebook diff preview for completed updates (#49402)
## Summary This is **PR 3 of 3** in the FE-4243 stack fixing "Notebook update proposal shows 'unapplyable' error for already-completed updates." - Consumes the `previous_content` field added by PR 2 (#49401) to reconstruct diffs for already-applied notebook updates - Restores the diff preview that PR 1 initially dropped — completed updates now show the full before/after instead of a generic "Notebook updated" message - Uses the same diff derivation function called pre-approval, guaranteeing the rendered diff matches what was shown during confirmation - Includes defensive fallback handling for older persisted chats (before `previous_content` existed) and edge cases **Depends on**: PR 2 (#49401) merging first — this PR consumes the `previous_content` field from that server change. Resolves FE-4243 ## Test plan - ✅ 19/19 tests pass in NotebookProposalRenderer.test.tsx (2 confirmed as real regressions) - ✅ 118/118 tests pass in full AIAssistantPanel suite - ✅ Typecheck: clean on modified files - ✅ ESLint: zero errors/warnings on changed files - ✅ Lint ratchet: passes (some rules improved) - ✅ New regression tests cover: delete_cell, insert_cell, missing previous_content, and operations that no longer reconcile - ✅ No notebook fetch in completed update tests (proves no redundant re-fetching) <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Added visual previews showing notebook changes, including inserted and deleted cells, when prior content is available. * Prevented duplicate cells from appearing in update previews. * Retained a compact completion message when change details are unavailable or inconsistent. * Ensured previews are shown only for the relevant notebook. * **Tests** * Added coverage for notebook update previews, deletion and insertion diffs, duplicate prevention, notebook matching, and fallback behavior. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
8920439569 |
Expose previous notebook content in update_notebook (#49401)
## Summary - Plumb pre-update notebook snapshot through `update_notebook` tool response as `previous_content` - Add sanitizers in `tool-sanitizer.ts` to strip snapshot before model sees it - Add client-side stripping in `prepareMessagesForAPI` to avoid re-uploading snapshot on subsequent turns - This is PR 2 of 3 fixing Linear issue FE-4243 (notebook update proposal shows 'unapplyable' error for already-completed updates) - Ships no visible behavior change on its own; enables PR 3 to restore diff preview for completed updates ## Test plan - [x] Unit tests: 80/80 passing across notebook-tools.test.ts, tool-sanitizer.test.ts, generate-assistant-response.utils.test.ts, message-utils.test.ts, and mock-tools.test.ts - [x] Typecheck: clean for all changed files - [x] ESLint: zero errors, lint:ratchet passes (exit 0) - [x] Integration: previous_content is correctly populated with pre-update notebook, stripped before model context, and stripped on client-side re-upload <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit - **Bug Fixes** - Notebook updates now retain previous content for recovery and history. - AI responses expose only the notebook’s ID and name, keeping previous content out of model-visible data. - **Tests** - Added coverage for notebook update results, content sanitization, and message preparation, including cases where previous content is absent or preserved. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
c172195269 |
test(studio): add eval cases for list_databases-driven notebook creation (#49398)
## Summary - Adds three eval cases exercising the behavior this stack wires up: a happy path where the model calls `list_databases` before targeting a named read replica, a guard against fabricating an identifier when the user names a region/replica `list_databases` doesn't actually return, and a guard against targeting a non-primary database when the user never asked for one. Part 6/6 (final) of the stack for FE-4225 (expose valid database identifiers to the notebook AI agent). Stacked on #49334. ## Test plan - [x] Ran all three cases against the real model; inspected transcripts directly - [x] Re-verified reworded `correctAnswer` text against real outputs via the correctness evaluator - [x] `pnpm --filter studio exec tsc --noEmit` passes - [x] `prettier --check` passes <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Tests** * Added evaluation coverage for selecting the correct database replica when creating notebooks. * Verified primary-database defaults when no database is specified. * Added checks to prevent fabricated database identifiers when a requested replica is unavailable. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
27ff49c145 |
Stop re-deriving notebook update diffs after completion (#49399)
## Summary
- Fixes false-negative "This update can't be applied" warning for
completed notebook updates (FE-4243)
- When `state='output-available'` (tool completed), skips notebook fetch
and diff derivation
- Renders compact "Notebook updated: {name}" instead of a phantom/failed
diff
- `UnapplyableNotebookUpdateNotice` now accepts and forwards
`footerAction` prop, preserving "Open notebook" link
- Different warning copy for terminal confirm states
(success/error/denied): "Preview unavailable / notebook has changed"
instead of "can't be applied"
- Preserves diff derivation for non-completed states
(output-denied/error)
This is PR 1 of a 3-PR stack; PRs 2-3 restore the full diff preview for
completed updates (requires server snapshot).
Towards FE-4243
## Test plan
- [x] All 15 tests in NotebookProposalRenderer.test.tsx pass
- [x] Typecheck clean
- [x] ESLint clean
- [x] Regression tests added: completed updates with missing target
cells (auto & manual approval)
- [x] Confirms denied/error states still derive against live content
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
- **New Features**
- Added clearer status handling for AI-generated notebook updates.
- Completed updates now show a confirmation with the notebook name when
available.
- Update actions and footer controls now adapt to the proposal’s current
state.
- **Bug Fixes**
- Improved messaging when notebook changes prevent an update preview
from being reconstructed.
- Preserved accurate previews for denied or failed updates using the
notebook’s latest content.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
|
||
|
Charis
|
dd534d229b |
docs(studio): instruct the notebook agent to use list_databases (#49334)
## Summary - Adds a bullet to `NOTEBOOKS_PROMPT` instructing the assistant to call `list_databases` before setting a `database_cell`'s `database_identifier`, mirroring the existing `list_tables` schema-validation instruction immediately above it. Part 5/6 of the stack for FE-4225 (expose valid database identifiers to the notebook AI agent). Stacked on #49333. This is the last piece that makes the assistant actually *use* the tool and schema field wired up earlier in the stack, rather than just having them available. ## Test plan - [x] `pnpm --filter studio exec tsc --noEmit` passes - [x] `prettier --check` passes <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Bug Fixes** * Improved notebook database configuration by ensuring database identifiers are selected from available databases. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
ee59d28240 |
fix(studio): reject unknown database_identifier before writing a notebook (#49333)
## Summary - `create_notebook`/`update_notebook` now validate every `database_cell`'s `database_identifier` against the project's real database list (`getReadReplicas`) before writing, throwing an assistant-actionable `NotebookToolError` (same pattern as the existing `expected_updated_at` mismatch check) when it doesn't match. - Only fetches the database list when a cell actually sets `database_identifier` — no added cost for the common case. - `update_notebook` validates only the cells its own operations introduce (`insert_cell`/`replace_cell`), not the whole resulting notebook — otherwise an unrelated, untouched pre-existing cell whose replica was removed after the fact would block updates that never touch it. Part 4/6 of the stack for FE-4225 (expose valid database identifiers to the notebook AI agent). Stacked on #49332. This closes the gap that PR 3 reopened: a model can no longer invent an identifier that silently breaks a cell — it now gets a retryable error naming `list_databases` (added in #49328) as the way to find a real one. ## Test plan - [x] `create_notebook`/`update_notebook` reject an unknown `database_identifier` with `NotebookToolError` + `exposeToAssistant: true` - [x] Both succeed when the identifier matches a real database - [x] `create_notebook` never calls the databases endpoint when no cell sets `database_identifier` - [x] `update_notebook` succeeds without validating or fetching databases when no operation introduces a database_cell, even if an untouched existing cell carries a now-invalid identifier - [x] `pnpm --filter studio exec tsc --noEmit` passes <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Added a database-listing tool that provides database identifiers and metadata, including primary-database status. * Database results now return only the relevant fields. * **Bug Fixes** * Added validation to prevent notebooks from referencing unknown databases. * Create and update actions now provide clear errors for invalid database identifiers. * Updates validate newly added or replaced cells while preserving existing, untouched cells. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
7b04fc7d09 |
revert(studio): reinstate database_identifier on the agent notebook schema (#49332)
## Summary - Reverts #49326's temporary mitigation, which stripped `database_identifier` from the agent-facing notebook cell schema (`agentCellSchema`) because the assistant had no legitimate source of truth for valid read-replica identifiers. - The previous PR in this stack (#49328) added the `list_databases` tool, so that source of truth now exists — `database_cell`s can carry `database_identifier` again. Part 3/6 of the stack for FE-4225 (expose valid database identifiers to the notebook AI agent). Stacked on #49328. ## Test plan - [x] Existing schema/preview tests (reverted alongside the mitigation) pass - [x] `pnpm --filter studio exec tsc --noEmit` passes <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * AI-generated notebook database cells now support database identifiers. * Notebook previews display the associated database source, including transitions between primary and replica databases. * **Bug Fixes** * Improved database metadata handling to preserve identifiers when updating notebook cells. * Database information is now shown only when available, preventing inaccurate or missing metadata displays. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
144c2eadb6 |
fix(studio): expose notebook diff validation errors for auto-retry (#49331)
## Summary - Automatically deny notebook tool proposals with specific error reasons when client-side diff validation fails (e.g., unknown cell ID) - Enables the AI Assistant to see the actual failure reason and retry automatically instead of requiring manual user intervention - Exposes the same `describeNotebookOperationError` helper used server-side for consistent error messaging - Adds `denyWithReason()` to manual tool approval handlers for flexible denial messaging ## Test plan - Run `pnpm --filter studio exec vitest run apps/studio/components/ui/AIAssistantPanel/Confirm.utils.test.ts` to verify denyWithReason tests - Run `pnpm --filter studio exec vitest run apps/studio/components/ui/AIAssistantPanel/NotebookProposalRenderer.test.tsx` to verify auto-deny behavior, Skip fallback, and no re-fire after approval is already handled - Confirm no regressions in existing notebook tool approval flows ## Manual testing - Get assistant to create a notebook. - Open the notebook and manually delete a cell yourself. - Ask the assistant to delete the cell you just deleted. - Assistant should automatically recover from the error. <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit ## Summary by CodeRabbit * **New Features** * Notebook proposals now display clear success, error, and denial outcomes. * Tool errors for SQL, Edge Functions, and notebooks now appear in their respective result views. * Output links are supported in notebook proposal results. * Approval panels remain visible after completed actions with standardized status messages. * **Bug Fixes** * Specific denial reasons are preserved instead of showing a generic skipped message. * Unapplyable notebook updates are automatically denied with an explanation. * Prevented duplicate denial responses after approval decisions. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
0f730de9e0 |
feat(studio): add list_databases tool for the AI assistant (#49328)
## Summary
- Adds a `list_databases` tool to the notebook AI tools
(`getNotebookTools`), returning `{ identifier, is_primary, region,
status }` for the project's primary and any read replicas, via
`getReadReplicas`.
- Registers `list_databases` in `tool-filter.ts` (opt-in validation
schema + SCHEMA category), and adds a mock fixture in `mock-tools.ts`
for evals.
Part 2/6 of the stack for FE-4225 (expose valid database identifiers to
the notebook AI agent). Stacked on #49327. A later PR in the stack
reinstates `database_identifier` on the agent-facing notebook schema and
requires the agent to call this tool first.
## Test plan
- [x] New test coverage in `notebook-tools.test.ts` for
`list_databases`, including `is_primary` computation
- [x] Existing exact-tool-set assertion updated
- [x] `pnpm --filter studio exec tsc --noEmit` passes
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **New Features**
* Added an AI notebook tool for listing a project’s databases.
* Results include each database’s identifier, region, status, and
whether it is the primary database.
* Supports projects with read replicas.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
|
||
|
Charis
|
7a2e237892 |
Add update_notebook evals; fix prompt gaps they surfaced (#49324)
## Summary - Add `update_notebook` eval cases (insert/replace/delete/move, a combined delete+insert, and guard/safety cases) mirroring the existing `create_notebook` cases, targeting the notebooks already seeded in the mock tool harness. - Fix two behavior gaps in `NOTEBOOKS_PROMPT`/`LIMITATIONS_PROMPT` that these cases surfaced when run live: the assistant asking the user for a notebook id instead of resolving it via `list_notebooks`, and the destructive-operations warning rule not being connected to SQL written into notebook cells. - Soften the destructive-SQL case's `correctAnswer` to match `update_notebook`'s real approval-gated behavior — a warning accompanying the reported change is acceptable, not only one strictly preceding the tool call. ## Test plan - [x] `pnpm run typecheck` (apps/studio) — clean - [x] `pnpm exec prettier --check` on both changed files — clean - [x] `evals/scorer.test.ts`, `evals/transcript.test.ts`, `evals/trace-utils.test.ts` — 21/21 pass - [x] Ran the new eval cases live against OpenAI (bypassing the Braintrust proxy) via Braintrust MCP; confirmed via trace inspection that the prompt fix resolved the id-resolution gap (Tool Usage 0% → 100% across 3 trials) and that the assistant now includes an explicit irreversibility warning when destructive SQL is written into a notebook cell <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit - **New Features** - Improved notebook creation and editing support across SQL, query, chart, and time-range cells. - Added clearer handling for saved notebooks, recurring requests, and one-time SQL execution. - Enhanced validation for database cells and notebook configuration. - **Bug Fixes** - Improved safeguards and warnings for destructive queries, including saved notebook queries. - Better handling of missing tables and notebooks. - More precise notebook cell updates and tool usage validation. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
717927f4f2 |
fix(studio): AI assistant notebooks no longer set an invalid database_identifier (#49326)
## Summary - The AI assistant's `create_notebook`/`update_notebook` tools could set a `database_cell`'s `database_identifier` to a value that doesn't correspond to any real database, because no tool exposes a project's actual read-replica identifiers to the model. - An unresolvable `database_identifier` silently breaks the cell: `QueryEditor`'s connection-string lookup fails to find a match, and running the cell fails with `Unable to run query: Connection string is missing` — even though the exact same SQL runs fine when pasted into a manually-created cell (which never sets this field). - Fix: strip `database_identifier` from the agent-facing schema (`agentCellSchema` in `notebook-schema.ts`) entirely, so the model can no longer emit it at all. **This is a temporary fix** until we wire in real read-replica support for the AI assistant (e.g. a tool exposing a project's valid replica identifiers) — the field can be reintroduced once the model has a legitimate source of truth to pull a valid identifier from. - Updated tests that relied on agent cells carrying `database_identifier` to reflect the new behavior, and added a regression test asserting `agentNotebookSchema` rejects a `database_cell` with that field set. Resolves FE-4224 ## Test plan - [x] `notebook-schema.test.ts`, `notebook-operations.test.ts`, `notebook-tools.test.ts`, `AssistantNotebookPreview.test.tsx`, `AssistantNotebookPreview.utils.test.ts` all pass - [x] `tsc --noEmit` clean - [x] Prettier clean on touched files <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Bug Fixes** * Improved handling of database notebook cells when database metadata is unavailable. * Cells without database identifiers now display “No metadata” instead of an incorrect replica identifier. * Prevented invalid database identifiers from being accepted in agent-generated notebook content. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
ebd616fa90 |
fix(studio): auto-retry notebook updates on stale/invalid conflicts (#49323)
## Summary
- **Removed dead client-side refresh UI** in
`NotebookProposalRenderer.tsx` and its test — the diff preview is always
computed from live data, so the check was redundant with the tool's
server-side re-validation
- **Added typed `NotebookToolError`** in `notebook-tools.ts` with
structured metadata (`{ exposeToAssistant: boolean }`) validated by a
zod schema with a literal discriminant tag (`tag:
'notebook_tool_error'`) — tracks the two retryable failures: staleness
conflict and invalid operations (unknown cell id)
- **Encoded errors in `generate-v4.ts` onError** — the one place in the
pipeline that holds the live `Error` before it becomes a string in the
persisted message
- **Extracted and fixed message history filter** into new
`generate-assistant-response.utils.ts` — any tool-error whose
`errorText` decodes against the `NotebookToolError` schema is let
through (with `errorText` rewritten to plain prose so the model sees the
message, not JSON), while other errors stay filtered as before
Net effect: the assistant detects the specific, actionable rejection
reason and retries on its own with no dead button or human intervention
needed.
## Test plan
- Existing unit tests in `NotebookProposalRenderer.test.tsx` pass (dead
button test removed)
- New unit tests in `notebook-tools.test.ts` cover encode/decode
round-trips and error discrimination
- New unit tests in `generate-assistant-response.utils.test.ts` cover
message history filtering with all error states
- `pnpm typecheck` is clean
- `pnpm --filter studio run lint:ratchet` passes (no new ESLint
warnings)
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **New Features**
* Notebook update errors now provide clearer, structured explanations to
the AI assistant.
* Assistant responses preserve relevant notebook error details while
filtering invalid or temporary tool states.
* **Bug Fixes**
* Improved handling of stale notebook revisions and invalid notebook
update operations.
* Notebook proposal rendering proceeds without an unnecessary refresh
step.
* **Tests**
* Expanded coverage for notebook errors, message filtering,
serialization, and error handling.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
|
||
|
Charis
|
e7c3cad8de |
feat(evals): add notebook eval cases and forbiddenTools scorer capability (#49104)
## Summary - Added ~11 new eval cases for Notebooks AI assistant evals, including: basic notebook cell creation, multi-cell composition (markdown+database+log), log cell time ranges, row_limit defaults, chart config, destructive SQL safety warnings, and hallucination guards for nonexistent tables - Extended `toolUsageScorer` with deterministic `forbiddenTools` field to score both required and forbidden tool usage, enabling eval cases to assert tool choice (e.g., `execute_sql` vs `create_notebook`) without LLM-as-judge - Extended SQL validators (`sqlSyntaxScorer`/`sqlIdentifierQuotingScorer`) to validate SQL inside `create_notebook` database cells (log cells deliberately excluded as they use ClickHouse dialect) - Fixed two real assistant issues in `NOTEBOOKS_PROMPT`: (a) reuse `CLICKHOUSE_LOGS_COMPLETION_INSTRUCTIONS` and schema section to prevent incorrect BigQuery-style SQL in logs queries, (b) require schema verification before writing `database_cell` to prevent queries against nonexistent tables Resolves FE-4087 ## Test plan - All 11 new eval cases run live against OpenAI via Braintrust; traces inspected and validated - Existing unit tests, lint, and typecheck pass <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Added support for creating notebooks with database, Markdown, chart, and log cells. * Improved handling of reusable notebook requests versus one-off SQL queries. * Added guidance for modern ClickHouse SQL and absolute log time ranges. * **Bug Fixes** * Improved SQL validation, row-limit enforcement, and destructive-query safety. * Prevented invalid or nonexistent-table queries from being accepted. * Improved validation of notebook cell types and tool usage. * Improved handling of ClickHouse log queries and database-cell SQL. * Improved evaluation reliability by limiting concurrent test execution. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
474bf5da4a |
fix(studio): reset rename form between same-named SQL snippets (#49275)
## 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? Bug fix ## What is the current behavior? In the SQL Editor, when two snippets are both named "Untitled query" and one is renamed, the rename modal's state is not reset afterwards. Opening the rename modal for the second snippet prefills the input with the first snippet's new name, and the second snippet can't be renamed at all because the "Rename query" button stays disabled. `RenameQueryModal` fed the snippet to react-hook-form through the `values` option, which only re-runs its reset when the values object deep-changes. Two snippets with the same name (and no description) produce a deep-equal object, so switching between them never resets the form — it keeps the previously renamed name and stays non-dirty. ## What is the new behavior? The form is mounted per snippet (`key={snippet.id}`) with plain `defaultValues`, so no form state can carry over between snippets regardless of name collisions. `SQLEditorNav` derives modal visibility from the selected snippet and clears it on cancel/complete, matching `SearchList`. Covered by a new component test in `RenameQueryModal.test.tsx` that renames one "Untitled query", reopens the modal for a second one, and asserts the field resets and the second rename submits. ## Additional context <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Bug Fixes** * Fixed the rename dialog retaining input from a previously renamed snippet. * Ensured the rename form resets correctly after successful submission and when switching between snippets. * **Tests** * Added regression coverage for renaming multiple untitled snippets with the same original name. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
CharisandJoshen Lim
|
8bdfe03fe7 |
refactor(studio): drop notebook type widening now that the API supports it (#49272)
## Summary - Regenerates `packages/api-types` for the content endpoints now that the Platform API's `notebook` content type has landed (list/get/upsert `type` enums, plus `UpsertContentBody`'s notebook cell shape with `_id`/`y_series`). Unrelated schema drift from the same regen (Warehouse, SSO, notification exceptions, etc.) is excluded — only the content-endpoint hunks are applied. - Removes every local widening cast added while the API support was pending (`content-query.ts`, `content-infinite-query.ts`, `notebook-query.ts`, `notebook-upsert-mutation.ts`, `sql-folders-query.ts`). - What remains is scoped and renamed to match: draft ids (`generateDraftId`/`isDraftId`), used only for cells created client-side in the editor before their first save, dropped before they'd ever reach the backend as a fake `_id`. ## Test plan - [x] `pnpm typecheck` — clean - [x] `pnpm --filter studio test` — full suite passes (518 files / 5471 tests) - [x] `pnpm --filter studio run lint:ratchet` — no new warnings - [x] `pnpm format` / prettier — clean <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Bug Fixes** * Improved notebook cell tracking during editing, reordering, insertion, and deletion. * Preserved existing cell identifiers while removing temporary draft identifiers before saving. * Improved chart configuration for selecting and displaying multiple Y-axis series. * Strengthened notebook validation and content persistence behavior. <!-- end of auto-generated comment: release notes by coderabbit.ai --> --------- Co-authored-by: Joshen Lim <joshenlimek@gmail.com> |
||
|
Charis
|
4343e21da0 |
feat(studio): tighten the notebook diff preview (#49218)
## 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? UI refactor of the notebook create/update preview in the AI Assistant panel, plus a small additive prop on the shared `CodeBlock`. ## What is the current behavior? The assistant's notebook diff renders each cell as its own bordered box with a gap between them, under a `6 cells` line that is easy to miss. Cells can't be collapsed, each one carries a repeated `ADDED` badge and a nested "Show more" toggle, and long markdown scrolls sideways instead of wrapping. ## What is the new behavior? <img width="796" height="1076" alt="CleanShot 2026-08-18 at 14 41 30@2x" src="https://github.com/user-attachments/assets/45e58c6c-48f2-404b-8699-757ee96a4a8d" /> - The whole diff is one card: a distinct header row (notebook name, summary, expand/collapse all) over cells glued together by dividers. - Every cell is a `Collapsible`. Added and replaced cells open by default; unchanged, moved, and removed cells stay as single rows but are now inspectable instead of being content-free. - The per-row badge is replaced by a colored gutter glyph (`+` `−` `~` `↕`) with a tooltip naming the change type. The change type reaches the accessible name via `aria-label` on the row. - The nested "Show more" toggle inside each cell is gone — the row itself is the only control. - `CodeBlock` gains a `wrapLongLines` prop (default `false`, no change for existing callers), used here so markdown and SQL soft-wrap. The highlighter sets `white-space` inline on the `<code>` element, so a class on the `<pre>` can't do this. ## Additional context Towards FE-4143 <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Notebook previews now display titles, notebook icons, and clearer bordered layouts. * Added per-cell expand/collapse controls, including “Expand all” and “Collapse all.” * Long code lines can now wrap for improved readability. * **Improvements** * Added mode-based fallback labels when notebook titles are unavailable. * Newly added and replaced cells expand by default, while unchanged cells remain collapsed. * Improved change markers, tooltips, removed-cell styling, and notebook proposal preview spacing. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
8c409e2df5 |
Fix eval scorer truncation via local transcript capture (#49151)
## Problem Scorers previously derived the assistant's final answer via Braintrust's `trace.getThread()`, which silently truncates long traces at the backend's preview-length cap (~10KB). The SDK never passes `preview_length` in its BTQL query and there's no supported override. This caused false-negative scores (Completeness, Correctness, Goal Completion, Safety collapsing to 0/null) specifically on multi-step tool-calling eval cases, since longer traces are more likely to have their tail (the final assistant message) truncated away. ## Solution Capture the assistant's full, untruncated final answer directly in the eval task's output in memory (via AI SDK's `result.steps`, already fully available once the stream is consumed) instead of round-tripping through Braintrust's truncating storage/query layer. Scorers now read `output.transcript` instead of calling `trace.getThread()`. ## Changes - **New**: `apps/studio/evals/transcript.ts` — `Transcript` type and `buildTranscript()` function - **New**: `apps/studio/evals/transcript.test.ts` — unit tests (5 passing) - **Modified**: `apps/studio/evals/assistant.eval.ts` — captures `result.steps` and returns transcript - **Modified**: `apps/studio/evals/scorer.ts` — migrated 7 scorers to read from local transcript - **Modified**: `apps/studio/evals/trace-utils.ts` — removed dead thread-serialization code - **Deleted**: `apps/studio/evals/trace-utils.test.ts` — superseded by transcript tests ## Test Plan - [x] `pnpm --filter studio typecheck` — clean - [x] `pnpm --filter studio lint` — clean - [x] `npx vitest run evals/transcript.test.ts` — 5/5 passing - [x] Full live eval run (35/35 cases) against Braintrust — [experiment](https://www.braintrust.dev/app/supabase.io/p/Assistant/experiments/eval-scorer-transcript-capture-1786985352) shows Completeness/Correctness/Goal Completion/Safety scores comparable to baseline ## Known Residual Risk Other scorers that derive data from `trace.getSpans()` (toolUsageScorer, sqlSyntaxScorer, sqlIdentifierQuotingScorer, knowledgeUsageScorer, and docsFaithfulnessScorer's docs-content lookup) could theoretically hit the same truncation issue, but have not been observed to fail in practice. This is not addressed in this PR. <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Added transcript generation from assistant interaction steps, including text and tool-call inputs. * Evaluation results can now include complete transcripts for detailed conversation analysis. * Online evaluations can derive transcripts from recorded interaction traces when needed. * **Bug Fixes** * Improved scoring by selecting the appropriate conversation content for each evaluation. * Ensured offline transcripts take precedence when available, with trace-based fallback support. * **Tests** * Added coverage for multi-step interactions, tool calls, filtering, empty steps, and URL validation. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
79fbe467ba |
feat(studio): wire notebook create/update proposals into assistant panel (#49159)
## Summary PR 4 of the notebook-approval-preview stack. - Adds `NotebookProposalRenderer`, wiring `create_notebook`/`update_notebook` into `MessagePartSwitcher` and rendering `NotebookPreview` across all 6 tool states (drafting, approval-requested, approval-responded, output-available, output-denied, output-error). - `update_notebook` fetches the live notebook via `useNotebookQuery`, checks `expected_updated_at` against the fetched `updated_at`, and gates the confirm action behind a refresh when stale. - A tool-input parse failure renders a raw-input admonition instead of returning `null`, so `ConfirmFooter` — and the ability to Skip/deny — stays available rather than leaving the chat stuck. Towards FE-4143 ## Test plan - [x] `tsc --noEmit` clean - [x] `eslint` clean on touched files - [x] `prettier --check` clean - [x] New `NotebookProposalRenderer.test.tsx` (create/update previews + approve, version-mismatch warning, parse-failure fallback with working Skip, output-available/output-denied summaries) - [x] Existing notebook test suites (`notebook-tools.test.ts`, `notebook-operations`, `NotebookPreview`) still pass <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Added AI-assisted notebook creation and updating with previews, approval controls, and operation summaries. * Added clear handling for loading, errors, denied actions, stale notebook versions, and invalid proposals. * Added links to open notebooks after successful creation or updates. * Preserved notebook SQL content when displaying proposed changes. * **Bug Fixes** * Improved notebook proposal handling for conflicts and incomplete tool responses. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
071a07118a |
feat(studio): add NotebookPreview diff component (#49112)
## Summary Stacked on #49109 (PR 1 — `deriveNotebookDiff`). This is PR 3 of the notebook approval-preview stack: a pure presentational component that renders the cell-level diff for a proposed notebook create/update, for use in the assistant approval UI (wired in a later PR). - `NotebookPreview` — header summary (`"6 cells"` for create, `"+2 −1 ~1 ↕1"` for update) + entry list + "Show N more cells" for long notebooks. - `NotebookPreviewCell` — dispatches per entry tag: `unchanged`/`removed`/`moved` collapse to a muted badge row; `added` renders source via `CodeBlock` (with a max-height/expand toggle); `replaced` renders a `DiffEditor` diff, plus a before → after metadata line when only `database_identifier`/`time_range` changed (SQL/text identical). - `NotebookPreview.utils` — pure helpers (labels, source/metadata extraction, language mapping, summary formatting), unit tested. - **Safety property**: cell content only ever renders through `CodeBlock`/`DiffEditor` (literal source), never through a markdown renderer — agent-authored text can't trigger image loads or link navigation before the user approves. Covered by an adversarial test (``, `[y](evil)`, `<img onerror>` → zero `img`/`[href]`/`[src]` DOM nodes). - Adds `'markdown'` as a supported `CodeBlock` language (small, additive change to `packages/ui-patterns`). Towards FE-4143 ## Test plan - [x] `pnpm --filter studio test` — NotebookPreview suite (21 tests) passes - [x] `pnpm --filter studio exec eslint components/interfaces/Explorer/NotebookPreview` — clean - [x] `pnpm --filter studio exec tsc --noEmit` — no new errors - [x] `pnpm exec prettier --check` — clean <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Added notebook previews showing create and update summaries. * Displayed added, removed, moved, unchanged, and replaced cells with metadata and source diffs. * Added expandable previews with truncation and a “Show more cells” option. * Added Markdown syntax highlighting to code blocks. * **Bug Fixes** * Safely render adversarial agent-authored Markdown as literal content. * **Tests** * Added comprehensive coverage for notebook previews, summaries, metadata, formatting, and truncation. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
2e68f2bf6e |
refactor(studio): derive notebook diff entries (#49109)
## 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? Refactor, plus one bug fix. Groundwork for showing the user a preview of what they are approving when the AI Assistant creates or edits a notebook. No UI in this PR. Towards FE-4143 ## What is the current behavior? `applyNotebookOperations` resolves an ordered list of notebook operations into the resulting cells and nothing else. Rendering a diff for the approval gate needs to know *what happened* to each cell position, not just where things landed, so there is no way to build the preview on top of it. Separately, replacing a cell dropped its id, so `[replace cell-2, insert after cell-2]` failed with a spurious `unknown_cell_id`. ## What is the new behavior? `deriveNotebookDiff` resolves operations into one annotated entry per cell position (`unchanged`, `added`, `removed`, `replaced`, `moved`). `applyNotebookOperations` becomes a thin projection over its result, so there is a single interpreter of notebook operations and the diff a user approves cannot disagree with the cells that get written. The pre-existing tests pass untouched, which is the evidence that the projection is faithful. Notes on the annotations: - `removed` entries stay in the position the cell used to hold so the list reads as a diff. This does not perturb insert-anchor arithmetic: prior inserts still sit contiguously after their anchor. - Moves that cancel out are downgraded to `unchanged`, since two moves can anchor on each other and leave every cell where it started. Badging those as moved would make the preview lie. - `fromIndex` is the cell's position in the original notebook rather than in the shifted working order, so `was #3` means what a reader expects. A replaced cell now stays addressable as an anchor. Anchoring and targeting are separate lookups: a replaced cell can be anchored on, but is never a legitimate target. <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Notebook changes now provide a structured view of added, removed, replaced, moved, and unchanged cells. * Replaced cells can be used as insertion anchors, while invalid or duplicate targets are rejected. * No-op moves are handled as unchanged cells. * Notebook edits preserve operation ordering and original cell positions for more predictable results. * **Bug Fixes** * Improved notebook operation handling and error reporting for complex cell edits. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
9be60cab63 |
refactor(studio): add optimistic locking to update_notebook (#49111)
## 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? Refactor / hardening ## What is the current behavior? The `update_notebook` AI tool re-fetches the notebook right before applying operations, but concurrent edits are last-write-wins: the model has no way to detect that the notebook changed since it planned the edit, so a stale diff can silently overwrite someone else's changes. ## What is the new behavior? - `get_notebook` now returns the notebook's `updated_at` timestamp. - `update_notebook` requires a new `expected_updated_at` input field (the `updated_at` the model got from `get_notebook`). - At execute time, after the existing re-fetch and before applying operations, `update_notebook` compares the fetched `updated_at` against `expected_updated_at` and throws a descriptive error if they don't match, telling the model to re-read the notebook and reissue the update. - The notebook system prompt and mock tools (used by the eval harness) are updated to match. ## Additional context <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Notebook retrieval now includes the latest update timestamp. * Notebook edits require confirmation that the content is current before saving. * **Bug Fixes** * Prevented stale edits from overwriting newer notebook changes. * Conflicting updates are rejected, allowing the latest content to be fetched before retrying. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
628473b3eb |
refactor(studio): extract notebook query-cell logic and give log cells display settings (#49075)
Final PR of the stack. #49069, #49070, #49072 and #49074 have merged, so this now targets `master` directly. **Rebased onto latest `master`**, which includes the centralized result-rendering work (#49096). See "Conflict resolution" below. ## What's left after master's own fixes `QueryCell` was written for database cells and adapted to log cells afterwards. Master has since fixed most of it directly: `handleUpdateCell` no longer bails on a non-database cell, the cell's own binding is read via `getQuerySourceBinding`, and `database_identifier` / `time_range` propagate across a source change. What remains: - **`display` was only passed for database cells**, so the `view` field on `log_cell` stayed unreachable and a logs query could never be charted. That is the one behavioral fix left in this PR. - The per-backend branching is inline and untested. ## What changed Per-backend logic moves into `QueryCell.utils.ts`, where it is unit-tested: `changeCellSource`, `setCellSql`, `cloneQueryCell`, `getCellDisplay`, `toQueryModel`. Each narrows on the cell tag exactly once, so the SQL brand and the backend's parameters stay correlated rather than being re-derived at each call site. `cloneQueryCell` also rebuilds the chart's series array, which valtio hands over as `readonly string[]`. `NotebookEditor` renders through `isQueryCell` (#49069) rather than a tag switch, so a new backend gets picked up by classifying it in `CELL_KINDS` instead of by remembering to add a `case`. ## Conflict resolution Two rounds of master's work landed in this file set. **`QueryCell/index.tsx` (master's own rework).** `changeCellSource` **subsumes the four source-change branches** master had inline, each covered by a test: | Master's branch | Test | |---|---| | database → database (replica change) | `keeps the query when only the database changes` | | logs → logs (time-range change) | `keeps the query when only the log time range changes` | | database → logs | `carries the query text over when moving from the database to logs` | | logs → database | `carries the query text over and restores a default row limit …` | Two improvements fall out of consolidating them: - A **logs → database** move now keeps the selected replica; pinned by `applies the selected database when moving from logs to the database`. - The row-limit default is **named** rather than a hard-coded `100`. `Explorer/utils.ts` now shares `DEFAULT_CELL_ROW_LIMIT` with `createQueryCellSkeleton`, so cell creation and backend conversion can't drift. Untouched from master: `snap.updateCell`, `AddCellDropdown`, `MoveCellDropdownContent`, the `SortableSection` grip props, and `NotebookEditor`'s add-cell buttons, skeletons, `reorderCells` and `insertCellAfter`. **Centralized result rendering (#49096).** That PR moved `QueryCell/QueryResultChart.tsx` up to `Explorer/`, split `QueryResultTable` into `QueryResultError`, and added `QueryResultRenderer`. Since this PR removes `QueryChartConfig`, the type swap had to follow the move and also reach `QueryResultRenderer`, which is new and referenced the removed type. `QueryResultRenderer`, `QueryResultError` and `DataGridResults` are otherwise untouched — the empty/error-state centralization is fully preserved, and `QueryEditor` still renders through it. ## Behavior worth a second opinion `changeCellSource` **carries the query text across a backend change** and rebrands it. This is probably not what a user wants — Postgres SQL and logs SQL are separate dialects over separate schemas, so a carried-over query will usually fail to run, and the rebrand asserts a dialect the text was never written in. Keeping it for now because it destroys nothing and needs no confirmation prompt. The tradeoff is written up at the function. Worth revisiting once we know whether people switch source to port an existing query or to start a fresh one — if it's the latter, clearing the body behind a confirmation is the better answer. Results *are* dropped on a backend change, since another engine returns unrelated columns. ## Incidental `Explorer/types.ts` drops `QueryChartConfig`, which duplicated the wire schema's `ChartConfig` field for field. `chart` stays persisted alongside `view`, so switching to the table and back returns the user's chart settings rather than rebuilding them. ## Verification Typecheck, Prettier, and the lint ratchet clean. 1013 tests pass across `state/`, the Explorer surfaces, notebooks, query sources, `data/sql`, the SQL editor, and `components/ui`; 13 of them are new coverage for the extracted helpers. <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit - **New Features** - Improved notebook cell rendering with more consistent handling of query and markdown cells. - Query cells now preserve SQL, source settings, display preferences, chart configuration, and query results when edited or switched between sources. - Added a default limit of 100 rows for applicable database queries. - **Bug Fixes** - Prevented stale query results from carrying over when changing query sources. - Improved chart configuration consistency across query results and display settings. - **Tests** - Added comprehensive coverage for query-cell updates, source transitions, SQL changes, display state, and chart data. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
b5bfc70c06 |
fix(studio): correlate a query's SQL brand with the backend it runs against (#49074)
Fourth of the stack; PRs 1–3 (#49069, #49070, #49072) have merged, so this now targets `master` directly. **Rebased onto latest `master`.** See "Conflict resolution" at the bottom for what was reconciled. ## The bug `QueryEditor` took `sql: string`, so a query's dialect brand died at the prop boundary and the component re-branded whatever it was handed based on a separately-passed `source`. Nothing tied the two together, which meant nothing stopped Postgres SQL from reaching the analytics endpoint. Explorer query drafts made it concrete. `explorer-query.ts` branded **every** draft with `untrustedSql` regardless of source: ```ts uncheckedSql: untrustedSql(sql) // even for a logs draft ``` and the editor then re-branded that same text with `untrustedLogSql` at run time for a logs draft — laundering a Postgres-branded value straight through the boundary that `safe-analytics-sql.ts` exists to defend. The brands are deliberately disjoint precisely so this can't happen; passing plain strings around defeated it. ## The fix Both carriers are now tagged by backend, so one `_tag` check narrows the SQL brand and that backend's parameters together. - **`ExplorerQueryDraft`** becomes `DatabaseQueryDraft | LogsQueryDraft`, and `toDraft` is the single place a persisted string re-enters the type system — branded for the backend its binding names. The draft is rebuilt rather than mutated in place, since a backend change changes which brand its SQL carries. - **`QueryEditor`** takes one discriminated `query` prop instead of `sql` + `source` + `rowLimit`. The tag picks both the brander at the editor boundary and the execution endpoint, so the mismatch is no longer expressible. - The two `acceptUntrusted*` promotions stay **inlined** in the run handler rather than factored into a shared helper, so each stays visible next to the user gesture that authorizes it, per the safe-SQL model. - **`rowLimit` moves onto the database member.** Logs execution has no use for it — `applyAutoLimit` is Postgres-specific — so it no longer sits on a shared type where it reads as meaningful for both. ## Local storage Existing query drafts shape-mismatch and fall back to a database binding via the existing `safeParse` guard — harmless, and notebooks are still behind the `explorer` flag so there is no saved server content in play. ## Conflict resolution `master` moved inside every file this PR touches. The type change is applied on top of that work; nothing was reverted. | Preserved from `master` | Where | |---|---| | zod parsing of persisted drafts (`persistedDraftsSchema`, `persistedDraftSchema`) | `explorer-query.ts` | | `MAX_PERSISTED_EXPLORER_QUERY_DRAFTS` cap, retaining most-recently-updated | `explorer-query.ts` | | debounced SQL persistence + `flushPendingPersistence`, immediate write-through for rename/source | `explorer-query.ts` | | `removeDraft` clearing pending timers | `explorer-query.ts` | | `getQuerySourceBinding(cell)` and the four source-change branches, incl. `database_identifier` / `time_range` propagation | `QueryCell/index.tsx` | | `restoredQueryKey` per `ref:id` and the `role="status"` loader | `QueryTab.tsx` | | `applyAutoLimit` relocated to `@/data/sql/utils` | `QueryEditor.tsx` | Two adaptations were needed: - `updateDraft` rebuilds the draft through `toDraft` instead of mutating it in place — required, because the object's shape depends on its tag. The debounced `persist` closure still re-reads `state.drafts[id]` at fire time, so behavior is unchanged. - Master's new test `falls back to the database source when persisted source data is invalid` asserted `draft.source`, which the tagged union replaces. Rewritten to assert the same intent against `_tag`. **Dropped from this PR's original description:** it previously claimed to fix a log cell always running against a synthesized default time range. Master fixed that itself by adopting `getQuerySourceBinding` (from #49072), so the claim no longer applies. ## Verification Typecheck, Prettier, and the lint ratchet clean. 736 tests pass across `state/`, the Explorer surfaces, notebooks, query sources, `data/sql`, and the SQL editor — including master's new `QueryTab.test.tsx`, `ExplorerQuerySourceMenu.test.tsx`, `ExplorerQueryTabCoordinator.test.tsx`, and the five draft-store tests added since this branch was cut. <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit - **Bug Fixes** - Improved Explorer query handling across database and logs backends. - Preserved query text when switching backends while clearing incompatible results. - Retained results when changing parameters within the same backend. - Improved restoration of saved drafts, including fallback handling for legacy or invalid sources. - Added validation before executing edited SQL to help prevent invalid requests. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
794e45378c |
test(studio): add get_notebook eval cases (FE-4088) (#49067)
## 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 (test coverage) — second PR of the notebooks-evals plan, covering FE-4088 (Evals: Assistant can read notebook). Follows #49010 (FE-4086, list_notebooks). ## What is the current behavior? `dataset.ts` has no coverage for `get_notebook`. Separately, `NOTEBOOKS_PROMPT` only covers *choosing* between `create_notebook`, `update_notebook`, and `execute_sql` — it says nothing about reading or describing an existing notebook. ## What is the new behavior? **Eval cases** (`evals/dataset.ts`), three of them: - Resolve a notebook by name via `list_notebooks`, then `get_notebook`, and report the queries it actually contains. - Summarize a smaller, single-log-cell notebook as a baseline. - Report a nonexistent notebook id as not found instead of hallucinating contents. The mock's `execute` throws, which the AI SDK surfaces to the model as a `tool-error` part rather than failing the eval task. No new scorers or tool changes — `toolUsageScorer` and `correctnessScorer` already cover these, and `get_notebook` has unit coverage in `notebook-tools.test.ts`. **Prompt change** (`lib/ai/prompts.ts`) — please review this one separately, it's the only production behavior change here: Running the first case surfaced a real gap. The assistant transcribed each cell's SQL correctly but was inconsistent about the configuration that changes what a cell returns — dropping the log cell's time range in some runs, miscounting markdown cells as queries in others. Adds one bullet to `NOTEBOOKS_PROMPT` telling it to report a query cell's configuration and not count markdown cells as queries. Gated behind the Explorer flag, same as the rest of that prompt. <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Added notebook evaluation scenarios for database and log queries, including concise summaries and nonexistent notebook handling. * Improved notebook descriptions by reporting query configurations that affect returned results. * Markdown cells are now excluded from query counts. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
8baaa517d0 |
test(studio): add list_notebooks eval cases (FE-4086) (#49010)
## 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 (test coverage) — first PR of the notebooks-evals plan, covering FE-4086 (Evals: Assistant can list notebooks). ## What is the current behavior? `dataset.ts` has zero notebook eval cases. Separately, two small gaps block writing them: `assistant.eval.ts` never sets `isExplorerEnabled`, so `NOTEBOOKS_PROMPT` never loads into the eval task's system prompt; and `list_notebooks`' `inputSchema` has no sort parameter, even though `getContent` already supports one, so "most recent" isn't answerable. ## What is the new behavior? - `assistant.eval.ts` passes `isExplorerEnabled: true` so `NOTEBOOKS_PROMPT` loads during evals. - `list_notebooks` gains a `sort_by: 'name' | 'inserted_at'` param, forwarded to `getContent`'s `sort`. Only creation order is exposed, since the underlying API has no `updated_at` sort key. - 3 new dataset cases: basic notebook enumeration, sorting by creation time (`sort_by`/`limit` args), and a nonexistent-notebook case guarding against hallucinated results. - A unit test covering `sort_by` forwarding to the content API's query param. ## Additional context <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Added notebook sorting options when listing notebooks, including by name or insertion date. * Added evaluation coverage for listing notebooks, finding the newest notebook, and avoiding fabricated results for nonexistent notebooks. * **Bug Fixes** * Ensured selected notebook sorting preferences are correctly applied when retrieving content. * Improved assistant evaluation coverage with Explorer mode enabled. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
b9835e419c |
test(studio): mock notebook tools (#48952)
## Summary - Adds deterministic `list_notebooks`/`get_notebook` fixtures (two seeded notebooks, each with markdown/database/log cells) and stateful in-memory `create_notebook`/`update_notebook` mocks to `apps/studio/lib/ai/tools/mock-tools.ts`, so Braintrust evals can exercise notebook tool calls without a real project. - Both write-tool mocks force `needsApproval: false`, matching the existing `execute_sql`/`deploy_edge_function` mock pattern — the eval harness filters out tool-parts in `'approval-requested'` state and can never answer an approval gate. - All four notebook tools are wrapped from the real `getNotebookTools()` definitions (only `execute`/`needsApproval` overridden), so evals validate the model's arguments against the exact production schemas. - Dedupes `describeOperationError` (previously duplicated between this new mock and `notebook-tools.ts`) into a single exported `describeNotebookOperationError` in `notebook-operations.ts`. **Stacked on #48949** (`feature/notebooks-update-tool`) — this PR's base branch is that PR, not `master`, because it reuses `update_notebook` and the shared error helper that only exist there. Merge #48949 first, then retarget/merge this one. ## Test plan - [x] \`pnpm --filter studio typecheck\` passes - [x] \`pnpm --filter studio test\` — all notebook-related suites pass (\`mock-tools.test.ts\`, \`notebook-tools.test.ts\`, \`data/content/notebooks/*\`) - [x] \`eslint\` / \`prettier --check\` clean on all touched files |
||
|
Charis
|
4da8bfe469 |
feat(studio): wire update_notebook tool (#48949)
## Summary - Adds `update_notebook`, an AI tool (`needsApproval: true`) that re-fetches a notebook, applies an ordered list of cell operations (insert/replace/delete/move, from `notebook-operations.ts`) via `applyNotebookOperations`, and PUTs the resolved content. Concurrent edits are last-write-wins, since re-fetching at execute time is the only mitigation per the RFC. - Registers `update_notebook` in `toolSetValidationSchema` / `TOOL_CATEGORY_MAP`, and documents it in the notebooks system prompt. - The untrusted→safe SQL promotion (`acceptUntrustedSql`/`acceptUntrustedLogsSql`) is inlined directly inside each `needsApproval: true` tool's own `execute` (both `create_notebook` and `update_notebook`), rather than behind a shared helper — kept auditable right next to the approval gate that authorizes it, and not reusable by an unrelated caller unaware it needs re-promotion. Resolves FE-4083 ## Test plan - [x] `notebook-tools.test.ts`: re-fetch → apply operations → PUT with promoted SQL; descriptive error (no PUT) when an operation targets an unknown cell id; `needsApproval` asserted for both create and update - [x] `pnpm vitest run` on `notebook-tools.test.ts`, `tools/index.test.ts`, `tool-filter.test.ts` — all pass - [x] `tsc --noEmit`, `eslint`, `prettier --check` — clean on all touched files <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * AI-assisted notebook editing now supports updating existing notebooks. * Notebook changes are refreshed before being applied to help preserve the latest content. * Updates report operation errors and return a summary of the modified notebook. * Notebook edits require user approval before being saved. * **Bug Fixes** * Invalid or unknown cell references are reported without applying a partial update. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
4ca17a23ce |
fix(studio): keep the new SQL editor tab from being pruned on first keystroke (#48956)
## What kind of change does this PR introduce?
Bug fix.
## What is the current behavior?
With a snippet already open in the SQL Editor, clicking the `+` button
and typing in the new tab made it look like the previously open tab was
being taken over: the tab bar showed the *old* snippet's name while the
editor showed the newly typed content. With several tabs open, the
rightmost one appeared to be the one taken over.
`useSqlEditorTabsCleanup` prunes any `sql-*` tab whose snippet is absent
from the server-fetched snippet list, so tabs for snippets deleted in
another session don't linger. But a snippet created by typing in a new
tab exists only in the local store until its first save lands, so it is
legitimately absent from that list — and the same keystroke that creates
it calls `setSql({ shouldInvalidate: true })`, invalidating the snippet
lists and triggering a refetch that pruned the tab that had just been
opened.
`removeTab` then reassigns `activeTab` to a neighbor, so the tab bar
fell back to whichever tab was open before, while the URL, sidebar, and
editor content all stayed correctly on the new snippet — none of them
read from the tabs store.
Confirmed from a user's persisted tab state, which showed a single tab
in `openTabs` pointing at the previous snippet while the URL pointed at
the new one.
No snippet content was ever lost — this was tab state only.
## What is the new behavior?
Tabs (and recent items) whose snippet is present in the local store with
a never-persisted status (`new`, `new_saving`, `new_save_failed`) are
preserved by the cleanup pass. Genuine stale-tab pruning for snippets
removed outside the session is unaffected.
## Additional context
The regression test in `Tabs.utils.test.tsx` fails without the fix
(`expected undefined to be defined`) and passes with it. The
pre-existing pruning tests still pass, confirming legitimate cleanup
still works.
Also adds two tests covering adjacent invariants that were verified
while narrowing this down: snippet content isolation between tabs, and
unique snippet id generation across `/sql/new` navigations.
Verified: 343 tests pass across `components/layouts/Tabs` and
`components/interfaces/SQLEditor`; typecheck clean; ESLint warning count
unchanged (ratchet safe); Prettier clean.
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
- **Bug Fixes**
- Preserved newly created SQL editor tabs and recent items before their
first save.
- Prevented unsaved snippets from being removed during tab cleanup.
- Ensured editing a new tab does not overwrite content in existing tabs.
- Ensured successive new SQL tabs receive distinct identities.
- **Tests**
- Added regression coverage for snippet editing, route changes, and tab
cleanup behavior.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
|
||
|
Charis
|
065468f8ac |
fix(studio): reset rename form after renaming a SQL snippet (#48951)
## 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? Bug fix ## What is the current behavior? The SQL snippet rename modal is mounted once per nav and reused for every snippet, so a single form instance is shared across renames. On a successful rename the form was never re-baselined, leaving it dirty, and the effect that synced the form to the selected snippet bailed out whenever the form was dirty. Renaming a second snippet therefore opened the modal pre-filled with the previous snippet's name, with the submit button enabled — one careless confirm renamed the wrong query. ## What is the new behavior? The form is reset after a successful rename, and the hand-rolled sync effect is replaced with react-hook-form's `values` option so the form follows whichever snippet is selected. `keepDirtyValues` keeps a background refetch from clobbering in-progress input, which is what the old dirty guard was protecting against. It has to be disabled explicitly on the resets that discard input, since `resetOptions` on `useForm` applies to every `reset` call — not just the `values`-driven one. Adds component tests covering the submit path, the rename-then-rename regression, and discarding an abandoned edit on cancel. ## Additional context Fixes FE-4114 <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit - **Bug Fixes** - Improved the rename query experience by ensuring the selected snippet name is displayed correctly when reopening the rename dialog. - Cancelled edits are now discarded reliably, preventing unsaved changes from persisting. - After a successful rename, the form reflects the updated query name and maintains consistent input and button behavior. - **Tests** - Added coverage for successful renaming, cancellation, reopening with a newly selected snippet, and submitted values. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
810d292121 |
feat(studio): notebook cell operations (#48940)
## Summary - Pure module (`data/content/notebooks/notebook-operations.ts`) for applying `update_notebook` cell edits client-side: `insert_cell` (`after_cell_id` incl. `'start'`), `replace_cell`, `delete_cell`, `move_cell`. - Never touches the safe-sql brands — SQL promotion still happens at the tool-execute boundary, matching `create_notebook`. - Stacked on #48938. No wiring yet — `update_notebook` tool wiring is next. Towards FE-4083 ## Test plan - [x] `pnpm vitest run data/content/notebooks/notebook-operations.test.ts` — 13 unit tests covering every op, combinations, and all three error cases. - [x] `pnpm exec tsc --noEmit` clean - [x] `pnpm exec eslint` clean on new files <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Added support for applying notebook cell operations, including insertion, replacement, deletion, and movement. * Operations are applied in a predictable order, with support for anchoring new cells at the beginning or near existing cells. * Added validation for invalid references, conflicting operations, and self-referential moves. * Added clear handling when operations produce an empty notebook result. * **Tests** * Added comprehensive coverage for individual, combined, ordered, conflicting, invalid, and empty-result notebook operations. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
ddb3e2c442 |
feat(studio): create_notebook AI tool (#48938)
## Summary - Adds a `create_notebook` AI assistant tool (`needsApproval: true`) that lets the assistant create a new notebook after explicit user approval. - Cell SQL is promoted from untrusted to safe via `acceptUntrustedSql`/`acceptUntrustedLogsSql` inside `execute`, using the approval gate as the confirming user gesture (same pattern as `execute_sql`). - Input is validated against the existing agent-writable notebook schema, which rejects any agent-supplied cell `id` at the schema level. - Threads an optional auth-headers param through `upsertContent`/`createNotebook`/`updateNotebook` so the tool can pass its own bearer token server-side. - Registers the tool in the tool-filter (`SCHEMA` category, alongside `list_notebooks`/`get_notebook`) and adds a `## Notebooks` prompt section guiding the assistant on when to use `create_notebook` vs. one-off `execute_sql`. Resolves FE-4082 ## Test plan - [x] `notebook-tools.test.ts` covers: tool registration, `needsApproval`, cell-id rejection, valid input, PUT body shape, and the returned id — all passing - [x] Typecheck clean - [x] Lint clean (no new warnings) <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Added AI-assisted notebook creation for saving multi-step investigations. * Added support for database and log SQL cells in newly created notebooks. * Notebook creation requires approval before saving and returns the notebook’s name and identifier. * Added support for custom request headers during notebook and content operations. * Added guidance for choosing between one-time SQL execution and reusable notebooks when Explorer is enabled. * **Improvements** * Improved validation and normalization of notebook content before saving. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
4587d177c3 |
Add optional title field to notebook cells (#48937)
## Summary - Adds optional `title` field to `databaseCellSchema` and `logCellSchema` in notebook schema - Allows database and logs notebook cells to carry descriptive titles - Field automatically propagates through derived schemas (wire, writable, agent, domain) via Zod inheritance <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Added optional titles to database and log notebook cells. * Cell titles are now preserved across notebook editing, viewing, and agent workflows. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
7798e42435 |
feat(studio): notebook read tools (#48908)
## Summary - Adds `list_notebooks` (cursor-paginated) and `get_notebook` AI tools in `lib/ai/tools/notebook-tools.ts`, modeled directly on `report-tools.ts`: server-side `getContent`/`getNotebook` with the `authorization` header forwarded, zod-validated input. - `get_notebook` resolves every cell and exposes `unchecked_sql` as a plain `sql` field for the agent to read — display only, per the `safe-sql-execution` skill; nothing here executes SQL. - Registers both tools in `lib/ai/tools/index.ts` (same platform branch as reports) and in `lib/ai/tool-filter.ts`'s `toolSetValidationSchema` + `TOOL_CATEGORY_MAP` (`SCHEMA` tier). - Adds an optional `headers` param to `content-infinite-query.ts`'s `getContent`, mirroring the sibling `content-query.ts`, so the cursor-paginated fetch can carry the `Authorization` header from a server context. - New tools are behind the Explorer feature flag. Stacked on #48907 (1.4 — notebook query and mutation hooks), per the Notebooks implementation plan (stack 2.1). Resolves FE-4081 Resolves FE-4080 ## Test plan - [x] `pnpm exec tsc --noEmit` — no new errors - [x] `pnpm exec vitest run lib/ai/tools/notebook-tools.test.ts lib/ai/tools/index.test.ts lib/ai/tools/report-tools.test.ts data/content/notebooks` — 36/36 passing - [x] `pnpm --filter studio run lint` — no new warnings - [x] `pnpm exec prettier --check` on changed files — clean <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Added AI tools to list project notebooks with pagination. * Added AI support for retrieving notebook markdown and resolved SQL cell content. * Notebook tools now respect project and authorization context. * Notebook features are available only when Explorer access is enabled. * Content requests can forward custom request headers. * **Tests** * Added coverage for notebook tools, Explorer access, feature flags, authorization, pagination, and error handling. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
1296a1c745 |
feat(studio): notebook query and mutation hooks (#48907)
## Summary Implements the "notebook query and mutation hooks" step of the notebooks data layer: - `data/content/notebooks/notebook-query.ts` — `getNotebook`/`useNotebookQuery`, wrapping the existing `getContentById` and narrowing to `type: 'notebook'`. - `data/content/notebooks/notebooks-infinite-query.ts` — `useNotebooksInfiniteQuery`, a typed wrapper over `useContentInfiniteQuery` narrowing pages to notebook rows. - `data/content/notebooks/notebook-upsert-mutation.ts` — `createNotebook`/`updateNotebook` + their mutation hooks, PUTting through the existing `upsertContent`. Write-path correctness, worked out while building the mutation hooks: - Cell `id`s are always backend-generated, never client-supplied — a brand-new cell has no `id` at all; an existing cell being kept/edited in an update keeps its real id so the backend can diff it against the previous version. `notebook-schema.ts` gains `writableCellSchema`/`writableNotebookSchema` (ids optional per cell) and `WritableCell`/`WritableNotebook` types, derived from `z.infer` of those schemas rather than hand-duplicated, with only the `sql` field re-branded per cell type via a small distributive conditional type. - Cell SQL at this write boundary must already be `SafeSqlFragment`/`SafeLogSqlFragment` (proven user-authored at a save/run event handler), not `unchecked_sql` — matching the `safe-sql-execution` skill's provenance model. - `content-remap.ts`'s notebook `unmapSqlContentField` branch is simplified to a passthrough: notebook writes only ever arrive already wire-shaped via `createNotebook`/`updateNotebook`, so there's nothing left to unmap. Note: this was originally stacked on `feature/notebooks-types-convergence`, but that branch merged into `master` (#48905) while this PR was in progress, so it's rebased directly onto `master` now. ## Test plan - [x] `pnpm --filter studio run typecheck` passes - [x] `pnpm --filter studio exec vitest run data/content/notebooks data/content/content-remap.test.ts` — 38/38 passing - [x] `pnpm --filter studio exec eslint` clean on all touched files <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Added notebook listing with pagination, filtering, sorting, and project-specific queries. * Added notebook retrieval for viewing individual notebooks. * Added notebook creation and editing with automatic content refresh. * Added support for preserving cell IDs and safely handling SQL content. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
0111aa371b |
ref(studio): converge notebook UI types with canonical content schema (#48905)
## Summary - Joshen's `state/notebooks/types.ts` (Explorer/notebook editor UI) redefined its own `TimeRange`, cell union, and `NotebookContent` shapes, duplicating the canonical schema from `data/content/notebooks/notebook-schema.ts` (#48813, #48815). - Points `Notebook.content` and `notebooksState.updateCells` at the canonical `Notebooks.Content` / `Notebooks.Cell` types (via `@/types`) instead, and fixes the handful of call sites that constructed notebook content by hand to match the real wire shape: `schema_version: 1` (not `'1.0'`) and `_tag`-discriminated cells (e.g. `{ _tag: 'markdown_cell', id, text }` instead of `{ type: 'markdown', content }`). - No behavioral changes — Joshen's state management, editor component, and hooks are untouched aside from the type-level fixes needed to compile against the canonical schema. ## Test plan - [x] `pnpm exec tsc --noEmit` — no new errors - [x] `pnpm exec vitest run state/notebooks/notebooks-state.test.ts components/interfaces/Explorer/__tests__/NotebookEditor.test.tsx` — 8/8 passing - [x] `pnpm exec eslint` on changed files — clean - [x] `pnpm exec prettier --check` on changed files — clean <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Bug Fixes** * Updated notebook content handling to use the current schema version format. * Improved compatibility for markdown cells, including their identifiers and text. * Standardized notebook content and cell updates for more consistent behavior. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
CharisandJoshen Lim
|
957e9fec67 |
feat(studio): notebook content at the API boundary (#48815)
## Summary Stacked on #48813 (1.2: notebook content schema). Part of [FE-4109](https://linear.app/supabase/issue/FE-4109/notebooks-data-model) — see that issue for the rest of the notebooks data-model stack. - Teach `content-remap.ts`'s wire↔domain dispatcher about the `notebook` content type, branding each cell's `sql` per `_tag` via the notebook schemas added in 1.2 (parses through `notebookDomainSchema` on the way in, unbrands per cell on the way out). - Add `{ type: 'notebook'; content: Notebooks.Content }` to the `Content` union in `content-query.ts`, plus a `ContentOfType<T>` helper for narrowing it. - Fix the resulting narrowing fallout at call sites that assumed `Content` only ever meant `sql`/`report`/`log_sql`: two generated-query-param casts, and four report/logs call sites now narrowed via `ContentOfType<'report'>` / `ContentOfType<'log_sql'>`. ## Test plan - [x] `pnpm --filter studio vitest run data/content/` — 35 tests pass, including new notebook coverage in `content-remap.test.ts` (per-cell brand separation, missing-field throw, remap↔unmap round-trip) - [x] `pnpm typecheck` — clean (pre-existing unrelated `ui-patterns` error aside) - [x] `pnpm --filter studio lint` — no new warnings/errors on changed files - [x] `pnpm format` — clean --------- Co-authored-by: Joshen Lim <joshenlimek@gmail.com> |
||
|
CharisandJoshen Lim
|
0b97e37ccf |
feat: notebook content schema (#48813)
Related to FE-4109. ## Summary - **API codegen workaround**: Platform API's `notebook` content type hasn't shipped to the OpenAPI spec yet, so `pnpm api:codegen` can't be run. Locally widened `ContentBase.type` to include `'notebook'` (marked with TODO for removal once spec publishes). - **Notebook schema & type system**: Introduced Zod schemas mirroring RFC-defined notebook shape (`schema_version: 1, cells: Cell[]`). Maintains wire/domain boundary (cell `sql` → `unchecked_sql` branded for security). Agent-writable schema for `create_notebook` tool omits cell IDs (backend-generated); future update operations will require them. All TypeScript types are `z.infer`'d from schemas (no hand-written parallel interfaces). - **IsoDateTimeString moved**: Extracted ISO datetime validator from `querySource.ts` to `lib/iso-datetime.ts` (data layer shouldn't import from components layer). Needed by notebook `time_range` fields. ## Test plan - [x] Unit tests: `notebook-schema.test.ts` (9 tests), `iso-datetime.test.ts` (3 tests), `querySource.test.ts` updated and passing (26 tests) - [x] Typecheck: no new errors - [x] Prettier: formatting clean <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit - **New Features** - Added support for validating and processing notebook content, including markdown, database, log cells, time ranges, and chart configurations. - Added compatibility for notebook content types in content handling. - Added reliable ISO date-time validation for notebook data and related features. - **Tests** - Expanded coverage for valid and invalid notebook structures, cell requirements, time ranges, chart settings, and date-time values. <!-- end of auto-generated comment: release notes by coderabbit.ai --> Co-authored-by: Joshen Lim <joshenlimek@gmail.com> |
||
|
Charis
|
8831e15fc3 |
Revert "refactor(studio): static tab kind registry and tab id codec (PR 1/9)" (#48762)
Reverts supabase/supabase#48718 |
||
|
Charis
|
ffd6d6636d |
refactor(studio): static tab kind registry and tab id codec (PR 1/9) (#48718)
## Summary First PR in the tab/snippet decoupling stack. Purely additive: no existing call sites change behavior except the `EntityTypeIcon` rewire, which preserves its API exactly. - **`state/tabs/kinds.ts`** — static `TAB_KINDS` descriptor table (`TabKind`, `TabSurface`, `surfaceOf`, `kindsOnSurface`, `isTabKind`). Leaf-safe: its only reference to the domain `Tab` type is a type-only import, so it stays importable at module scope (needed later for the valtio store and the persistence migration reader) without creating a runtime cycle. - **`state/tabs/kinds.icons.tsx`** — per-kind icon leaf module (`ui` + `lucide-react` + `TabKind` type only), plus the preserved `LogsSnippetIcon`. - **`state/tabs/tab-id.ts`** — `createTabId` / `parseTabId` / `toUrlSegment` / `parseUrlSegment(segment, surface)` codec. `parseUrlSegment` is surface-scoped: a bare segment only resolves to a kind when the surface has exactly one bare kind (true for `sql`). The table surface has five bare kinds (`r`/`v`/`m`/`f`/`p`) with no URL disambiguator between them, matching `/editor/[id]`, which learns kind from the fetched entity rather than the URL — so a bare table segment correctly resolves to nothing. - **`components/ui/EntityTypeIcon.tsx`** — rewired to a thin wrapper delegating to `kinds.icons.tsx`, preserving its exact prop API (`type`, `size`, `strokeWidth`, `isActive`, `sqlSource`) for all existing consumers. - **`state/tabs/tab-id.test.ts`** — codec round-trips per kind, bare-vs-prefixed URL segments, the `templates`/`examples`/`new` sentinels, and surface scoping (including the table-surface ambiguity above). Full plan: `apps/studio/TABS_DECOUPLING_PLAN.md` (not included in this PR). ## Test plan - [x] `pnpm typecheck` - [x] `pnpm lint --filter=studio` (0 errors) - [x] `pnpm --filter studio run lint:ratchet` (warning counts did not increase) - [x] `pnpm test:studio` (full suite green, including 50 new/updated tests in `state/tabs/`) - [x] `pnpm format` (no-op) <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Added consistent icons and metadata for SQL, notebook, chat, table, view, and related tab types. * Added support for creating, parsing, and converting tab identifiers to URL segments. * Improved handling of tab types across SQL and table surfaces. * **Bug Fixes** * Invalid, empty, or ambiguous tab identifiers and URL segments are now rejected. * **Tests** * Added coverage for tab identifiers, URL conversion, supported tab types, and edge cases. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
2165746784 |
fix(studio): keep SQL editor source menu open when switching sources (#48715)
## 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? Bug fix ## What is the current behavior? Selecting Database/Logs in the SQL Editor's query-source dropdown closes the menu (Radix's default select behavior), so switching to Logs gives no visible indication that a Time range control just became available until the dropdown is reopened. ## What is the new behavior? Selecting a source keeps the dropdown open, so the newly-available source-specific controls (e.g. Time range for Logs) are immediately visible. ## Additional context Fixes FE-4036 <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Bug Fixes** * Improved source switching in the SQL editor so the selection menu remains open while changing between database and logs sources. * Ensured source-specific controls update correctly after switching. * **Tests** * Added coverage for source selection, menu behavior, and source-specific control updates. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
5986fecc89 |
fix(studio): use table + card component in log retention upgrade prompt (#48713)
## Summary - Refactor the SQL Editor / Logs "Log retention" upgrade-prompt dialog to use the design system's `Card` + `Table` components instead of a custom flexbox-div table, matching the pattern in `apps/design-system/registry/default/example/table-demo.tsx` - No behavior or data change Fixes FE-4034 Non-blocking review nit from #48452 (comment: https://github.com/supabase/supabase/pull/48452#issuecomment-5128918096) ## Test plan - [x] `pnpm --filter studio typecheck` passes - [x] `pnpm --filter studio run lint:ratchet` passes (no new warnings) - [x] Prettier check passes <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Style** * Updated the upgrade prompt with a more consistent card and table layout. * Preserved existing plan names and log retention details. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
b9053794a0 |
fix(studio): add tooltip explaining why Prettify SQL is disabled (#48712)
## 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? Bug fix ## What is the current behavior? In the SQL Editor, the Prettify SQL action (both the "More actions" dropdown menu item and the toolbar button) is disabled for logs snippets, but gives no indication why. ## What is the new behavior? The disabled Prettify menu item now uses `DropdownMenuItemTooltip` and the disabled Prettify toolbar button uses `ButtonTooltip`, both showing "Can only prettify SQL queries" while disabled. Addresses review feedback from #48452 (Linear FE-4038). ## Additional context Resolves FE-4038 <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Improvements** * Updated SQL editor tooltips with clearer, consistent messaging. * Log-source users now see an explanation when SQL formatting is unavailable. * Regular users continue to see the SQL prettify keyboard shortcut. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
89010c1649 |
fix(studio): show selected custom time range in SQL editor time range menu (#48709)
## What Display the actual selected custom date range in the Time Range dropdown's secondary text, instead of the static "Custom range" label. The range is formatted as `DD MMM, HH:mm - DD MMM, HH:mm` to match the existing format in the Logs Explorer's date-picker trigger button, ensuring visual consistency across the Logs UI. Fixes FE-4035 ## Test plan - [x] Typecheck passes: `pnpm typecheck` - [x] Manually verify: Open SQL Editor with `sqlEditorLogsSource` flag enabled, open the Time Range dropdown menu, select a Custom range, and confirm the dropdown's secondary text now displays the selected date range instead of "Custom range" <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Custom time ranges now display formatted start and end timestamps for clearer time selection. * Preset time ranges continue to show their existing helper text. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
0791b04eb8 |
feat(sql-editor): roll out manual saving by default (#48706)
## 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 — progresses the SQL Editor manual saving rollout, and removes a fully rolled out feature flag. ## What is the current behavior? Manual saving (snippets save only on an explicit Save / Cmd+S rather than autosaving every edit) is opt-in. It requires both the `sqlEditorManualSave` ConfigCat flag and the user turning on the "Disable snippet auto-saving" feature preview themselves. That flag is now fully rolled out. ## What is the new behavior? - `sqlEditorManualSave` is removed, so the feature preview is listed for everyone. (Delete the flag in ConfigCat after a few months.) - New `sqlEditorManualSaveForced` flag progresses the rollout. It forces manual saving on regardless of what the user stored previously, including an explicit opt-out, via a new `isForced` field on `FeaturePreview` that `initializeFlags` resolves ahead of the localStorage lookup — so the feature preview modal reflects the forced state too, not just the save behavior. Turning the flag off reverts everyone who never opted in, so it remains a working kill switch. - Users the rollout switches over get a one-time dialog on their first SQL Editor route, explaining that snippets no longer autosave. Dismissal persists in `sql-editor-manual-save-notice-dismissed` (allowlisted, so it survives sign-out). - Users who opted into the preview themselves never see the dialog — it records their dismissal up front instead, since the notice needs to outlive the feature preview and once the preview is retired there's no stored opt-in left to recognize them by. - The preview keeps its switch so users who lose their local storage can opt in early, but once the rollout reaches them the "Disable feature" button is disabled with a tooltip explaining why. ### To test Turn on `sqlEditorManualSaveForced` on via the dev toolbar. - No `supabase-ui-sql-editor-manual-save` and no dismissal key → dialog appears on `/project/<ref>/sql`, toolbar shows the Save button. Dismiss, reload → no dialog. - `supabase-ui-sql-editor-manual-save` set to `false` (previously opted out) → still forced onto manual saving, and still gets the dialog. - `supabase-ui-sql-editor-manual-save` set to `true` → no dialog, and the dismissal key is written automatically. - Flag off, no opt-in → autosave, no dialog, and the "Disable autosave" power-off button still opens the preview modal. - Preview modal while forced → "Disable feature" is disabled with a tooltip; another preview (e.g. Column-level privileges) still disables normally. ## Additional context |
||
|
Charis
|
21511042a3 |
feat(studio): assistant logs context and reports guard (#48514)
## 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 — final PR (9/9) of the SQL editor logs-source stack. **Base branch:** `charislam/sql-editor-inline-ai-clickhouse-dialect` (PR 8). Nothing here is user-visible: entry points stay behind `sqlEditorLogsSource` + `otelLegacyLogs`, and flag rollout happens after the whole stack merges. ## What is the current behavior? - The Assistant has no idea a SQL editor snippet targets the logs backend. Ask it about a logs snippet and it answers in Postgres, because the attached query is fenced as ` ```sql ` and nothing tells the model otherwise. - Because the `sql` fence is what `MessageMarkdown` treats as runnable Postgres, an attached ClickHouse query is rendered with a Run-against-Postgres affordance and branded with `untrustedSql`. - "Debug with Assistant" on a failed logs query produces a dialect-less prompt, so both the in-app assistant and the copyable version get debugged as Postgres. - A report referencing a `log_sql` snippet runs its ClickHouse SQL against the user's Postgres database and surfaces the resulting error. ## What is the new behavior? **Assistant panel.** The "Current Query" chip records which backend the attached query targets. That reaches the model two ways: each attachment is fenced with its own dialect (` ```clickhouse ` vs ` ```sql `), and a `containsLogsSnippets` flag rides on the user message as AI SDK `metadata`. The server reads the flag off the conversation and prepends the ClickHouse dialect rules plus the logs schema reference as a non-cached context message. Two design points worth calling out in review: - The flag lives on the **message**, not the request body, so Retry and the tool-approval continuation reproduce the context a message was originally asked in — neither of those passes a per-call body. - It's derived from **what's actually attached**, so detaching the chip drops the claim rather than leaving the two able to disagree. The `clickhouse` fence also keeps a logs query out of `MessageMarkdown`'s `sql` branch, so it's no longer offered as runnable Postgres or branded with `untrustedSql` — a boundary this stack's distinct brands exist to prevent crossing. **Debug flow.** `buildDebugChatArgs` attaches its query with a source for the same reason, and names the dialect in the prompt text so the copyable version stands on its own outside the app. **Reports.** A report only stores a snippet id, so whether it queries the logs backend is only knowable once the content loads. `ReportBlock` guards on the fetched type and renders a `LogsSnippetReportBlock` placeholder instead of executing. Double-guarded: no `sql` for a logs snippet (so it's out of the query key and `queryFn` short-circuits even on an explicit `refetch`) and `enabled` excludes it. **Incidental cleanups.** `buildAssistantContextMessages` extracted out of `generate-assistant-response`; a schema-access sentinel that was duplicated as a string literal across two files (and compared against) replaced with one exported constant; `SqlSnippet` deduplicated to a single declaration; `resolveSnippetSource` / `isLogsSource` shared instead of re-implemented per surface. **Tests.** 4 new/extended suites. Notable cases pinned: a message with no metadata must validate (`safeValidateUIMessages` applies `metadataSchema` to *every* message, so a required schema would 400 every existing conversation); only *user* messages count, so a model reply can't talk the server into a different dialect; a mixed-attachment message is flagged without overclaiming a single source; and `ReportBlock` registers no pg-meta mock for the logs cases, so an unhandled request failing the test *is* the assertion that logs SQL never reaches Postgres. Verified: `pnpm typecheck`, `lint:ratchet` (no regression), Prettier, and the full Studio suite (459 files / 4969 tests). ## Additional context <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit - **New Features** - Added support for recognizing log snippets in reports, with clear guidance to open them in the SQL editor or remove them. - AI Assistant now understands log snippets and provides ClickHouse-specific context, formatting, and troubleshooting guidance. - Snippets retain their source information when shared with the AI Assistant. - **Bug Fixes** - Prevented unsupported log snippets from being executed as regular database queries. - Improved source detection when opening snippets directly from links. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
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 --> |
||
|
Charis
|
b3c5c9fc04 |
feat(studio): logs snippets in SQL editor nav, search, and tabs (#48457)
## What PR 7 of the SQL-editor query-source (Database vs Logs) stack. Surfaces `log_sql` snippets as a distinct query source across the SQL editor sidebar. Stacked on **`charislam/toolbar-ui-creation-flow`** (PR 6 — toolbar UI + creation flow); review/merge that first. Nothing is user-visible until the flags roll out — every entry point requires **both** `sqlEditorLogsSource` **and** `otelLegacyLogs`. ## Changes - **Nav** — a flag-gated **Logs** section (`LogsSnippetsSection`) backed by its own single-type `log_sql` query. The active snippet is injected only into the section it belongs to, via a shared `withActiveSnippet(snippets, active, belongsPredicate)` helper (also DRYs the private/favorites/shared injections). - **Search** (`SearchList`) — a **Logs** result group with a shared, extracted `SqlSnippetTree`; the "N results found" count now sums database + logs, with loading/empty states covering both queries. - **Tabs** — an immutable `sqlSource` field on tab/recent-item metadata (set at tab creation, lazily backfilled once the snippet loads via `useEffectEvent`), and a distinct `ScrollText` icon via a shared `LogsSnippetIcon`. Tab cleanup treats `log_sql` tabs as live and only prunes them when logs data is authoritative (`canPruneLogsTabs`), so a disabled/erroring logs query never wrongly deletes logs tabs or blocks database-tab cleanup. - **Data layer** — `useSqlSnippetsQuery` gains an optional `type` param so logs reuse the same `SnippetWithContent` shape as the other sections (no casts). ## Tests - `state/tabs.test.ts` — `sqlSource` backfill + creation-time carry-through. - `components/layouts/Tabs/Tabs.utils.test.tsx` — cleanup prunes stale database/logs snippets, keeps live ones, and preserves logs tabs when logs data isn't authoritative. ## Verification - `pnpm --filter studio typecheck` ✓ - `pnpm --filter studio run lint:ratchet` ✓ - `pnpm test:studio` (affected suites) ✓ <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Added a collapsible Logs section to the SQL editor sidebar for browsing, sorting, selecting, renaming, and deleting log queries. * Expanded SQL search with separate, paginated results for database and log queries. * Added dedicated log-query icons across navigation, tabs, previews, and recent items. * **Bug Fixes** * Improved tab and recent-item cleanup while preserving active log queries and accurate source metadata. * **Tests** * Added coverage for log tab cleanup and SQL source metadata synchronization. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
ec64135f9d |
perf(pg-meta): scope column privileges query to a single table (#48553)
## 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? Performance improvement ## What is the current behavior? The column privileges page in Studio only ever renders one table, but the underlying query still `aclexplode`s every column in the whole schema and filters the result client-side. ## What is the new behavior? Adds a scoped SQL path that prunes `pg_class`/`pg_namespace` to the requested schema+table before exploding ACLs, gated behind the `pgMetaScopedIntrospection` flag, with a plan-guard test asserting `pg_class`/`pg_attribute` stay index-driven. Studio's query hook and cache keys now thread the selected table through so column-privilege invalidation and cold-load races are scoped correctly, and the page fetches per-table instead of per-schema. ## Additional context <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Improvements** * Column privileges are now scoped to the selected schema and table for more accurate results. * Changing schemas automatically updates the table selection and refreshes the displayed privileges. * Privilege updates now refresh only the relevant schema, table, and column data. * Loading states are handled more accurately when no table is selected. * **Bug Fixes** * Improved consistency between scoped and unscoped column privilege results, including table-, column-, and grant-option privileges. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
50e1eb7436 |
chore(eslint): bump eslint-config-next to v16 for useEffectEvent (#48458)
## 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? Chore / build (ESLint config upgrade + lint cleanup). ## What is the current behavior? `eslint-plugin-react-hooks` v5 (pulled in transitively by `eslint-config-next` v15) doesn't recognize stable `useEffectEvent`, so every effect that calls an effect-event handler needs an `eslint-disable react-hooks/exhaustive-deps` to silence a false positive. There are 30 such dead disables across Studio. ## What is the new behavior? Bumps `eslint-config-next` to v16, which pulls in `eslint-plugin-react-hooks` v7 whose `exhaustive-deps` understands `useEffectEvent`, and removes the 30 now-dead disable directives (and their orphaned explanatory comments). Supporting changes: - **Flat-config migration**: v16 is a native flat-config array (v15 was eslintrc), so `eslint-config-supabase` now spreads it directly instead of bridging through `FlatCompat`. - **React Compiler rules off**: v16 enables react-hooks v7's `recommended`, which layers the React Compiler lint rules on top of the two classic rules. These are switched off (derived dynamically from what next enables) to keep this change scoped to the `exhaustive-deps` improvement. - **Plugin-registration fallout** (v16 scopes plugin registration to a file glob rather than registering globally like FlatCompat did): stop re-registering `@typescript-eslint` (shared) and `jsx-a11y` (studio); scope our react / react-hooks / jsx-a11y rule overrides (studio, www) to v16's plugin glob so they don't error on files outside it (e.g. `.cjs`). - **Lint surface preserved**: v16's glob newly includes `.mts`/`.cts` (v15 didn't lint them), which surfaced pre-existing errors in tooling scripts. The shared config keeps the prior surface by leaving `.mts`/`.cts` unlinted; linting them is left as a separate change. - **Ratchet**: rebaselines `@tanstack/query/exhaustive-deps` 9 → 89. v15 forced next's `@babel/eslint-parser` onto `.ts` files, hiding these deps; v16 parses `.ts` with `@typescript-eslint/parser` and correctly surfaces the intentional `connectionString`-excluded-from-`queryKey` pattern. Worth a follow-up to review whether any are real cache-correctness bugs. - Drops three now-dead devDeps from `eslint-config-supabase`: `@eslint/eslintrc`, `@eslint/js`, `@typescript-eslint/eslint-plugin`. Verified locally: `turbo run lint` → 7/7 packages pass with 0 errors; Studio `lint:ratchet` passes; Prettier clean on changed files; typecheck unaffected. ## Additional context <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Chores** * Refined linting configuration and removed outdated lint suppressions across Studio. * Updated Next.js linting support and refreshed related development configuration. * Expanded lint baseline coverage for query-related code. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
CharisandJoshen Lim
|
c16c7e94cc |
feat(studio): SQL editor logs source — toolbar UI + creation flow (#48452)
## What PR 6 of the SQL-editor "query source (Database vs Logs)" stack (builds on the merged PR 5, #48414). Adds the user-facing toolbar surface for the logs query source and consolidates the SQL-editor toolbar into a single **source menu**. Everything stays behind `sqlEditorLogsSource` + `otelLegacyLogs` (dual-flag gated); with the flags off the toolbar is unchanged. ## Changes - **Consolidated source menu** (`QuerySourceMenu`) — one `Database ▾` / `Logs ▾` dropdown that both labels the snippet's source and hosts the source-specific controls as flyout submenus: - Database: database selector (`Primary` / read replicas), `Run as` (role impersonation), and `Row limit`. - Logs: `Time range` — the same relative presets as the Logs Explorer plus a `Custom range…` calendar dialog. - **Source is immutable** — the Database/Logs rows aren't a toggle. An existing (materialized) snippet opens a *fresh* tab of the target source (never reinterpreting a query against the wrong backend); a blank new tab re-flavors in place. Extracted as the pure, unit-tested `resolveSourceSwitch`. - **New-snippet-with-source** threaded through `/sql/new?source=`, the nav "Create a new logs query" entry, and the duplicate flow. Logs snippets hide the (db-dialect) Export action. - **Run-affordance guard** — the Run button is disabled + annotated for a logs snippet on a non-ClickHouse org (sits above PR 5's execution short-circuit). - **Retention entitlement gating** — both preset and custom logs ranges past `log.retention_days` surface the upgrade prompt instead of applying silently. Prettify is disabled for logs (sql-formatter mangles ClickHouse). ## Tests - `querySource.test.ts` — `logDateRangesEqual` (structural relative/absolute matching, incl. the "Last hour" vs "Last 1 hour" label case). - `QuerySourceMenu.utils.test.ts` — `resolveSourceSwitch` push-vs-replace / no-op behavior. `pnpm --filter studio typecheck` · `lint:ratchet` · Prettier · SQL editor suite (307 tests) all green. ## For reviewers To test manually, enable the `sqlEditorLogsSource` feature flag for yourself on local/staging. There is no nav for Log SQL snippets currently (that is by design, this PR is big enough as-is), so to check an existing logs snippet, you can create one using the existing Logs Explorer, copy its UUID, and force navigate to that snippet in the SQL editor via URL. <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Added support for creating and switching between database and logs queries. * Added log time-range presets and custom date-range selection. * Added database, run-as role, and row-limit controls. * Added read-replica selection, including options to create a new replica when available. * **Improvements** * Added clearer explanations when query execution is unavailable. * Disabled SQL formatting and query export where unsupported for logs queries. <!-- end of auto-generated comment: release notes by coderabbit.ai --> --------- Co-authored-by: Joshen Lim <joshenlimek@gmail.com> |
||
|
Charis
|
4c8ed105d2 |
feat(studio): logs SQL execution wiring + source-aware run gestures (#48414)
## 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 (SQL editor: execution wiring for logs-source snippets). Part of the stacked SQL-editor "Database vs Logs" query-source series. ## What is the current behavior? The SQL editor only ever runs queries against the user's Postgres database. There is no execution path for a logs (`log_sql`) snippet, and the run-button telemetry event carries no backend discriminator. ## What is the new behavior? - `useRunSource(id)` derives the run backend from the snippet type; a `log_sql` snippet resolves to `{ type: 'logs', dateRange }`, pairing the run with its session time range (default: last hour). - `useLogsSqlExecution` runs a promoted `SafeLogSqlFragment` against the analytics OTEL (ClickHouse) endpoint with the resolved time range as `iso_timestamp_start`/`iso_timestamp_end` request params. The endpoint is **pinned to OTEL** — a snippet's dialect must not flip with org migration. - The run gestures (toolbar button and Cmd+Enter) branch on the source and promote with the matching `acceptUntrusted*` right at the user action, preserving the auditable promotion-at-gesture boundary. pg intellisense is gated off for logs snippets. - The `sql_editor_query_run_button_clicked` telemetry event gains a required `{ source: 'database' | 'logs' }` property, fired from both execution paths. - Capability guard: a `log_sql` snippet is reachable by direct URL regardless of the (later) entry-point flag gating, so `executeLogsQuery` short-circuits when `otelLegacyLogs` is off — recording a clear "not available yet" result message instead of firing a request that would only return an opaque backend error on a non-ClickHouse project. This is a guard on the gesture, not endpoint selection. - Tests: `useRunSource` routing, `useLogsSqlExecution` endpoint/range/structured-error/capability-guard, and a reusable `flags` option on `renderSqlEditorHook`. No UI entry points are added — the feature runs dark until the flag-gated creation/nav PRs later in the stack. ## Additional context Stacked on the query-source series; base branch is `master` now that PR 4 (log date range domain + session state, #48401) is merged. Follow-ups in the stack add the toolbar/creation UI (with a run-affordance gate on `otelLegacyLogs`), nav section, AI dialect support, and reports guard. <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Added support for running log queries directly from the SQL editor. * Log query results, errors, and time ranges are now handled within the editor session. * Added automatic selection between database and log query execution, including support for custom date ranges. * SQL assistance is disabled while editing log queries where database definitions do not apply. * **Tests** * Added coverage for log query execution, date ranges, feature availability, and execution source selection. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
d5436ae826 |
feat(studio): log date range domain + session logRange state (#48401)
## 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 (+ a small refactor and a docs/convention note). PR 4 of the stacked SQL-editor query-source series (Database vs Logs). ## What is the current behavior? The SQL editor has no representation of a logs query's time range: `querySource.ts` only knows how to map a snippet type to a source (`getSnippetSource`), and session state (`sql-editor-session-state.ts`) tracks results and the row limit but not a per-snippet time range. The Logs date picker's pure range helpers (`parseCustomInput`, `generateDynamicHelper`, the `Unit` type) are trapped inside the `Logs.DatePickers.tsx` React component. ## What is the new behavior? - **Logs time-range domain** in `querySource.ts`: branded `IsoDateTimeString` + `isoDateTimeString()`, `RelativeTimeUnit`, a `LogDateRange` discriminated union (relative/absolute), `DEFAULT_LOG_DATE_RANGE`, a single date-picker parser (`datePickerValueToLogDateRange` / `logDateRangeToDatePickerValue` — handles the five presets *and* dynamic `2h`/`30m` helpers; `calcTo === ''` means "now"; unparseable helpers degrade to absolute), and `resolveLogRunRange` which re-resolves relative ranges against `now` at run time (reusing the existing `ResolvedLogDateRange` shape). - **Session state**: per-snippet `logRange` + `setLogRange` — session-only, never written to snippet content, so it works on read-only shared snippets and is cleaned up in `clearForSnippet`. - **Refactor**: extracted the picker's framework-free helpers into a new pure `Logs.datePickerHelpers.ts`; the logs domain now shares the `Unit` type and reuses `generateDynamicHelper` instead of duplicating them. Importers point at the new module directly (no re-export shim). Hardened the amount parse against `NaN`. - **Full unit coverage** in `querySource.test.ts`. Recorded the no-shim refactoring convention in the `studio-best-practices` skill. Verification: `pnpm typecheck` clean, lint ratchet improved, 43 tests pass (querySource + Logs.Datepickers), Prettier clean. ## Additional context <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit - **New Features** - Added robust Logs date-range modeling with support for relative (e.g., last N units) and absolute time periods. - SQL Editor sessions now remember log date ranges per snippet. - **Bug Fixes** - Safer handling of invalid or missing date inputs, with sensible fallback to default/current time. - **Tests** - Added/expanded automated coverage for date-range conversion, helper parsing, and resolution behavior. - **Refactor** - Centralized date-picker helper utilities for reuse across the Logs and SQL query experience. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
fa5eb17277 |
feat(studio): discriminated snippet union + source-aware writes (#48313)
Stacked on #48305. ## What PR 3 of the stacked SQL-editor query-source series (Database vs Logs). Stacked on the PR 2 branch `charislam/log-sql-content-shape`. Turns `SnippetWithContent` into a discriminated union on `type` and makes all snippet writes source-aware: - `data/content/sql-folders-query.ts`: `SnippetWithContent` is now `{ type: 'sql'; content?: SqlSnippets.Content } | { type: 'log_sql'; content?: LogSqlSnippets.Content } | { type: 'report'; content?: never }`. `report` is kept (the content endpoints' wire type carries it) but has no SQL content — its body is `Dashboards.Content`, loaded through the separate `Content` union. - `setSql` brands per type (`untrustedLogSql` vs `untrustedSql`). - `buildUpsertPayload` persists `snippet.type` (no longer hardcoded `'sql'`). - `createSqlSnippetSkeletonV2({ source })` emits the matching type + content shape with the `as any` cast removed. - New `components/interfaces/SQLEditor/querySource.ts`: `SqlSnippetSource` + `getSnippetSource`. - `seedSnippet` test helper gains a `source` arg. - New `remapWireSnippet` boundary helper in `content-remap.ts` concentrates the single wire->domain assertion, so `content-id-query` / `content-upsert-mutation` call sites are cast-free (no `as unknown as`). - Collateral: query result types aligned to the union; `updateSnippet` no longer accepts `type` (source is immutable); db-only editor read paths narrow away `log_sql`. ## Why Impossible-states-impossible typing: a snippet's brand follows its content type, so logs SQL and database SQL can never cross execution paths. No behavior change for existing database snippets. ## Testing - \`pnpm typecheck\` — clean - \`pnpm --filter studio run lint:ratchet\` — no new warnings - \`pnpm test:studio\` (data/content, SQLEditor, state/sql-editor) — passing, including new tests for \`getSnippetSource\`, source-aware \`setSql\`, type-aware \`buildUpsertPayload\`, and both skeleton shapes. <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Added source-aware creation for SQL editor snippets, including log-based SQL snippets. * Introduced backend source mapping so log snippets are treated as log_sql. * **Bug Fixes** * Improved SQL retrieval/prettification so log snippets no longer use the wrong fallback content. * Ensured log snippets are sanitized and preserve correct type, content, identifiers, and statuses during save/upsert flows. * **Tests** * Expanded unit and integration coverage for log snippet creation, source mapping, editing, prettification, and upsert payloads. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
7743fee3ab |
feat(studio): log_sql content shape + remap content.sql to unchecked_sql (#48305)
## What PR **2 of 9** in the SQL-editor query-source (Database vs Logs) stack. **Base:** `charislam/snippet-source-typing` (#48301) — this is a stacked PR; review/merge that one first. Client-side rename only — **the wire format is unchanged** (the platform API still stores and returns `content.sql`). This moves the frontend `LogSqlSnippets.Content` field to the branded `unchecked_sql`, matching `SqlSnippets.Content`, and hardens the remap boundary so the rename can't silently drop saved query text. ## Changes - **`types/userContent.ts`** — `LogSqlSnippets.Content`'s plain `sql: string` becomes `unchecked_sql: UntrustedLogSqlFragment` (the brand added in PR 1). Shape kept minimal: `{ content_id, unchecked_sql, schema_version }`. - **`data/content/content-remap.ts`** — extend `remapSqlContentField`/`unmapSqlContentField` to `log_sql`, branding **per type** (`untrustedLogSql` for logs, `untrustedSql` for database) and never mixing brands. **Defensive unmap**: content missing `unchecked_sql` is never clobbered with `sql: undefined`; a residual raw `sql` field (a missed save-path rename) throws in development to surface the bug loudly, while production no-ops safely. - **Legacy Logs Explorer consumers** updated to the branded field: the explorer save/update paths, `SavedQueriesItem`, `RecentQueriesItem`, and the recent-queries page. - **Two db-only write sites** that leaned on `LogSqlSnippets.Content.sql`: `EditorPanel` now saves `unchecked_sql`, and `MoveQueryModal` switches to the SQL-editor-specific `getSqlSnippetById` so its content is typed as `SqlSnippets.Content` — no narrowing or casting. ## Tests - **content-remap**: `log_sql` remap/unmap round-trip with the logs brand; the defensive-unmap no-op (prod) and dev throw. - **content-upsert-mutation**: a `log_sql` payload reaches the wire as a plain `content.sql` and the response remaps back to `unchecked_sql` (the data-loss-critical round-trip shared by both explorer save-new and `SavedQueriesItem` update). ## Verification - `pnpm --filter studio run typecheck` ✓ - `pnpm --filter studio run lint:ratchet` ✓ (no new warnings) - `pnpm test:studio` for `data/content` + `Settings/Logs` — 139 passing ✓ - Prettier ✓ Nothing is user-visible yet — logs snippet entry points arrive later in the stack behind the `sqlEditorLogsSource` flag. <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit - **Bug Fixes** - Improved handling of saved and recent log queries across the SQL editor and Logs Explorer. - Log SQL now uses `unchecked_sql` (branded as untrusted) consistently when creating, editing, moving, and reopening queries, with correct remapping to/from the API boundary. - Fixed saved-query update payloads to preserve the right query content and omit legacy fields. - **Tests** - Added/expanded Vitest coverage for saved log query editing, recent-log normalization, and `log_sql` remap/upsert request/response behavior. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
ec1c889349 |
feat(studio): logs SQL brands + execution data layer (#48301)
## 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 (data layer only — PR 1 of the SQL-editor query-source stack; nothing user-visible yet, no consumers). ## What is the current behavior? The Studio SQL editor only runs queries against Postgres. There is no type-safe brand for user-authored logs SQL and no execution/normalization layer for running SQL against the logs/analytics (ClickHouse) backend. ## What is the new behavior? Pure additions, no behavior change: - `data/logs/safe-analytics-sql.ts` — adds distinct untrusted/safe brands for user-authored logs SQL (`UntrustedLogSqlFragment`, `untrustedLogSql`, `acceptUntrustedLogsSql`), mirroring pg-meta's `UntrustedSqlFragment` but kept intentionally disjoint so Postgres and logs SQL can never cross boundaries. - `data/logs/execute-logs-sql-mutation.ts` (new) — `executeLogsSql` wraps `executeAnalyticsSql`, attaches the resolved time range as request params (`iso_timestamp_start/end`, never spliced into SQL), and normalizes to `{ rows, error? }`; `mapLogsError` normalizes the analytics backend's structured 200-body error into the `{ message }` shape the result pane reads; `useExecuteLogsSqlMutation` collapses transport and 200-body errors into React Query's single `onError` path. - Unit tests for `mapLogsError`, the brands (including compile-time disjointness vs pg-meta brands), and safe composition. Verification: `pnpm test:studio` (new suites, 26 passed), `pnpm typecheck`, `lint:ratchet` (no new warnings), and Prettier all pass. ## Additional context <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Added the ability to run user-authored logs SQL with resolved start/end timestamps. * Normalized query error handling so failures surface a clear message (including sensible fallbacks) and integrates with mutation error flows (with a default error toast when not customized). * Introduced safety branding for logs SQL fragments, including promotion to runnable safe SQL. * **Tests** * Added tests covering error normalization across multiple malformed/empty error shapes. * Added tests ensuring logs SQL branding preserves/accepts only the intended types and rejects unsafe inputs. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
0bef8e7d90 |
test(sql-editor): e2e coverage + delete jsdom test + merge Results.utils tests (Steps 5-6) (#48217)
## Summary Steps 5 and 6 of the SQL editor test refactor plan (the final two steps). **Step 5** — extends `e2e/studio/features/sql-editor.spec.ts` (real browser, zero mocks) with cases that need the real Monaco editor / full app render: - destructive-query warning modal: confirm actually re-runs the forced query (previously only `Cancel` was exercised) - debug button opens the AI Assistant with the query error pre-filled Deletes `apps/studio/tests/components/SQLEditor/SQLEditor.test.tsx` — its logic-level cases are now covered mock-free by the Step 4 hook tests, and its integration cases by e2e. Deleting rather than narrowing is the honest consequence of "no mocking": every remaining assertion it could make in jsdom requires a Monaco mock. **Step 6** — merges `apps/studio/tests/components/SQLEditor/Results.utils.test.ts` (`formatClipboardValue`/`formatCellValue`) into the colocated `apps/studio/components/interfaces/SQLEditor/UtilityPanel/Results.utils.test.ts` (`formatResults`/`convertResultsToMarkdown`/`convertResultsToJSON`/`getResultsHeaders`/`isLargeValue`/`convertResultsToCSV`) — both tested disjoint exports of the same source file. Deletes the `tests/` copy. This is the last step in the plan. ## Test plan - [x] `pnpm --filter studio typecheck` — no new errors in changed files - [x] `npx prettier --check` on all changed files - [x] Ran the new/changed e2e cases locally end-to-end against a live local stack — both pass - [x] `cd apps/studio && npx vitest run components/interfaces/SQLEditor/UtilityPanel/Results.utils.test.ts` — 42/42 passing after the merge <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Tests** * Added end-to-end coverage for destructive SQL query warning modal flow before forced execution. * Added end-to-end coverage for the AI Assistant debug flow when SQL execution fails. * Expanded unit test coverage for SQL editor results formatting utilities (clipboard and cell value formatting). * Removed the prior SQLEditor unit test suite and the older results-formatting unit tests. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
08c4f64c42 |
test(sql-editor): add mock-free hook tests (Step 4) (#48214)
## What Step 4 of the SQL editor testability plan: **mock-free hook tests** for the extracted SQL editor hooks, built on the Step 3 renderHook harness (`tests/lib/sql-editor-test-utils.tsx`) — in-memory editor port + real valtio stores + MSW. **Zero `vi.mock`.** | File | Tests | Covers | |------|-------|--------| | `useSqlEditorExecution.test.tsx` | 8 | destructive-query gating (`potentialIssues` vs. forced run), auto-limit suffixing, connection-string → `x-connection-encrypted` header, `onSuccess`/`onError` session-store writes, error-line highlight, diff-open short-circuit | | `useSqlEditorAi.test.tsx` | 7 | one-shot diff-request drain (empty vs. non-empty editor), drain-exactly-once across remounts, accept/discard diff, `onDebug` opening the assistant chat + debug prompt | | `usePrettifyQuery.test.tsx` | 2 | in-place format + write-back, diff-open no-op | | `useSnippetIdentity.test.tsx` | 2 | generated identity + store-driven loading state | | `useSnippetTitleGenerator.test.tsx` | 2 | untitled-snippet naming via the title endpoint | Every test exercises real dependencies at the seam where they're real: network via MSW, stores used real and reset per test, Monaco via the in-memory editor port. ## Test plan - [x] `pnpm test:studio -- SQLEditor` → **286/286 passing** (21 new tests included) - [x] `pnpm --filter studio typecheck` clean - [x] Confirmed zero `vi.mock` in the new files <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Tests** * Added comprehensive automated coverage for SQL query formatting, snippet identity, and AI-generated titles. * Added coverage for AI-assisted SQL editing, including diff acceptance, rejection, debugging, and request handling. * Added coverage for query execution, result persistence, safety checks, replica selection, error highlighting, and diff-state behavior. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
4d793a708e |
test(sql-editor): shared renderHook harness with in-memory editor port (#48209)
## Summary - Add `renderSqlEditorHook()` test harness that eliminates mocking Monaco by injecting a real, deterministic in-memory editor port (EditorController/DiffController backed by plain JS state) - Include `createInMemoryEditor()`, `resetSqlEditorStores()`, and `setupSqlEditorMocks()` utilities to provide isolation and mock-free network testing via MSW handlers - Export `CustomWrapper` from custom-render and add optional `editor`/`diff` injection points to SQLEditorProvider (production unaffected via null-coalesce fallback) This is **Step 3** of an in-progress SQL editor testability refactor (Step 2 finished EditorController/DiffController port; this harness has no consumers yet — hook tests land in a follow-up step). ## Test plan - [x] `pnpm --filter studio typecheck` passes (already verified) <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Tests** * Added reusable SQL Editor test utilities for in-memory editing, selections, error highlighting, snippets, and diff content. * Added helpers for resetting editor state, configuring API mocks, and rendering SQL Editor hooks in a complete test environment. * Enabled SQL Editor providers to accept optional controller overrides for isolated testing. * Exported the shared test wrapper for reuse across test suites. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
3d83e026f9 |
refactor(sql-editor): finish EditorController/DiffController port (Step 2) (#48166)
## Summary Step 2 of the SQL Editor testability plan. `SQLEditorContext` already wrapped the Monaco refs and exposed a few semantic imperative helpers (`getEditorSql`, `clearHighlights`, `applyErrorHighlight`, `refocusEditor`, …). This finishes that abstraction so no hook or controller touches `editorRef.current`/`diffEditorRef.current` directly anymore — they only call the port. The port is what will let Step 3's test harness inject a real in-memory editor adapter instead of mocking Monaco; production wires it to the real Monaco refs, unchanged. - Extends the context value with two semantic controllers, backed by the existing refs: - `editor: EditorController` — `isReady`, `getValue`, `getSelectionStartLine`, `getSql` (today's `getEditorSql`), `replaceAll` (wraps the repeated `executeEdits(...)` pattern), `focus`, `revealLineInCenter`, `highlightErrorLine` (today's `applyErrorHighlight`), `clearHighlights`. - `diff: DiffController` — `isMounted`, `getModifiedValue`, `setDiff` (the diff-sync effect body), `attach` (today's `handleDiffEditorMount`). - Migrates every touch point off raw refs onto the port: `useSqlEditorExecution`, `usePrettifyQuery`, `useSqlEditorShortcuts`, `SQLEditorControllers`' `readEditorSql`, and `useSqlEditorAi`'s `acceptAiHandler`/`drainDiffRequest`/`handleDiffEditorMount`/diff-sync effect. - `SQLEditorEditorPanel.tsx` is intentionally left untouched — it wires the raw refs into the real Monaco/DiffEditor React components for rendering, which isn't decision logic to abstract. Behavior-preserving. ## Test plan - [x] `pnpm --filter studio typecheck` - [x] `pnpm test:studio -- SQLEditor` (265 tests passing) - [x] `pnpm --filter studio run lint:ratchet` |
||
|
Charis
|
cdc843dadd |
refactor(sql-editor): extract deriveSnippetIdentity, debug/completion/diff-key helpers (#48014)
## Summary Pure-fn extraction pass across the SQL editor hooks. - Extracts `deriveSnippetIdentity` out of `useSnippetIdentity`'s inline id + `isLoading` derivation into `SQLEditor.utils.ts`. - Extracts `extractDebugContext` (shared snippet/result/error extraction) and `buildDebugChatArgs` (the `aiSnap.newChat(...)` payload builder) out of `useSqlEditorAi`'s `buildDebugPrompt`/`onDebug` into `SQLEditor.utils.ts`. - Extracts `buildCompletionRequestBody` (the AI completion endpoint's request body builder) and `planDiffRequestApplication` (the pending-diff-request application decision: replace vs. open a diff, depending on whether the editor is currently empty) out of `useSqlEditorAi` into `SQLEditor.utils.ts`. The `drainDiffRequest` effect now just applies the plan instead of branching inline. - Extracts `resolveDiffKeyAction` out of `useSqlEditorShortcuts`'s window-keydown Enter/Escape branch into `SQLEditor.utils.ts`. ## Test plan - [x] `pnpm --filter studio typecheck` - [x] `pnpm test:studio -- SQLEditor` (265 tests passing) - [x] `pnpm --filter studio run lint:ratchet` |
||
|
Charis
|
1c827c5cbb |
refactor(sql-editor): extract title-gen/execute-params + merge auto-limit functions (#48013)
## Summary Part 2/6 of the SQL Editor testability follow-up, stacked on #47980 (the analyzeQueryIssues/resolveConnectionString PR). - Extracts `shouldAutoGenerateTitle` and `buildExecuteParams` out of `useSqlEditorExecution`'s inline logic into `SQLEditor.utils.ts`. - Merges `checkIfAppendLimitRequired` and `suffixWithLimit` into a single `applyAutoLimit` function — the two were only ever called together and re-parsed the same query twice at every call site. `applyAutoLimit` only accepts `SafeSqlFragment` (never a plain string) and composes the `LIMIT` suffix through `safeSql`/`literal` rather than raw template concatenation, so the only place in the file that reasserts the `SafeSqlFragment` brand on a derived string is the small, dedicated `trimTrailingSemicolons` helper — removing existing terminators can't introduce unsafe content, unlike gluing new text onto the fragment. - Updates the two other `checkIfAppendLimitRequired`/`suffixWithLimit` call sites (`EditorPanel.tsx`, `ReportBlock.tsx`) accordingly; `ReportBlock` now promotes its report SQL once and reuses the result for both its display-only auto-limit hint and its execution, instead of promoting twice. ## Test plan - [x] `pnpm --filter studio typecheck` - [x] `pnpm test:studio -- SQLEditor ReportBlock EditorPanel` (239 tests passing) |
||
|
Charis
|
11fb715149 |
refactor(sql-editor): extract query-issue analysis + connection string resolution (#47980)
## Summary Part 1/6 of the SQL Editor testability follow-up (extracting pure decision logic out of the SQL editor hooks so it can be exhaustively unit-tested without mocking). - Extracts `analyzeQueryIssues` and `hasBlockingIssues` out of `useSqlEditorExecution`'s inline destructive-query / warning-modal-gating logic into `SQLEditor.utils.ts`. - Extracts `resolveConnectionString`, deduping the `databases?.find(...)` lookup that was duplicated verbatim in both `useSqlEditorExecution` and `useSqlEditorExplain`. - Behavior-preserving — same runtime logic, now unit-testable in isolation. ## Test plan - [x] `pnpm --filter studio typecheck` - [x] `pnpm test:studio -- SQLEditor.utils` (159 tests passing, includes new exhaustive-permutation cases for the three extracted functions) <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Bug Fixes** * Improved SQL safety checks before execution, including destructive queries, unsafe updates, database-altering commands, and tables missing row-level security. * Correctly recognizes tables protected by active security triggers. * Improved database connection selection for primary and read-replica databases. * Preserved the ability to run queries with force enabled when safety warnings are present. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
4df3c86e93 |
feat: add admin studio link (#47976)
Add link to Admin Studio in top bar for internal use only. Resolves FE-3864 <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Added a conditional Admin Studio shortcut to the desktop project header. * The shortcut opens the relevant project in Admin Studio and is available only when configured and enabled. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
9e80a159b5 |
refactor(sql-editor): decompose into controller contexts + presentational components (decompose 6/6) (#47938)
## What Final step (**6 of 6**) of the `SQLEditor.tsx` decomposition. Splits the two `ResizablePanel` bodies out of the `SQLEditorContent` monolith into presentational sub-components: - **`SQLEditorPane`** — the editor panel: loading state, `DiffEditor` + diff ask-AI widget, `MonacoEditor` + ask-AI widget. Reads the Monaco refs (`editorRef`/`monacoRef`/`diffEditorRef`) from `SQLEditorContext`, and receives the `diff`/`prompt`/`ai` controllers plus reactive state as props. - **`SQLEditorResults`** — the results panel: loading state + `UtilityPanel`. `SQLEditorContent` is now a composition root: shared-ref provider, hook composition, and the run-query warning modal. ## Result `SQLEditor.tsx` goes from the original **1056-line** monolith down to **259 lines** (≈75% reduction). The remaining size over a bare ~140-line root is the run-query warning modal, kept inline **deliberately**: its handlers hold the `acceptUntrustedSql` promotion, which must stay at the explicit user-action boundary in the root rather than moving into a presentational pane. ## Behavior-preserving The moved JSX is byte-identical aside from prop threading. No logic, effects, dependency arrays, or `eslint-disable`s changed. The two **render-time ref reads** — the editor placeholder (`!promptState.isOpen && !editorRef.current?.getValue()`) and the ask-AI widget gate (`editorRef.current && promptState.isOpen && !isDiffOpen`) — are preserved verbatim in `SQLEditorPane`, which re-renders whenever the `prompt`/`diff` props change, keeping those reads fresh (the guardrail from the plan). Verification (all green): - `SQLEditor.test.tsx` characterization suite — 11/11 pass - `tsc --noEmit` — no new errors - eslint — clean (no new ratchet entries) - prettier — clean ## Stack Builds on decompose 5 (#47935). This is the last PR in the series — the decomposition is complete after this merges. <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Added a resizable SQL editor layout with run warnings, query results, and an Explain view. * Introduced centralized SQL editor controllers and expanded AI-assisted prompt/diff workflows. * Added keyboard support for running and Explain analysis. * **Bug Fixes** * Restored editor focus after accepting or discarding AI changes. * Improved handling and validation of SQL Explain actions. * Preserved editor scroll position when switching snippets. * **Refactor** * Streamlined the SQLEditor into a composed layout and improved memoization to reduce unnecessary re-renders. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
52a25c2ebb |
refactor(sql-editor): extract AI/diff + shortcuts hooks (decompose 5/6) (#47935)
## What Decompose step **5 of 6** for `SQLEditor.tsx`. Extracts the Assistant / diff cluster and the keyboard-shortcut wiring out of the `SQLEditorContent` monolith into two focused, individually-testable hooks: - **`useSqlEditorAi`** — SQL completion (`complete`), the ask-AI prompt flow (`handlePrompt`), accept/discard diff handlers, `onDebug` / `buildDebugPrompt` helpers, `handleDiffEditorMount`, and the fragile diff lifecycle effects (one-shot diff-request drain, diff-editor value sync, ask-AI widget visibility). - **`useSqlEditorShortcuts`** — the registered shortcuts (focus editor, new snippet, format, explain) plus the accept/discard/escape keydown handling. `SQLEditorContent` now composes these hooks alongside the execution/explain hooks landed in decompose 4. ## Behavior-preserving This is a pure extraction. The moved function bodies, effect logic, dependency arrays, and JSX are unchanged from the previous monolith (verified via `git diff` against the pre-decomposition source). In particular: - `useEffectEvent` is preserved for `drainDiffRequest` / `resetDiff`. - `editorMountCount` remains single-owner (passed into the AI hook to drive the one-shot drain). - The untrusted→safe SQL promotion (`acceptUntrustedSql`) continues to happen in the run/explain gesture and warning-modal handlers in `SQLEditorContent`, as close to the explicit user action as possible. The Phase-1 characterization suite (`SQLEditor.test.tsx`, 11 tests) remains green. ## Stack Part of the SQLEditor decomposition stack (1/6 … 6/6). Builds on decompose 4 (execution + explain hooks, #47923). Next: PR6 splits the JSX into panes + final cleanup. <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Improved SQL editor AI assistance, including completion prompts, debugging support, and diff review controls. * Added keyboard shortcuts for accepting or discarding AI-generated SQL changes. * Added shortcuts for focusing the editor, creating snippets, formatting queries, and explaining SQL. * **Bug Fixes** * Prevented SQL execution while reviewing AI-generated differences. * Improved handling of AI diff state during editor loading and interaction. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
7832d9f716 |
chore: dependabot autobump gha major versions (#47924)
We were ignoring major version updates for Dependabot updates of GitHub Actions, so we are now several major versions behind on several actions. Turning major version updates back on so we can update everything. <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Chores** * Dependency update configuration now includes major version updates. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |
||
|
Charis
|
3fa7f086df |
refactor(sql-editor): extract execution + explain hooks (decompose 4/6) (#47923)
## Summary PR **4 of 6** in the SQLEditor decomposition stack. Extracts the two query-run concerns out of the composition root into cohesive hooks. Behavior-preserving — the characterization suite stays green. ## New hooks - `useSqlEditorExecution` — the execute mutation, the `executeQuery` pipeline (destructive-query gating, lazy title generation, connection resolution), and the `potentialIssues` warning-modal state. - `useSqlEditorExplain` — the execute-explain mutation and the `executeExplainQuery` pipeline. Both pipelines accept an **already-promoted `SafeSqlFragment`**. ## Safe-SQL boundary Per review guidance, `acceptUntrustedSql` promotion stays in the **composition root's** run/explain gesture handlers and warning-modal confirm handlers — as close to the explicit user action as possible, so it's auditable. The hooks never call `acceptUntrustedSql`; they only accept safe SQL. `executeQuery` lists its (now all-stable) dependencies directly rather than suppressing `react-hooks/exhaustive-deps`. ## Verification - `vitest` — 11 characterization tests pass - `pnpm --filter studio typecheck` — clean for SQL editor files - `eslint` / `lint:ratchet` — pass (0 new violations) - `prettier --check` — clean `SQLEditorContent` is down to ~530 lines (from 1056 at the start of the stack). Remaining: PR5 (AI/diff + shortcuts), PR6 (JSX pane split + cleanup). <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Bug Fixes** * Improved SQL query execution safeguards for potentially destructive statements, unsafe updates, and missing row-level security coverage. * Improved EXPLAIN handling, including clearer validation for unsupported multi-statement queries. * Preserved query results, errors, and EXPLAIN output more reliably in the SQL editor. * **Improvements** * Added clearer execution status handling and more consistent editor focus, highlighting, and utility-tab behavior. * Improved connection selection and SQL execution reliability across supported database configurations. <!-- end of auto-generated comment: release notes by coderabbit.ai --> |