dima/supabase
1
0 Fork 0
mirror of https://github.com/supabase/supabase.git synced 2026-10-08 19:05:06 +03:00
Code Issues Packages Projects Releases Wiki Activity
37,363 Commits 1,791 Branches 62 Tags
cli/docs-example-how-to
Commit Graph
4814 Commits
Author SHA1 Message Date
Jordi EnricandClaude Opus 4.8 4096267623 feat(api-keys): migrate last-used indicator to ClickHouse endpoint (#47458) 
## Problem

The "last used" indicator for the legacy `anon` / `service_role` API
keys (Project API keys settings) was disabled because it ran a BigQuery
`edge_logs` query. It is now re-enabled against the ClickHouse-backed
`api_keys.last_used.otel` analytics endpoint.

## Current behavior

- The `anon` / `service_role` "last used" indicator is off (the
BigQuery-backed query was disabled).

## New behavior

- New `useApiKeysLastUsedQuery` hook calls the `api_keys.last_used.otel`
endpoint (timestamp params only, no SQL sent), plus its query key and
the generated platform API type.
- `DisplayApiSettings` reads last-used from this hook instead of posting
BigQuery `edge_logs` SQL. The pure `getLastUsedAPIKeys` shaper is kept
and unit-tested. Still gated by the `showApiKeysLastUsed` flag.
- Removed the disabled secret-keys (`sb_secret_`) BigQuery last-used
path, which has no ClickHouse endpoint to migrate to: drops the dead
`useLastSeen` query, the `APIKeyRow` "Last Used" column, and the unused
`showLastSeen` prop.
- Reworded the delete-confirmation copy to be accurate for both secret
and publishable keys.

## Additional context

- Backed by the platform endpoint in supabase/platform#34892 (merged and
deployed).
- Scope: `anon` / `service_role` legacy keys. Secret/publishable and JWT
signing-key "last used" are follow-ups, pending the endpoint returning
those key types.

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

* **Improvements**
* Updated API key settings to show “last used” activity for the past 24
hours using a dedicated data source and time window.
  * Added clearer messaging when recent API key activity fails to load.
  * Removed the “Last Used” column from API key management tables.
* **Bug Fixes**
* Improved mapping so “last used” values correctly match the intended
key and role.
* Updated API key deletion confirmation to explain required backend
changes and resulting unauthorized behavior.
* **Tests**
* Added unit tests to validate “last used” computation and edge-case
filtering.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->

---------

Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
 2026-07-14 14:35:32 +02:00
Andrew ValleteauandClaude Fable 5 768ea1001b fix(studio): scope table editor introspection CTEs to target table OID (#47894) 
## 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 (performance), plus a regression-guard test suite and docs.

## What is the current behavior?

Studio's introspection queries in `@supabase/pg-meta` do `O(catalog)`
work for per-table requests. On databases with very large catalogs
(hundreds of thousands of relations/constraints — real deployments reach
this) they take tens of seconds per dashboard interaction, trip
`statement_timeout`, and create heavy CPU/memory pressure when several
tabs open concurrently. Two instances of the same bug class:

**1. Table Editor query (`getTableEditorSql`)** — fetches metadata for
ONE table by OID, but five catalog scans are unscoped and only filtered
at the top-level join:

- `primary_keys` CTE — scans all of `pg_index` (`where i.indisprimary`)
- `index_cols` CTE — scans all unique indexes
- `relationships` CTE — scans every FK in `pg_constraint` (and is
scanned twice by the two subplans)
- `uniques` subquery (inside `columns`) — scans all single-column unique
constraints
- `check_constraints` subquery (inside `columns`) — scans all
single-column check constraints

The planner cannot push the outer join qual into grouped / `distinct on`
subqueries, so each is computed over the full catalog and thrown away.
`tables-paginated.ts` was previously rewritten to avoid exactly this
pattern; the single-table query never got the same treatment.

**2. Entity definitions (`getTableDefinitionSql` /
`getEntityDefinitionsSql`)** — the vendored `pg_get_tabledef` plpgsql
function scans the entire `information_schema.columns` view once **per
column** (plus `information_schema.tables` once per call) just to decide
whether a name needs double-quoting — a pure string property of a name
it already holds — and its per-index partial-index lookup casts
`relnamespace::regnamespace::text` across every `pg_class` row. On a
12K-table catalog this makes a single entity's DDL cost ~3.7s and a
default 100-entity definitions page ~6 minutes.

## What is the new behavior?

**Fix 1 — scope the Table Editor CTEs to the requested OID** (`id` is
validated non-null and interpolated via `literal()`, same as the
existing `base_table_info` filter):

- `primary_keys` / `index_cols`: `and i.indrelid = <id>`
- `relationships`: `and (c.conrelid = <id> or c.confrelid = <id>)`
- `uniques` / `check_constraints`: `and conrelid = <id>`

Semantics are unchanged: the top-level select already filtered every CTE
to the target table, so rows for other tables were computed and
discarded. The `pg_index`/`pg_constraint` lookups become index scans
returning a handful of rows. One residual scan is structural: PostgreSQL
has no index on `pg_constraint.confrelid`, so the incoming-FK half of
`relationships` is a single filtered seq scan of `pg_constraint` — still
one cheap pass instead of materializing every FK row twice.

**Fix 2 — remove the O(catalog) scans inside `pg_get_tabledef`**: the
information_schema uppercase checks are replaced with direct regex tests
on the name in hand (preserving the original's `quote_ident` behavior
for schemas that need quoting), and the partial-index lookup is scoped
by the already-resolved table OID. Original statements are kept as
comments, matching the vendored file's convention.

**Regression guard** — so this bug class stays out:

- `test/db/stress-catalog.ts` builds a synthetic catalog (default 2,000
tables with PKs, unique + check constraints, FK chains and an FK hub;
`PG_META_STRESS_TABLES` scales it to incident size).
- `test/db/plan-guard.ts` provides `EXPLAIN (ANALYZE, FORMAT
JSON)`-based budget assertions: a query's plan may only seq-scan a
scaling catalog if its budget entry carries a written structural
justification (e.g. no index on `pg_constraint.confrelid`; no index on
`pg_class.relnamespace` for per-schema listings), plus a per-query time
bound (the only guard available for opaque plpgsql internals like
`pg_get_tabledef`).
- `test/sql/studio/catalog-plan-guard.test.ts` applies budgets to the
hot-path studio queries: table editor, constraints, FK listing, entity
types, tables-paginated, columns, indexes, table/entity definitions,
views. Reverting either fix makes the suite fail immediately with the
offending scans listed.
- `test/sql/studio/table-editor.test.ts` (new — none existed) asserts
the Table Editor query's semantics: primary keys, unique indexes, both
FK directions, `is_unique`, check definitions, column comments.
- A new package `README.md` documents the plan-guard budget entry as a
requirement for any new introspection query.

### Validation (synthetic 12,000-table catalog, PostgreSQL 17.6)

- **Output equivalence, fix 1:** for 12 relation types (regular,
composite PK, partitioned parent + partition, view, materialized view,
constraint-free table, FK hub/chain/tail, and a fixture with
enums/domains/generated/identity columns and duplicate check
constraints), the `entity` jsonb from the old and new query is
byte-identical.
- **Output equivalence, fix 2:** byte-identical DDL across 13 fixture
combinations (serial/identity/generated/array columns, case-sensitive
and keyword names, mixed-case schemas, partitions, unlogged +
reloptions, partial/expression indexes, external PK/FK/comments/trigger
variants).
- **Performance, fix 1:** Table Editor query `EXPLAIN ANALYZE` ~1,630ms
→ ~30ms (~50×); the gap grows with catalog size since the old query is
O(catalog) per call.
- **Performance, fix 2:** single entity definition 3,672ms → 63ms; a
100-entity definitions page ~6min → 0.87s. The plan-guard bound for
`getEntityDefinitionsSql` tightens accordingly from 15s/25 entities to
3s/100 entities (330ms measured at default test scale).

Verified locally: `catalog-plan-guard` (12 tests), `table-editor`,
`tables-paginated` (16 tests) pass; `typecheck` clean.

### Rollout

Per review, the new behavior ships **dark** behind the
`pgMetaScopedIntrospection` ConfigCat flag (default off = legacy SQL,
kept as full duplicated templates in pg-meta and verified byte-identical
to the pre-PR queries). Studio reads the flag in the query hooks and
threads it through (flag state is part of the React Query keys). The
rollout is staged in the ConfigCat dashboard via user-email targeting
(like every other ConfigCat flag): target the reporting user's email
first, then a percentage rollout, then 100%. Server-side AI callers of
`getEntityDefinitionsSql` stay on the legacy path. Once fully rolled
out, delete the legacy templates + flag in a cleanup PR.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Closes: PGMETA-122

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

- **Bug Fixes**
- Improved table editor SQL to correctly scope primary keys, indexes,
uniques, checks, and relationships to the selected table.
- Optimized table definition SQL to reduce unnecessary catalog scanning
for uppercase-name detection and partial-index detection.

- **Tests**
- Added SQL generator tests for table editor metadata (keys, indexes,
relationships, comments, and constraints).
- Added catalog query plan guard coverage with a stress catalog and
EXPLAIN-based scoping/performance budgets.

- **Documentation**
- Expanded documentation on catalog query plan safeguards and how to
keep new introspection queries properly scoped.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->

---------

Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
 2026-07-14 13:16:11 +02:00
Jordi Enric b84aafd1a6 fix(ui-patterns): fix chart y-axis label clipping (#47890) 
## Summary
- Chart y-axis tick labels were clipped (e.g. edge function overview
execution time charts) because `chart-line.tsx`/`chart-bar.tsx`
hardcoded a `-40` left margin regardless of the actual
`YAxisProps.width` passed in.
- Margin now scales with the configured axis width.

## Test plan
- [ ] Visually check edge function overview performance/usage charts
render full tick labels (e.g. "195ms" instead of "ms")

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

## Summary by CodeRabbit

- **Bug Fixes**
  - Improved chart layout and alignment across line and bar charts.
- Adjusted Y-axis spacing so labels display more consistently when axes
are shown or hidden.
- Removed unnecessary fixed spacing from Edge Function performance, CPU,
and memory charts.
- Tightened spacing around chart timestamp rows for a more compact
presentation.

<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-07-14 10:05:07 +02:00
+4
Joshen LimAlaister YoungAlaister Youngkanadsupabase-supabase-autofixer[bot]github-actions[bot]Jeremias MenichelliMiranda LimonczenkoCharisGildas GarciaDavid_CAli WaseemNik RichersNik Richers
 		503bdb5a6f Fix filter button in table editor (#47867) 
## Context

Realised that the filter button in the table editor is broken so this PR
fixes it
<img width="385" height="393" alt="image"
src="https://github.com/user-attachments/assets/94332f1b-cd69-4a8c-a822-3b9096d06ee3"
/>


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

* **UI Improvements**
* Updated the entity-type filter to use a clearer, labeled “Filter
entity types” button with revised popover trigger and sizing.
* Refreshed the table header filter controls, including the entity-type
dropdown behavior and updated styling when filters are applied.
* Improved handling when no entity types are currently visible by
showing a dedicated empty-state within the filter menu.
* **New Features**
* Added a “No results based on filters” empty panel with a “Reset
filters” action to restore all entity types.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->

---------

Co-authored-by: Alaister Young <alaister@users.noreply.github.com>
Co-authored-by: Alaister Young <10985857+alaister@users.noreply.github.com>
Co-authored-by: kanad <github@kanad.dev>
Co-authored-by: supabase-supabase-autofixer[bot] <248690971+supabase-supabase-autofixer[bot]@users.noreply.github.com>
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Co-authored-by: Jeremias Menichelli <jmenichelli@gmail.com>
Co-authored-by: Miranda Limonczenko <miranda.limonczenko@supabase.io>
Co-authored-by: Charis <26616127+charislam@users.noreply.github.com>
Co-authored-by: Gildas Garcia <1122076+djhi@users.noreply.github.com>
Co-authored-by: David_C <110653663+YuDavidCao@users.noreply.github.com>
Co-authored-by: Ali Waseem <waseema393@gmail.com>
Co-authored-by: Nik Richers <nrichers@gmail.com>
Co-authored-by: Nik Richers <nik@validmind.ai>
 2026-07-14 15:15:40 +08:00
K-Dog (Kevin) 00ecb53059 feat(etl): ETL usage insights+pricing docs (#47873) 
Adds pipeline usage insights to summary and daily breakdowns + usage billing docs
 2026-07-14 14:03:39 +07:00
Joshen Lim 7f8fb85caf Joshen/fe 3879 schema not exposed warning is unnecessarily repeated in (#47868) 
Just a tiny nit i came across - realised that if the schema is not
exposed via the API, the warning that we show on the policies page RE
data not being selectable is repeated for each table which imo seems
unnecessary.

Opting for a single admonition at the top instead

## Before
<img width="1085" height="744" alt="image"
src="https://github.com/user-attachments/assets/7bd8cf7b-f5a8-47d6-b41f-13fc4782ed8b"
/>

### After
<img width="1080" height="673" alt="image"
src="https://github.com/user-attachments/assets/c839f502-42ff-4184-ad07-0ff2fd2d2676"
/>


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

* **New Features**
* Added an on-screen warning for tables whose schemas aren’t exposed via
the project Data APIs, including a link to Data API settings.
* **UI Improvements**
* Refined the “filter entity types” control’s tooltip behavior and
updated the popover header text.
* **Bug Fixes**
* Improved Data API/RLS status messaging by removing the prior “schema
not exposed” outcome and showing “unknown” when access can’t be
determined.
* Consolidated policy warning rendering to avoid duplicated or
inconsistent messages.
* **Tests**
* Updated policy/admonition helper tests to match the revised status and
message behavior.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-07-14 13:47:05 +08:00
Joshen Lim c793352036 Add some keyboard shortcuts for the Assistant (#47872) 
## Context

Part of some minor improvements to the AI Assistant - this one's about
adding some keyboard shortcuts

## Changes involved
- Added keyboard shortcut for "New chat"
<img width="212" height="96" alt="image"
src="https://github.com/user-attachments/assets/e9c3bd63-adbc-4b05-8c52-1baed67e365a"
/>
- Also added a small animation for the "How can I assist you?" text for
visual indication when moving between chats that might not have a
conversation yet
- Added keyboard shortcut for "Permission settings"
<img width="236" height="86" alt="image"
src="https://github.com/user-attachments/assets/337da009-5979-4910-9292-73cc4d7f7cce"
/>
- Show keyboard shortcut for "Close Assistant"  
<img width="165" height="88" alt="image"
src="https://github.com/user-attachments/assets/b45a4f5a-8e12-45f1-8fdb-a18c0deedb02"
/>
- Fix `ExpandingTextArea` height calculation logic issue
- If you open and close the Assistant panel a number of times, the
height of the input field isn't consistent, so this fixes that


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

## Summary by CodeRabbit

- **New Features**
- Added keyboard shortcuts for starting a new AI Assistant chat and
opening permission settings.
- Header actions now display shortcut hints and support keyboard access.

- **Improvements**
- Enhanced accessibility with labels for chat edit controls (save,
cancel, edit, delete).
  - Chat onboarding now remounts when switching active chats.
  - Improved chat popover alignment.
- Escape now blurs the message input; textarea resizing is more reliable
during content/layout changes.

- **Bug Fixes**
- Updated onboarding loading behavior based on the lints loading state.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-07-14 13:45:38 +08:00
Ali Waseem 652311feb5 fix(studio): guard Auth Performance allocation strategy select against invalid values (#47896) 
## Summary
- The Connection management "Allocation strategy" select on the Auth >
Performance page called its `onValueChange` handler's conversion logic
with whatever value it was given, with no validation. If that handler
ever fired with a value outside the `'percent' | 'connections'` enum, it
would silently overwrite a correctly loaded config, converting it to the
wrong absolute connection count and leaving the strategy dropdown in a
blank/inconsistent state.
- Extracted the percent/connections conversion into a pure, unit-tested
`convertPoolSize()` helper (`PerformanceSettingsForm.utils.ts`) and
added a guard so `onValueChange` ignores any value that isn't a
recognized allocation unit.

## How to test
1. Under **Connection management**, switch **Allocation strategy** back
and forth between "Absolute number of connections" and "Percent of max
connections" — the value should convert correctly each time and the
dropdown should never render blank.
2. Save, then hard-reload the page — the saved strategy and value should
persist as shown.

## Test plan
- [x] `PerformanceSettingsForm.utils.test.ts` — unit tests covering both
conversion directions, clamping, and the invalid-value guard
- [x] `PerformanceSettingsForm.test.tsx` — MSW-backed component test
verifying persisted percent/absolute configs render correctly on load
- [x] `pnpm test:studio`
- [x] `pnpm typecheck`

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

## Summary by CodeRabbit

* **Improvements**
* Switching database pool allocation strategies now automatically
converts values between percentage and connection-based units.
* Values are rounded and constrained appropriately to remain within
supported limits.
  * Allocation settings now handle invalid or zero values more safely.
* **Tests**
* Added coverage verifying persisted allocation strategies and pool-size
conversion behavior.

<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-07-13 14:16:52 -06:00
Charis c9a649cf73 refactor(sql-editor): extract snippet-identity/mount/prettify/title hooks (decompose 3b/6) (#47893) 
## Summary

Continues the SQLEditor decomposition. Pulls four cohesive concerns out
of the `SQLEditorContent` composition root into co-located `use*` hooks.

## New hooks

- `useSnippetIdentity` — derives `id` / `generatedNewSnippetName` /
`isLoading` from the URL + snippet store (keeps the `[urlId]` memo dep
verbatim).
- `useEditorMount` — the editor `onMount` (scroll restore/track) + the
mount counter that lets a pre-mount diff request re-run.
- `usePrettifyQuery` — formats the editor SQL in place and writes it
back to the store.
- `useSnippetTitleGenerator` — the title-generation mutation +
`setAiTitle`.
 2026-07-13 15:35:30 -04:00
David_C f2dece5b54 fix(studio): added bottom margin to schema graph's Minimap that was previously colliding with SchemaGraphLegend (#47858) 
## 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 (UI): fixes #47857 

## What is the current behavior?

SchemaGraph's Minimap component currently collide with the
SchemaGraphLegend component:
<img width="710" height="357" alt="Image"
src="https://github.com/user-attachments/assets/e6a7c30b-da00-455b-b224-e47898ada272"
/>

## What is the new behavior?

Added bottom margin, SchemaGraph's Minimap component does not collide
with the SchemaGraphLegend component anymore:
<img width="609" height="297" alt="Screenshot 2026-07-11 at 7 38 05 PM"
src="https://github.com/user-attachments/assets/6e0c5be6-ba86-40cd-b0c7-e24ca8c16f4c"
/>

## Additional context

this mb-11! is consistent with the "load more tables" button below (the
entire chunk looks like this):

```js
				<MiniMap
                    pannable
                    zoomable
                    nodeColor={miniMapNodeColor}
                    maskColor={miniMapMaskColor}
                    className="border rounded-md shadow-xs mb-11!"
                  />
                  <SchemaGraphLegend />
                  {hasNextPage && (
                    <Panel position="bottom-center" className="mb-11!">
                      <Button
                        variant="default"
                        size="tiny"
                        loading={isFetchingNextPage}
                        onClick={() => {
                          fitViewOnNextLayout.current = true
                          fetchNextPage()
                        }}
                      >
                        Load more tables
                      </Button>
                    </Panel>
                  )}
```


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

## Summary by CodeRabbit

* **Style**
* Adjusted the schema graph minimap spacing to improve layout and
visibility within the interface.

<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-07-13 17:48:16 +00:00
Gildas Garcia 2ec8e3b135 fix: ensure sidebar has valid html (#47826) 
## Problem

The sidebar in both org or project pages has invalid html, preventing
screen reader users to understand them.
See https://dequeuniversity.com/rules/axe/4.12/list

Besides, we were using some shadcn components incorrectly (groups inside
menus when it should be the other way around)

## Solution

- Invert group/menu relation so that we don't have `div` inside `ul`
- No visual changes
- No functional changes

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

* **Style**
* Improved sidebar navigation by splitting routes into clearer sections
(Home/Tools, Products, Other routes, and Settings).
* Updated the advisors link so its active indicator renders directly
within the navigation item, while keeping the existing
disabled/error/warning behavior.
* Adjusted organization navigation layout for more consistent spacing
and alignment. Animations remain smooth.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-07-13 19:09:37 +02:00
Charis 9726940dba refactor(sql-editor): own shared editor refs in SQLEditorContext (decompose 3/N) (#47837) 
## Summary

PR **3** in the SQLEditor decomposition stack. The keystone structural
step: introduce a context provider that owns the shared, mutable Monaco
state, so later PRs can extract logic hooks and presentational panes
without threading the same three refs through every signature.

Behavior-preserving — the characterization suite (merged in PR 1) stays
green with identical assertions.

## What changed

New `SQLEditorContext.tsx`:
- `SQLEditorProvider` owns the shared refs (`editorRef`, `monacoRef`,
`diffEditorRef`, `scrollTopRef`), the run-refocus flag, and the
error-highlight decorations.
- `useSQLEditorContext()` guard hook (React 19 `use()`), mirroring the
canonical `PoliciesDataContext` pattern.
- Stable helpers: `refocusEditor`, `clearPendingRunRefocus`,
`markRefocusAfterRun`, `refocusEditorAfterRunIfNeeded`, `getEditorSql`,
`clearHighlights`, `applyErrorHighlight`.

`SQLEditor.tsx`:
- Split into
`<SQLEditorProvider><SQLEditorContent/></SQLEditorProvider>`.
- `SQLEditorContent` reads refs/helpers from context instead of local
`useRef`/`useCallback`.
- `lineHighlights` moves from React state to a ref inside the provider
(decorations are purely imperative; nothing renders off them). This also
removes the stale-closure hazard the old `executeQuery` dep-array
omission worked around.

The context value holds only stable identities (refs + `useCallback`s,
memoized once), so the provider never re-renders its consumers — it's a
dependency-injection channel, not reactive state.

## Verification

- `vitest` — 156 pass (characterization + utils), assertions unchanged
- `pnpm --filter studio typecheck` — clean for all SQL editor files (the
two `@sentry/tanstackstart-react` errors are a pre-existing
local-install gap on master, unrelated)
- `eslint` — 0 errors
- Equivalence: single `useEffectEvent`, `drainDiffRequest`
driving-effect deps `[diffRequest.pending, editorMountCount]`
byte-identical, all 8 `eslint-disable` dep arrays preserved

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

* **Bug Fixes**
* Improved SQL editor focus restoration after running or explaining
queries.
* Centralized run/explain highlighting so errors are highlighted
consistently and old highlights are reliably cleared.
* Ensured the SQL snippet used for run/explain is derived consistently
from the current editor state.
* **Refactor**
* Restructured the SQL editor to use a shared internal context for
editor UI, refs, and imperative highlight/focus helpers, keeping the
external editor interface unchanged.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-07-13 13:02:03 -04:00
Joshen Lim 1d29b4c5b4 Clean up RLS Tester artifacts (#47866) 
## Context

As per PR title - we're pausing the development of the RLS Tester
feature preview while we re-evaluate its direction. Have also updated
the GH discussion
[here](https://github.com/orgs/supabase/discussions/45233) RE this! 🙏

Removes the RLS Tester UI + Sandbox functionality

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

## Summary by CodeRabbit

* **Removed Features**
* Removed the RLS Tester feature preview, banner, and database policy
testing workflow.
* The related SQL testing, role selection, policy summaries, sandbox
management, and result views are no longer available.
* **Bug Fixes**
* Improved accessibility on the database policies page by adding a label
to the clear-filter button.

<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-07-13 17:01:05 +08:00
Jordi Enric dbdbe1540b fix(studio): make paused project messaging easier to scan DEPR-581 (#47807) 
## Problem

The free-plan paused project notice was a dense paragraph, so the key
points (data is safe, resume window, download-after-expiry, upgrade)
were easy to skip.

## Fix

Present those points as a scannable bullet list, keeping the dynamic
day-count tooltip, restore-deadline timestamp, and the existing
Pro-wording variant.

Also adds a "Project Status" tab to the dev toolbar (local and staging
only) with a select for forcing the current project's status, so the
paused state and other statuses are easy to preview without touching the
backend. It overrides the status in the React Query cache and reverts on
close, project navigation, or reset.

## How to test

- Run Studio locally against the platform API
- Open the dev toolbar, go to the Project Status tab, and select
INACTIVE
- Navigate to a project page and confirm the paused screen renders
- On a free-plan org, confirm the notice now shows the details as bullet
points with the day-count tooltip and restore-deadline date intact
- Click "Reset to real data" (or close the toolbar) and confirm the
status reverts

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

* **New Features**
* Added a development toolbar “Project Status” tab to simulate project
status and pause states in non-production environments.
* Status and pause-state overrides persist locally, can be reset, and
are reflected across project detail and paused-state views.
* **Style**
* Refined paused-project messaging: updated the heading and reworked the
free-plan explanation into bullet points, while keeping paid-plan
messaging paragraph-based for readability.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-07-13 10:30:28 +02:00
Saxon Fletcher 05b3a6a3c0 Fix feature preview dialog selected (#47870) 
Use bg-accent for selected state across elements. This fix is specific
to the feature preview dialog.

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

## Summary by CodeRabbit

* **Style**
* Updated feature preview selection styling for clearer visual
distinction between selected and unselected items.

<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-07-13 16:24:55 +08:00
Charis b6cdd5cce3 refactor(sql-editor): extract pure SQL helpers + tighten safe-SQL boundary (decompose 2/6) (#47831) 
## Summary

1. **Extract pure logic** out of the 1056-line `SQLEditor.tsx` monolith
into unit-tested functions in `SQLEditor.utils.ts`.
2. **Remove `rawSql()` from the SQL editor** and tighten the
untrusted→safe boundary per the `safe-sql-execution` model.

## Extracted functions (+ tests)

- `getEditorSql(editor, snippetContent?)` — selection → full value →
snippet fallback. Returns an **`UntrustedSqlFragment`** so
editor/snippet SQL keeps its provenance.
- `computeErrorHighlightLine(error, startLineNumber)` — parses the `LINE
n:` marker + selection offset.
- `assembleCompletionDiff(meta, text)` — before/selection/after assembly
for the AI completion diff.
- `buildExplainSql(sql, impersonatedRoleState)` — takes an already-safe
fragment; EXPLAIN ANALYZE + role impersonation + rollback wrapping.
- `buildDebugPromptText(sql, errorMessage)` — the assistant debug prompt
string.

## Safe-SQL boundary

- `rawSql()` is no longer used anywhere in the SQL editor.
- `executeQuery` / `executeExplainQuery` now **require a
`SafeSqlFragment`** — safe by construction, so they can never auto-run
untrusted SQL.
- `acceptUntrustedSql` promotion happens **only in the small run/explain
gesture handlers** (`executeQueryFromButton`, `handleRunShortcut`,
`handleRunExplain`, and the warning-modal confirm handlers), never
buried in the long helpers.

## Verification

- `vitest` — 156 pass (11 characterization + 145 utils, incl. new cases)
- `pnpm --filter studio typecheck` — clean for all SQL editor files (two
unrelated `@sentry/tanstackstart-react` module-resolution errors exist
on current master pre-install; not touched by this PR)
- `eslint` — 0 errors

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

## Summary by CodeRabbit

* **Enhancements**
* Improved SQL execution and EXPLAIN workflows with safer handling at
run and analysis actions.
  * Enhanced SQL selection and snippet handling in the editor.
* Improved error highlighting to more accurately identify affected
lines.
* Refined completion previews and debugging prompts for clearer results.
* EXPLAIN ANALYZE now supports rollback-wrapped execution and avoids
duplicate wrapping.

* **Bug Fixes**
* Improved behavior when working with selected, empty, or missing SQL
content.
* Prevented existing EXPLAIN statements from being unnecessarily
modified.

<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-07-10 15:00:58 -04:00
Danny White 539b66f5c4 fix(studio): improve unified logs checkbox hit area and align icons (#47832) 
## 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 / polish

## What is the current behavior?

Unified logs row selection checkboxes only respond to clicks on the
checkbox itself, so it's easy to miss. Log type icons in the table also
don't match the ServiceFlow panel (16px / muted vs 14px / lighter /
strokeWidth 1.5).

## What is the new behavior?

- Expand the select checkbox tap target with `hit-area-2`, and add a
visible `hover:border-foreground-muted` affordance (matching the older
logs explorer intent; the base Checkbox hover is a no-op after the
colour-system token collapse).
- Align log type icons with ServiceFlow: 14px,
`text-foreground-lighter`, `strokeWidth={1.5}` on both the table column
and ServiceFlow section headers.

## Additional context

Older logs used an `absolute inset-0` wrapper for the same hit-area
problem; unified logs uses the design-system `hit-area` utility instead.

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

## Summary by CodeRabbit

* **Style**
* Refined unified log visuals with more consistent icon sizing, stroke
weight, and muted coloring.
* Improved checkbox hover styling and expanded its clickable area for
easier selection.


<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-07-10 18:20:38 +00:00
Danny White e5df232b44 fix(studio): polish unified logs row alignment and success colours (#47829) 
## What kind of change does this PR introduce?

UI polish

## What is the current behavior?

Unified logs row chrome is slightly misaligned (checkbox vs filter
toggle, uneven gaps around the level dot), success grey is too dark and
doesn’t match the Level key, and log-type icons read a bit heavy.

## What is the new behavior?

- Aligns the row checkbox with the filter sidebar toggle and spaces the
level dot evenly between checkbox and timestamp
- Drops the checkbox `translate-y` nudge in favour of normal middle
alignment
- Introduces `--chart-success` so the chart and Level key/dots share a
lighter grey
- Softens log-type icon colour on each row

| Before | After |
| --- | --- |
| <img width="1024" height="759" alt="1293"
src="https://github.com/user-attachments/assets/af7ab83f-8917-41cb-99f3-1c1f92df769e"
/> | <img width="1024" height="759" alt="52159"
src="https://github.com/user-attachments/assets/9b859308-2101-4a02-bdc1-75e5750f84fa"
/> |


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

## Summary by CodeRabbit

* **Style**
* Improved Unified Logs table spacing and alignment, including narrower
selection and level columns.
  * Refined checkbox and date-cell presentation for a cleaner layout.
  * Updated log type icons to use muted foreground styling.

* **Bug Fixes**
* Success statuses and chart indicators now consistently use the
dedicated success color across light and dark themes.


<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-07-11 03:13:41 +10:00
Noah Moss 3fda760fe1 Fix R2NP disk size multiplier in FE copy (#47827) 
## 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?

Fixes the copy here to accurately reflect the [constant defined on the
backend](https://github.com/supabase/platform/blob/7653f016838fc1bba15bbce201a1c58566f1751a/packages/api-core/src/shared/disk.ts#L32).
No behavior change.


https://linear.app/supabase/issue/INDATA-906/ensure-fe-r2np-disk-size-multiplier-is-accurate

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

## Summary by CodeRabbit

* **Bug Fixes**
* Updated restore guidance to accurately indicate that the restored disk
size may be approximately 1.5× larger.

<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-07-10 18:40:46 +02:00
Jordi EnricandClaude Opus 4.8 17ee3e6977 feat(studio): add Multigres log type to unified logs FE-3785 (#47560) 
## Problem

The Multigres log type is available in the legacy logs collections but
was missing from the new unified logs, so Multigres logs could not be
selected or viewed there.

## Fix

Wire the `multigres_logs` source into unified logs the same way the
other single-source types (Realtime, Supavisor, PgBouncer) are: a
display label, a filter condition, the derived `log_type` expression, a
display-casing entry, and a sidebar icon.

## How to test

- Open a project with Multigres logs and go to the new unified logs view
- Open the Log Type filter and confirm "Multigres" appears as an option
- Select "Multigres" and confirm rows from the `multigres_logs` source
are returned and labeled "Multigres" with the network icon
- Expected result: Multigres logs are filterable and display correctly,
matching the legacy logs behavior

## Notes

Level/severity uses the shared `severity_text` fallback that all
non-HTTP sources rely on. If Multigres rows come back always classified
as success, the OTEL pipeline may not populate `severity_text` for this
source (legacy logs read the level from a JSON `event_message`), which
would need a source-specific level branch.

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

* **New Features**
* Added support for the **Multigres** log type in Unified Logs (labels,
icon, and derived filtering/grouping/counting).
* Unified Logs now renders Multigres **event_message** by extracting the
`msg` field from valid JSON, with correct capitalization.
  * Unified Logs row click telemetry now recognizes **Multigres**.
* The **Multigres** log type option is hidden when the selected project
is not high-availability.
* **Tests**
* Added/updated unit tests for Multigres event-message parsing and
shared event-message display behavior.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->

---------

Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
 2026-07-10 16:38:41 +00:00
Danny White 79b2eeb8b2 fix(studio): make replica pricing learn more an inline link (#47825) 
## What kind of change does this PR introduce?

Bug fix / polish

## What is the current behavior?

On the add-destination sheet for read replicas, “Learn more” sits as a
stray sibling next to the cost copy.

## What is the new behavior?

Cost copy and “Learn more” are one paragraph, with the link inline and
styled like `InlineLink` (underline, pointer cursor, hover colour):

> New replica will cost an additional $16.25/month. Learn more

| Before | After |
| --- | --- |
| <img width="1258" height="120" alt="CleanShot 2026-07-10 at 12 21
21@2x"
src="https://github.com/user-attachments/assets/bb47461e-e856-4cf2-b81a-f8e9aa9ebf75"
/> | <img width="1256" height="114" alt="CleanShot 2026-07-10 at 12 23
48@2x"
src="https://github.com/user-attachments/assets/094332be-ad2e-4782-94ac-08c8775a4665"
/> |

_Note that the grey icon square is being fixed separately in
https://github.com/supabase/supabase/pull/47794._

## To test

On the staging preview, open Database → Replication → Add destination →
Read replica, and check the footer cost line in light and dark mode —
“Learn more” should sit inline, show a pointer cursor, and change colour
on hover.

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

## Summary by CodeRabbit

## Summary by CodeRabbit

* **New Features**
* Added estimated monthly pricing to the read replica pricing dialog
trigger (“additional {cost}/month”).

* **UI Improvements**
* Moved the pricing/cost impact messaging from the form footer to the
pricing dialog area.
* Refined the “Learn more” link/button styling and layout for a cleaner
presentation.



<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-07-10 16:34:16 +00:00
Joshen Lim 0747af0afe Adjust PITR UI (#47821) 
## Context

Original intention was to fix the PITR UI for smaller viewports, but
realised that the layout of the UI could be improved as well. (Tbh this
UI could do with a bit of revisiting, but just making patches for now to
improve what's existing)

Fixes also apply to the restore to new project page since they share the
same component

## Changes involved

- Am opting to change the layout of the UI a little such that date
selection is on the left, and time selection is up top
- Fits the user flow a bit better - you select the parameters you want,
and the final message at the bottom is the summary
<img width="1033" height="525" alt="image"
src="https://github.com/user-attachments/assets/614c0327-f5db-4e1f-a2a1-ae8a9ea89978"
/>
- In the confirmation dialog, we were originally showing the top label
as "Local time" which I feel is inaccurate especially if the user has
selected a different timezone from where they're located at.
  - Opting to display the full name of the selected timezone instead
<img width="546" height="261" alt="image"
src="https://github.com/user-attachments/assets/cd8838a1-8476-4492-bc86-cac229685e5a"
/>
- RE mobile layout - am currently just opting to have them in a column
fashion although I feel like this isn't ideal either (requires
revisiting of the UI as a whole to adjust the layout on desktop too)
- e.g the Date picker here could be a popover like Unified Logs to
streamline what is essentially a form
<img width="507" height="831" alt="image"
src="https://github.com/user-attachments/assets/c8e055bd-49c3-4d0a-9d8c-e54f53e9fcb4"
/>
- Also fixed the CTA URL for read replicas, was still pointing to
/settings/infrastructure, should point to database/replication




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

* **New Features**
* Point-in-time recovery now includes the selected timezone throughout
the restore workflow.
* Restore confirmations display the chosen timezone and a clear summary
of the target recovery date and time.
* Recovery details now show the earliest and latest available backups
for the selected date, including the two-minute matching window.
* **Bug Fixes**
* Updated the read-replica management link to direct users to database
replication settings.
* **Style**
* Improved restore form layout, calendar presentation, and timezone
selector alignment.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-07-10 22:37:41 +08:00
Charis 3461d60aba feat(sql-editor): consolidate controls into a single top bar (#47792) 
## 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 / UI improvement

## What is the current behavior?

SQL editor controls are split across two rows: a bar between the editor
and results (tabs, download,
run/save/format/db-selector/role/favorite/intellisense) and a separate
footer at the very bottom (row count + limit dropdown).

## What is the new behavior?

All controls are consolidated into a single top bar above the editor (⋮
overflow menu, database selector, role impersonation, limit, save, run).
The bar between the editor and results now holds only result-relevant
information: the Results/Explain/Chart tabs, the row-count summary, and
the Export button. The limit dropdown is now always visible as a
persistent query setting.
 2026-07-10 10:27:43 -04:00
Jordi EnricJoshen LimIvan Vasilov
 		a75a3b22b6 fix(studio): align PITR calendar dates in the first week (#47806) 
## What

Fixes misaligned day cells in the PITR calendar widget for the first
week of the month.

## Why

The PITR calendar draws a \`border\` on each day cell via
\`classNames.day\`. The day \`<td>\` has no explicit width, so
\`box-sizing: border-box\` doesn't apply and the 1px borders add to its
size (36px → 38px), while the weekday header cells stay pinned at
\`w-9\` (36px). Bordered day cells therefore drift right of their
headers, which is most visible in the first partial week where
unbordered leading cells sit flush next to the wider bordered ones.

## How

Pin each day cell to a fixed \`w-9 box-border\` so the border is drawn
inside the 36px box, and let the day button fill the cell (\`w-full\`).
Column pitch now matches the weekday headers regardless of border state.

Class-only change, no logic touched.

Closes FE-3886

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

* **Style**
* Improved calendar day cell sizing and layout for more consistent
rendering.
* Ensured day buttons use full-width styling where applicable, while
preserving existing hover, border, background, and corner behavior.
* **Documentation**
* Added documentation for “Calendar with disabled days,” including a new
interactive preview.
* **New Features**
* Introduced a calendar example demonstrating disabled-day behavior with
mid-week month start and restricted date selection.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->

---------

Co-authored-by: Joshen Lim <joshenlimek@gmail.com>
Co-authored-by: Ivan Vasilov <vasilov.ivan@gmail.com>
 2026-07-10 11:40:33 +01:00
Saxon FletcherJoshen LimAlaister Young
 		4fd5f8adf1 Vercel project connect layout (#47623) 
Bring the Vercel project connect layout up to date with other connect
screens. Follows up from https://github.com/supabase/supabase/pull/47550
which is required.

| Before | After |
| --- | --- |
| <img width="2368" height="1680" alt="image"
src="https://github.com/user-attachments/assets/f5c0b8ef-8fb5-4176-b0cf-98e33958dcc8"
/> | <img width="848" height="808" alt="image"
src="https://github.com/user-attachments/assets/e64c289f-0b73-4605-ad38-4552d3d6f934"
/> |

<img width="883" height="733" alt="image"
src="https://github.com/user-attachments/assets/7ccd9b72-de6a-4c21-ac8e-abe773bb8116"
/>


## Testing

- Open the deploy preview or staging URL for
`/dashboard/integrations/vercel/acme-production/marketplace/choose-project`
with callback params from a real Vercel Marketplace install redirect
(see _Vercel_ subheading below).

### Vercel

- Run through the [install
flow](https://github.com/supabase/supabase/pull/47550) on a real project
until you get to the project connection screen, copy the url and paste
into staging preview


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

## Summary by CodeRabbit

* **New Features**
* Introduced interstitial-style Vercel integration experiences for both
install and choose-project flows, with improved page titles and clearer
environment-variable guidance (including tooltip details).

* **Bug Fixes**
* Improved Vercel integration matching by configuration id and
strengthened loading/error handling with more specific “not found”
messaging.

* **Refactor**
* Enhanced the Vercel/GitHub project chooser with a dedicated
interstitial mode.
* Updated partner logo rendering to support optional styling
customization.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->

---------

Co-authored-by: Joshen Lim <joshenlimek@gmail.com>
Co-authored-by: Alaister Young <alaister@users.noreply.github.com>
 2026-07-10 18:22:36 +08:00
Pamela ChiaandAlaister Young 9858562b8b fix(telemetry): dedupe funnel toast error events (#47802) 
## Summary
Since #47293, an API failure on a signup / org-creation /
project-creation form emitted `dashboard_error_created` twice:
`useTrackFunnelError` fired the origin-tagged event and the global
`ToastErrorTracker` independently fired the legacy untagged
`source:'toast'` event for the same toast, each behind its own 10%
sampling draw. I verified the twin rate empirically at 8-11% of
origin-tagged funnel toasts, exactly the floor for two independent 10%
draws, meaning the twin co-fires for effectively every funnel error
([Hex
thread](https://app.hex.tech/supabase/thread/019f3bc1-3a5c-7200-9122-8e3439bfbe8c)).
Any consumer counting funnel errors without an `origin IS NOT NULL`
filter saw ~2x inflation.

The fix makes `ToastErrorTracker` the sole emitter of `source:'toast'`
events, so the duplicate is unrepresentable rather than suppressed.
Funnel call sites pass the id returned by `toast.error()` into
`trackFunnelError`, which registers the funnel properties against that
toast id instead of firing its own event – the tracker then emits a
single `dashboard_error_created` enriched with `origin` /
`errorCategory` / `errorReason` / `errorCode` for registered toasts, and
the plain untagged event otherwise. The `'toast'` overload of
`trackFunnelError` requires the toast id, so a missed pairing is a
compile error rather than a silent double count. Registration is
unconditional and there's only one sampling draw, so suppression can't
lose a sampling race. `'form'`-sourced funnel events are unchanged.

## Changes
- `lib/toast-errors.tsx`: toast-id → funnel-properties registry
(`registerFunnelErrorToast`); `ToastErrorTracker` emits one (optionally
enriched) event per error toast under a single 10% draw, deleting
entries once consumed
- `lib/telemetry/use-track-funnel-error.ts`: overloaded signature –
`'toast'` requires the id returned by `toast.error()` (type-enforced),
`'form'` keeps direct emission with its own sampling
- Update the 7 funnel `toast.error` call sites in `NewOrgForm`,
`SignUpForm`, and `pages/new/[slug]` to pass the toast id
- Component tests for the tracker (previously uncovered), including an
end-to-end test through `useTrackFunnelError`
- Code hygiene (also flagged by CodeRabbit): all four
`dashboard_error_created` emitters (toast, form, `AlertError`,
`ErrorMatcher`) independently encoded the 10% draw – downstream analysis
assumes a uniform sampling multiplier across sources, so one site
drifting would silently skew comparisons. The rate and the draw now live
in one place (`isDashboardErrorSampled()` in
`lib/telemetry/error-sampling.ts`). No behavior change.
- Mount `ToastErrorTracker` in the TanStack root (`routes/__root.tsx`),
mirroring `pages/_app.tsx`. The TanStack tree mounted `Toaster` but
never the tracker, so untagged toast error telemetry has never fired in
that flavour – and with the tracker now the sole emitter, the missing
mount would have silently dropped funnel toast events there too. Side
effect once the TanStack flavour ships: untagged `source:'toast'` volume
from it goes from zero to normal.

## Testing
Component-tested (`apps/studio/lib/toast-errors.test.tsx`):
- [x] Unregistered error toast fires exactly one untagged
`dashboard_error_created {source:'toast'}`
- [x] Registered funnel toast fires exactly one event, enriched with
`origin`/`errorCategory`/`errorReason`/`errorCode`
- [x] `useTrackFunnelError` with a toast id routes through the tracker
as a single enriched event
- [x] Non-error toasts ignored; the 10% sampling gate still applies

Full Studio unit suite passes (392 files / 4371 tests), plus typecheck
and lint.

Also verified end-to-end in a local browser (TanStack flavour, sample
rate temporarily forced to 1): a failed signup produced exactly one
`dashboard_error_created` with `{source:'toast', origin:'signup',
errorCategory:'api', errorReason:'email_already_registered',
errorCode:403}` and no untagged twin (two independent trials); an
unregistered error toast produced exactly one plain `{source:'toast'}`;
a client-side validation failure produced exactly one `{source:'form',
origin:'signup', errorCategory:'validation',
errorReason:'email_invalid'}`; success toasts produced nothing.

Post-deploy I'll re-run the twin-rate query from the Hex thread; the
untagged-twin rate on funnel pages should decay to ~0 as stale bundles
reload over 2-3 days.

## Notes
- Origin-tagged funnel toast events now ride the tracker's single 10%
draw instead of their own independent draw – statistically identical
volume, but the event fires on the tracker's next effect rather than
synchronously at the call site (irrelevant for PostHog)
- Registration must happen in the same synchronous block as
`toast.error()` (documented on the `TrackFunnelError` type) – all
current call sites comply
- The invalid Postgres version toast in `pages/new/[slug].tsx` (~line
416) needs no special-casing: unregistered toasts keep the plain
untagged event, so its telemetry is preserved
- Heads-up for `dashboard_error_created` consumers: overall untagged
`source:'toast'` volume will dip slightly after this deploys, since
funnel-page twins disappear. A volume monitor seeing that drop is this
fix landing, not a tracking regression (same class as the intended
GROWTH-893 sampling-unification drop).

## Linear
- fixes GROWTH-965


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

## Summary by CodeRabbit

* **Bug Fixes**
* Enhanced error telemetry for organization creation, sign-up, payment,
and project-creation flows by associating failures with toast
identifiers and enriched funnel context.
* Standardized dashboard error sampling logic across error handling
components for consistency.

* **Tests**
* Added comprehensive test coverage for toast error tracking, including
funnel registration, deduplication, filtering, and sampling behavior.

<!-- end of auto-generated comment: release notes by coderabbit.ai -->

---------

Co-authored-by: Alaister Young <10985857+alaister@users.noreply.github.com>
 2026-07-10 17:35:27 +08:00
ChloeGarciaMillerandandGildas Garcia 59a82c52f8 Fix: improve accessibility for icon button (SQL Editor menu) (#47674) 
## 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 (accessibility improvement)

## What is the current behavior?

Icon-only button (not visible on widescreen displays) does not have
explicit accessible name for screen readers and tooltip.

## What is the new behavior?

The icon-only button now has explicit accessible name using visually
hidden text (sr-only), ensuring proper screen reader support.

## Additional context

Tooltip text is added for visual users.
No visual changes were introduced


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

* **UI Improvements**
* Added a tooltip to the SQL editor’s “More actions” dropdown button,
improving discoverability.
  * The tooltip now shows “More actions” when hovering over the trigger.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->

Co-authored-by: Gildas Garcia <1122076+djhi@users.noreply.github.com>
 2026-07-10 11:30:45 +02:00
Joshen Lim 66bfc5fdc3 Refactor ConnectSheet + Add unit tests to cover various logic (#47764) 
## Context

PR here mainly breaks up the files under `ConnectSheet` to separate the
functional logic so that we can write unit tests.
No behavior changes intended beyond the bug fixes

## Changes involved

- **Test organization:** moved all root-level `ConnectSheet` test files
into `ConnectSheet/__tests__/` for consistency with other parts of the
codebase that use this convention.
- **Bug fix:** read replica label had a stray `}` / missing `)`,
rendering as e.g. `Read Replica (us-east-1 - abc123})` instead of `Read
Replica (us-east-1 - abc123)`.
- **`ConnectSheet.tsx`:** extracted the "hydrate sheet state on open"
`useEffect` logic (mode/field/URL param resolution from URL vs.
localStorage) into a new `ConnectSheet.utils.ts`, with unit tests
- **`useConnectServerEnv.ts`:** fixed two race conditions in the secret
reveal/hide flow:
- `toggle()` and `getValue()` could each fire a separate reveal request
if triggered close together — now deduped to share one in-flight
request.
- `getValue()` could hide a secret that had just been explicitly
revealed by a concurrent `toggle()`, due to reading a stale closure
value — now reads the live state via `useLatest`.
- Also stopped swallowing the original error on reveal failure (now
attached via `cause`).
- Added tests for the above, plus the 10s auto-hide timer (previously
untested).
- **`ConnectStepsSection.tsx`:** extracted `resolveContentPath` and the
three inline "show notice" booleans (IPv4 addon, session pooler,
self-hosted MCP) into `ConnectStepsSection.utils.ts`, matching the
existing pattern for the Data API notice. Added unit tests for all of
them.

## To test
- [ ] Just a basic smoke test of the Connect sheet should do

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

* **New Features**
* Improved connect setup hydration so saved preferences and URL values
are applied more consistently when opening the sheet, including
automatic URL backfilling where needed.
* Refreshed connection guidance notices (IPv4 add-on, session pooler,
and self-hosted MCP) with more consistent logic.
* **Bug Fixes**
* Fixed secret reveal behavior to keep concurrent reveal actions in
sync, handle failures more safely, and ensure auto-hide works reliably.
  * Corrected the read-replica option label formatting.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-07-10 15:24:34 +08:00
Han Qiao ce81c2f6ec feat(studio): allow deleting a branch from general settings page (#47677) 2026-07-10 14:22:52 +08:00
Saxon Fletcher 75bb899f01 Fix mobile toolbar behaviour (#47800) 
**Before**

<img width="340" height="104" alt="image"
src="https://github.com/user-attachments/assets/48a434de-21c2-4e3f-8cad-3c6408f7348d"
/>

**After**

<img width="508" height="317" alt="image"
src="https://github.com/user-attachments/assets/e52c6ba6-c173-4bb8-a466-1697585286c7"
/>


Fixes 
- Rendering issue of the menu toggle on mobile when menu is open
- You can now switch between panels when open (e.g. from help to
advisor). There was previously a bug that would just close the drawer
rather than switch.

**To test**
- Reduce screen size in a project view
- Click top right menu item
- Notice menu item is active in light mode
- Click "Assistant" and notice it switches vs closes the drawer
- Click close or outside the toolbar to close the drawer

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

* **Bug Fixes**
* Improved the mobile menu button’s visual state by updating the menu
icon color when the menu is open.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-07-10 12:14:34 +08:00
Joshen Lim 87c5f74ceb Add CTA to update tax ID if run into tax ID issues when upgrading plan (#47767) 
## Context

There's a chance users may run into tax ID issues when upgrading a plan
as such:
<img width="412" height="120" alt="image"
src="https://github.com/user-attachments/assets/4c559ae0-d942-4c10-b83a-c5944a2a49ee"
/>

Adding a CTA here to guide users and mitigate confusion on how to
proceed to remediate

<img width="399" height="147" alt="image"
src="https://github.com/user-attachments/assets/3bcaca14-bd77-4168-b9ab-cd6b75f79e24"
/>


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

* **Bug Fixes**
* Improved billing-profile validation feedback by standardizing error
messaging for tax ID vs. billing address country mismatches.
* Added an in-toast “Update tax ID” action to guide users directly to
the fix.
* Updated the billing dialog’s close behavior so it returns users to the
relevant address section (including scrolling to it automatically).
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-07-10 10:06:29 +08:00
Riccardo Busetti f45389138b ref(pipelines): Remove unnecessary restarts (#47732) 2026-07-10 01:34:19 +00:00
Gildas Garcia 72aa214b0c fix: new project form accessibility issues (#47785) 
## Problem

The new project form has accessibility issues:
- labels are not linked to inputs
- description are not linked to inputs

## How to test

Navigate through the form inputs with voice over and make sure every
input makes sense

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

* **Bug Fixes**
* Improved form field identification consistency across project creation
screens (compute size, database password, project name, PostgreSQL
version, region, and organization).
* Enhanced selector/input accessibility by adding explicit element
identifiers to key controls.
* Updated region and repository UI structure to improve reliable
rendering without changing setup behavior.
* Preserved existing password, version, and routing logic while making
dropdowns and fields easier to locate and interact with.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-07-09 18:11:57 +02:00
Ali Waseem d949a19f97 fix(studio): extend downgrade error toast duration (#47780) 
## Summary
- Downgrading to Free tier is blocked server-side when an org has an
active branch, but the resulting error toast in `ExitSurveyModal.tsx`
used the default 4s toast duration, making it easy to miss.
- Adds `duration: 10_000, dismissible: true`, matching the pattern
already used for other important billing error toasts
(`org-subscription-update-mutation.ts`, `NewOrgForm.tsx`).

Fixes FE-3882

## Test plan
- [ ] Attempt to downgrade an org with an active branch to Free tier and
confirm the error toast stays visible for 10s and can be dismissed

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

## Summary by CodeRabbit

* **Bug Fixes**
* Improved the downgrade error message to stay visible longer and be
easier to dismiss, making failures clearer for users.

<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-07-09 07:58:03 -06:00
Gildas Garcia 1aa23f9f64 fix: fix several accessibility issues on the organization home page (#47769) 
## Problem

- Organizations links are not accessible with keyboard
- Project list buttons are missing labels
- Headings should be sequential


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

## Summary by CodeRabbit

* **Bug Fixes**
* Improved keyboard and screen-reader accessibility for project actions
and project reference copy controls.
  * Added clearer tooltip guidance for copying a project reference.
* Updated project and organization card interactions for more consistent
click and focus behavior.

<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-07-09 15:14:05 +02:00
Charis 1987f19d0a feat(sql-editor): add manual save feature preview (#47745) 
## What

Adds an opt-in **SQL Editor manual save** feature preview that switches
the SQL Editor from autosaving every edit to saving only on demand, and
hardens the tab-close flow so unsaved edits are handled correctly.

## Changes

**Feature preview**
- New `sqlEditorManualSave` flag + `UI_PREVIEW_SQL_EDITOR_MANUAL_SAVE`
local-storage toggle, wired into the Feature Preview modal with an
explanatory panel.
- `useIsSqlEditorManualSaveEnabled` gates behavior on both the flag and
the user's preview opt-in.

**Editor toolbar**
- Save button (with `Cmd+S`) next to Run, plus an autosave status
indicator showing dirty/saving/saved state and a shortcut to disable
autosave (emits a `sql_editor_autosave_disable_clicked` telemetry
event).

**Discard on close**
- Closing a snippet tab with unsaved edits prompts for confirmation and,
on confirm, actually discards the local edits and evicts the cached
server copy so the snippet reopens clean.

**Decouple tab layout from SQL specifics**
- Tabs store gains a generic per-type close-handler registry
(`registerTabCloseHandler` / `getCloseConfirmation` / `closeTabs`). The
SQL editor registers its discard + confirmation behavior from the save
coordinator.
- Low-level `removeTab`/`removeTabs` (rename/move re-keying, stale
cleanup) intentionally do **not** trigger discard.
- Adds `statusOnDiscard` lifecycle transition and `clearSnippetContent`
store action.

## Testing
- `pnpm --filter=studio typecheck` — clean.
- Added unit tests for the close-handler registry (fires on single/multi
close, skips re-keying/cleanup removals, respects tab type, selects
confirmation copy, unregisters cleanly).

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

* **New Features**
* Added a SQL editor manual-save preview with a “Save” button and
`Cmd+S`, plus a modal option to disable manual-save/preview.
* Added “unsaved changes” tab status indication when manual-save is
enabled.
* Introduced tab-type-specific close confirmations (shown only when
needed).
* **Bug Fixes**
* In manual-save mode, closing a SQL tab with unsaved edits now clears
local snippet content and refreshes it on reopen.
* **Tests**
  * Added coverage for tab close handlers and confirmation behavior.
* **Chores**
* Added a persisted setting allowlist entry and tracked autosave-disable
clicks via telemetry.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-07-09 08:37:38 -04:00
Joshen Lim f34fdd6c8f Skip using count estimate function for retrieving row counts if in read only context (#47761) 
## Context

Currently when retrieving row counts of a table in the Table Editor,
we're using a `COUNT_ESTIMATE` pg function
([ref](https://github.com/supabase/supabase/blob/master/packages/pg-meta/src/sql/studio/database/get-count-estimate.ts#L5))
to retrieve an estimate (instead of checking `pg_class` -> `reltuples`)
as that would theoretically provide a more accurate representation.

However, in a read only context, that function can't be used - users
will run into `cannot execute CREATE FUNCTION in a read-only
transaction`, so we need to fallback to just checking `pg_class` in this
scenario.

The logic's already set up as we were previously looking into allowing
users to use a read replica to power the dashboard, but we also need to
consider members with read-only roles within the organization, so this
PR updates the logic a little to factor that in.

## To test

- [ ] With a read-only role, open the table editor and verify that we're
not using the count estimate function to retrieve the table row counts

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

## Summary by CodeRabbit

* **New Features**
* Updated the invite member dialog to open in a larger size for better
usability.

* **Bug Fixes**
* Improved table row count behavior so it now respects read-only access
and permission limits more reliably.
* Count estimates should now be shown more consistently across different
database contexts.

<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-07-09 18:19:39 +08:00
Joshen Lim 644fe0821b Add create org CTA for authorize route if no org found (#47760) 
## Context

As per PR title - also left a comment that this is a short term solution
for now, so we know where to clean up after the long term solution is
implemented

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

## Summary by CodeRabbit

- **New Features**
- Added a clear action in the empty organizations state so users can
create an organization directly from the authorization flow.

- **Bug Fixes**
- Improved authorization error messaging for clearer, more consistent
display.
- Refined invalid authorization guidance so the retry prompt and
missing-parameter details are shown more cleanly.

<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-07-09 18:18:08 +08:00
Joshen Lim 4901f081e5 Migrate remaining requests to pg-meta API to use query endpoint (#47758) 
## Context

Migrates the remaining API requests to the pg-meta endpoint to use the
query endpoint directly with the SQL from the pg-meta package. This
touches the following:
- policies
- publications
- triggers
- views
- materialized views
- types

## To test
Just need to verify that we're still fetching the data correctly on
these pages
- Database policies
- Database publications
- Database triggers
- Database tables (views + materialized views)
- Database types

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

* **Bug Fixes**
* Improved and stabilized loading of database metadata (views, triggers,
RLS policies, publications, materialized views, and enum types),
including more reliable schema-scoped filtering.
* Updated policy loading behavior and related UI queries to consistently
use schema arrays, improving cache correctness and consistency.
* **Tests**
* Updated end-to-end test synchronization to wait for the correct
metadata responses using more specific request identifiers.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-07-09 17:09:03 +08:00
Joshen Lim 073cada53a Fix observability custom reports menu item (#47759) 
## Context

More action button should be flushed to the right here
<img width="294" height="156" alt="image"
src="https://github.com/user-attachments/assets/65017960-3edb-4268-bb0e-1e2c26937d4b"
/>

## Changes involved
- Adjust `Menu.Item` in `packages/ui` to use a `div` instead of a `span`
- Was otherwise causing HTML validation issues as we were trying to nest
a `div` within a `span`
- Having a `div` is a bit more flexible as well since `Menu.Item`
expects `children` to be of any type (e.g a react node)

<img width="279" height="149" alt="image"
src="https://github.com/user-attachments/assets/12730cef-b077-4ef4-93c9-c21def939888"
/>


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

* **Refactor**
* Standardized the Observability menu component to use named exports,
ensuring consistent usage across the app.
* Updated the mobile observability menu registration to reference the
correct exported component.
* **Style**
* Refined Observability menu item layout, spacing, truncation, and
dropdown sizing for a cleaner presentation.
* Enhanced menu item rendering to allow custom `className` styling and
full-width content layout.
* **Accessibility**
* Added an aria-label to the “more actions” button for improved screen
reader support.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-07-09 17:08:11 +08:00
Monica KhouryandJoshen Lim 9e17c41771 Chore: Link Delete Project doc from the Delete Project confirmation m… (#47637) 
Fixes FE-3801. 

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

* **Bug Fixes**
* Updated the project deletion confirmation dialog so the “learn more /
documentation” text and external link are always shown consistently.
* Paid projects now include the full warning (“All project data will be
lost, and cannot be undone.”) alongside the documentation link, while
free projects show the simplified message without the additional
warning.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->

---------

Co-authored-by: Joshen Lim <joshenlimek@gmail.com>
 2026-07-09 11:41:16 +03:00
Monica Khourysupabase-autofix-botAli Waseem
 		fd8a37b1d0 feat: add toggle for sensitive data visibility in table columns (#46180) 
## Fixes

FE-2619

## What is the new behavior?

This PR adds support for marking table columns as sensitive and masking
their values in the grid view.

Sensitive columns:

- Display an 8-dot mask instead of the underlying value
- Remain masked across page refreshes
- Can be temporarily revealed for 5 seconds via the **Show data** action
- Display a warning when copying rows containing sensitive data

This helps prevent accidental exposure of sensitive information when
sharing screens, recording demos, or taking screenshots.

## Testing

- [x] Toggle sensitivity ON → save → refresh → remains masked
- [x] Toggle sensitivity OFF → save → refresh → remains unmasked
- [x] Toggle sensitivity multiple times → state remains consistent
- [x] Copy row with sensitive columns → warning shown
- [x] Click **Show data** → value revealed for 5 seconds then re-masked
- [x] Text, Boolean, Binary, JSON, and Foreign Key columns all display a
consistent 8-dot mask

### Test data

SQL fixture covering multiple PostgreSQL data types:

https://gist.github.com/monicakh/2485e9054bf21045912359871e9a1cb4. 

### UI

<img width="1284" height="554" alt="CleanShot 2026-06-09 at 12 01 33@2x"
src="https://github.com/user-attachments/assets/4aec0ba7-c874-42d7-9442-d2c704b319cc"
/>

<img width="1200" height="560" alt="CleanShot 2026-06-07 at 10 43 40@2x"
src="https://github.com/user-attachments/assets/b9569484-6fcc-47de-bc3d-881d0edc4060"
/>

The **Show data** action is only available for sensitive columns.

<img width="450" height="400" alt="CleanShot 2026-06-07 at 10 42 18@2x"
src="https://github.com/user-attachments/assets/d48849a2-ec0b-4522-a787-561a1d204ec9"
/>

Warnings on Copy command

<img width="450" height="80" alt="CleanShot 2026-06-09 at 11 58 42@2x"
src="https://github.com/user-attachments/assets/374e7d6b-b82a-4923-b035-2ec9b2f7bb7d"
/>

<img width="450" height="80" alt="CleanShot 2026-06-09 at 11 58 58@2x"
src="https://github.com/user-attachments/assets/ecd951bb-e9e2-47ae-9ddd-d32969e01c12"
/>


<!-- review_stack_entry_start -->

[![Review Change
Stack](https://storage.googleapis.com/coderabbit_public_assets/review-stack-in-coderabbit-ui.svg)](https://app.coderabbit.ai/change-stack/supabase/supabase/pull/46180?utm_source=github_walkthrough&utm_medium=github&utm_campaign=change_stack)

<!-- review_stack_entry_end -->

<!-- end of auto-generated comment: release notes by coderabbit.ai -->

---------

Co-authored-by: supabase-autofix-bot <noreply@supabase.com>
Co-authored-by: Ali Waseem <waseema393@gmail.com>
 2026-07-09 10:44:42 +03:00
Ali Waseem 000bdd0684 fix(studio): trim leading whitespace in site URL form (#47748) 
## Changes

- **SiteUrl.tsx**: Added `.trim()` to the Zod schema so whitespace is
stripped before validation and before the value reaches the mutation.
All-whitespace input now correctly fails with "Must have a Site URL"
instead of being silently accepted. This matches the existing pattern in
the sibling Redirect URLs form (AddNewURLModal.tsx).
- **SiteUrl.test.tsx** (new): MSW component test with two cases:
- Trims leading/trailing whitespace before submitting to PATCH
/platform/auth/:ref/config
- Shows a validation error and does not submit when the value is only
whitespace

## Test plan

- [x] `npx vitest --run
components/interfaces/Auth/SiteUrl/SiteUrl.test.tsx` — 2/2 pass
- [x] `npm run typecheck` — clean
- [x] `npx eslint` on both files — no new warnings


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

## Summary by CodeRabbit

* **Bug Fixes**
* Improved site URL validation so leading and trailing whitespace is
ignored before saving.
* Prevents whitespace-only values from being submitted and shows a
validation error instead.

* **Tests**
* Added coverage for site URL saving, including trimmed input,
validation failures, request payloads, and success feedback.

<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-07-08 13:06:56 -06:00
Joshen Lim 944c5862f3 Chore/small refactors (#47740) 
## Context

Just extracting the fixes which I think are applicable from this
[PR](https://github.com/supabase/supabase/pull/47695)

Main files are
- `apps/studio/hooks/analytics/useLogsQuery.tsx`
- `packages/common/auth.tsx`
- `packages/common/feature-flags.tsx`

## Changes involved
- Adjust `useLogsQuery` to accept an object as prop, rather than 4
individual params
- This one doesn't address any Sentry issues, but is just a improvement
to the function's API imo, more readable
- Adjust how user email is retrieved in `feature-flags`
- Related Sentry issue
[here](https://supabase.sentry.io/issues/7592718607/?project=5459134)
- The error is a bit vague, but Claude's attempt to fix looks alright in
general IMO
  - Minimally verified that feature flags are loading as expected still

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

## Summary by CodeRabbit

* **Bug Fixes**
* Improved log-related screens and queries for more reliable loading and
filtering across the app.
* Fixed profile and account data handling so identity details are
retrieved more consistently.
* Improved authentication handling to better recognize missing user data
and keep the app stable.
* Updated feature flag personalization to use more accurate account
information.

<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-07-08 23:21:48 +08:00
Vaibhav c070893475 fix: limit regex (#47717) 
- closes https://github.com/supabase/supabase/issues/47712 

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

## Summary by CodeRabbit

* **Bug Fixes**
* Improved SQL query handling so automatic row limits are no longer
added when a query already ends with `LIMIT`, even if there’s whitespace
before the semicolon.
  * Preserved correct behavior for queries using `LIMIT ... OFFSET ...`.
* **Tests**
* Expanded coverage for SQL limit detection and limit-suffix behavior
around whitespace and semicolon placement.

<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-07-08 08:11:39 -06:00
Joshen LimandAli Waseem 0421b1001d Flip show tooltip to true for supavisor connections chart (#47730) 
## Context

Realised that tooltips were not showing up for supavisor charts in
database reports - just needed to flip a boolean
Although - i don't have any projects with supavisor connections data
(even on prod) so I can't visually verify this atm

Also fixes a small issue in which docs url for the chart wasn't showing
if the chart had no data, e.g:
<img width="996" height="311" alt="image"
src="https://github.com/user-attachments/assets/926febe4-9e3d-4975-9278-e7582d6ae12d"
/>

Should have docs button like this
<img width="949" height="351" alt="image"
src="https://github.com/user-attachments/assets/7561c1c5-94ae-405b-bd54-6bc94be0dd0a"
/>


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

* **New Features**
* Enabled tooltips for the “Shared Pooler (Supavisor) client
connections” chart so the metric can be inspected directly.
* **UI Improvements**
* Adjusted the tooltip positioning in the chart header for clearer
readability.
* When charts have no data, the “Learn more”/documentation link now
follows the provided docs URL.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->

---------

Co-authored-by: Ali Waseem <waseema393@gmail.com>
 2026-07-08 12:52:42 +00:00
Gildas Garcia 62160939a8 fix: make status hovercard trigger on focus (#47731) 
## Problem

Status lists only appear on mouse hover and disappear when panning at
200%+ zoom.

## Solution

Make hover-triggered information available via click or focus in line
with WCAG 1.4.13

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

## Summary by CodeRabbit

* **Bug Fixes**
* Improved keyboard accessibility for the service status hover card by
making the trigger focusable.

<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-07-08 14:52:36 +02:00
Carel de WaalandAlaister Young 0acc0eb8b3 feat: Support Form - Sync AI assistant conversation to Front (#46778) 
# Sync AI assistant conversation to Front

## What & why

When a user submits a support ticket, an AI assistant chat opens so they
get help
immediately while waiting for a human agent. This PR mirrors every turn
of that chat into
the Front conversation the support form already created, so the support
team sees the full
context and Front automations (routing, emails, CSAT) can act on it.

Studio holds no Front credentials — it calls the platform endpoints (see
the platform PR)
to do the syncing. The assistant card is gated behind the
`supportAssistantFollowUp`
ConfigCat flag.

## How it works

1. **Submit** — `SupportFormV3` generates a stable `threadRef` (via the
`uuid` package —
`crypto.randomUUID()` is `undefined` in insecure contexts like
non-localhost HTTP and
would throw, silently aborting the submit) and sends it on
`/platform/feedback/send`.
The response returns the Front `conversationId`. Both are stored on
`SubmittedSupportRequest`.
2. **Open chat** — `SupportAssistantSuccessCardContent` opens a chat
seeded with
`supportMetadata` (`threadRef`, `frontConversationId`, subject,
category, severity, …).
   The first message is a `<support>…</support>` XML block.
3. **First user message** — the chat is tagged `isSupportChat = true`;
the `onFinish` hook
   fires `syncSupportChatToFront`.
4. **Subsequent turns** — each `onFinish` slices the unsynced delta,
strips the XML
metadata block from the seed message, and posts to the platform messages
endpoint.
5. **Escalation / resolve** — the `escalate_to_human` /
`resolve_support_conversation` tools
(and manual **Escalate**/**Resolve** buttons in the assistant input)
flip lifecycle status
via `setSupportLifecycleStatus` → `syncSupportLifecycleToFront`, which
calls the
escalation/resolve endpoints. Front rules act on `ai_support_status`.
The assistant only
   resolves after the user explicitly confirms the issue is fixed.

## Key design decisions

- **`threadRef` as the shared key** — one UUID travels as `threadRef` on
submit and as
`chatId` on every sync, so all messages thread into a single Front
conversation.
- **`conversationId` from the form response** — passed to all
sync/lifecycle calls so the
  platform skips lazy derivation and PATCHes custom fields directly.
- **Delta-only sync** — `lastSyncedMessageCount` tracks what's been
sent; the boundary is
snapshotted before the async call to avoid skipping messages that arrive
mid-flight.
- **Server-side de-dup** — stable `external_id` (`chatId:msg.id`) means
retries don't
  duplicate in Front.
- **Fire-and-forget** — sync failures log to Sentry, never break the
chat; `isSyncing`
resets on rehydration so the next `onFinish` retries the same delta.
Message and lifecycle
syncs use separate guards (`isSyncing` / `isLifecycleSyncing`) so an
in-flight message
  sync can't drop an escalate/resolve.
- **Lifecycle queued until the conversation exists** — if a lifecycle
transition is requested
before the initial message sync has returned a `frontConversationId`,
it's stored as
`pendingLifecycleStatus` and flushed once the id is assigned, rather
than dropped.
- **Tools return immediately** — the lifecycle tools return a stub to
the AI SDK; the real
Front call happens in `onFinish`, keeping async I/O out of the tool
execute path.
- **XML seed stripped before sync** — only the user's actual `<message>`
is sent to Front
  (or dropped entirely if the form already created the conversation).

## Changes

| Area | File(s) |
| --- | --- |
| Support form state | `SupportForm.state.ts` — `threadRef` /
`frontConversationId` on `SubmittedSupportRequest` |
| Support form submit | `support-ticket-send.ts` — sends `threadRef`,
reads `conversationId` |
| Support form UI | `SupportFormV3.tsx` — generates `threadRef`, stores
`conversationId` |
| AI assistant state | `ai-assistant-state.tsx` — `SupportChatMetadata`,
`setSupportLifecycleStatus`, `onFinish` wiring, tool handling |
| Message sync | `state/ai-chat-front-sync.ts` — delta tracking, message
filtering, initial vs. incremental |
| API data layer | `data/feedback/ai-chat-front-sync.ts` — typed
platform-client wrappers for the three conversation endpoints |
| Support tools | `lib/ai/tools/support-tools.ts` — `escalate_to_human`,
`resolve_support_conversation` |
| Tool integration | `lib/ai/tool-filter.ts`, `tools/index.ts`,
`generate-assistant-response.ts` |
| Success card | `SupportAssistantSuccessCardContent.tsx` — tags chat on
first engagement |
| Assistant panel UI | `AIAssistant.tsx` — Escalate/Resolve buttons,
disabled input on closed chats, support placeholders |



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

## Summary by CodeRabbit

- **New Features**
  - Support chats now include “Escalate to human” and “Resolve” actions.
- Support submissions can be associated with a stable Front thread via a
generated `threadRef`, preserving linkage across follow-ups.
- AI assistant responses and input hints adapt when support mode is
active.

- **Bug Fixes**
- Improved support chat state management and lifecycle handling to keep
conversation metadata and message history synchronized more reliably
with Front.

- **Chores**
- Added/updated coverage to reflect the new support-chat state and
syncing behavior.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->

---------

Co-authored-by: Alaister Young <10985857+alaister@users.noreply.github.com>
 2026-07-08 12:41:53 +02:00
Gildas Garcia 0eeeb758d8 fix: homepage accessibility fixes (#47729) 
## Problem

On the organization home page:
- you can't tab to a project card and navigate to the project
- the status filter popover cannot be open with keyboard
- the feedback popover cannot be open with keyboard

## Solution

- make the project card (which is a link) accessible with Tab
- fix the popover trigger buttons

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

* **Bug Fixes**
* Improved keyboard accessibility so project cards can be focused with
Tab navigation.
* Updated dropdown and filter popover trigger wiring for more consistent
click behavior.
* Reset the feedback flow to its starting step whenever the trigger is
clicked.
* **Bug Fixes**
* Made the home icon link explicitly focusable via keyboard navigation.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-07-08 11:44:48 +02:00
hallidayo f1c8187d17 fix: api docs not found (#47304) 
## 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?

Supabase Studio > Integrations > Data API >
[Docs](https://supabase.com/dashboard/project/_/integrations/data_api/docs)

## What is the current behavior?

Going to a route that does not exist the user just gets a blank page and
no warning.

## What is the new behavior?

User now gets redirected back to the intro docs page and and error toast
appears

## Additional context

Closes #34721


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

## Summary by CodeRabbit

* **Bug Fixes**
  * Improved handling of invalid Data API documentation links.
* If a requested table, view, or function can’t be found after loading,
users now see an error message and are redirected back to the main Data
API docs page.
* This helps prevent blank or broken documentation views when route
parameters are incorrect.

<!-- end of auto-generated comment: release notes by coderabbit.ai -->
 2026-07-08 10:00:40 +02:00