This PR migrates the whole monorepo to use Tailwind v4:
- Removed `@tailwindcss/container-queries` plugin since it's included by
default in v4,
- Bump all instances of Tailwind to v4. Made minimal changes to the
shared config to remove non-supported features (`alpha` mentions),
- Migrate all apps to be compatible with v4 configs,
- Fix the `typography.css` import in 3 apps,
- Add missing rules which were included by default in v3,
- Run `pnpm dlx @tailwindcss/upgrade` on all apps, which renames a lot
of classes
- Rename all misnamed classes according to
https://tailwindcss.com/docs/upgrade-guide#renamed-utilities in all
apps.
---------
Co-authored-by: Jordi Enric <jordi.err@gmail.com>
## Problem
We used to have a `_Shadcn_` suffix for all the shadcn form components
because we also had `formik` form components.
This is not needed anymore.
## Solution
- Remove the suffix
- Update all usages
Projects that upgrade from free → paid keep their original sub-8 GB disk until the first usage-driven auto-expand fires. During that window, the compute-and-disk settings page is broken in several ways. This PR fixed it.
## Problem
The input groups components introduced in #44282 don't have the
validation attributes when invalid. This hurts accessibility and also
break the design:
<img width="1730" height="324" alt="image"
src="https://github.com/user-attachments/assets/a3fb8d86-f3a8-46bb-aa53-d0599c11f056"
/>
## Solution
This is because the wrapper `<FormControl_Shadcn_>` passes the
validation props to its direct child.
The solution is to avoid applying them on the `<InputGroup>` and to
apply them manually on the inputs.
I also fixed a small accessibility issue by moving the addon texts after
the input so that screen readers announce them in the correct order. No
visual change for this
<img width="587" height="158" alt="image"
src="https://github.com/user-attachments/assets/1f8858ea-6659-45f9-964e-8c43a7fe14ba"
/>
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **Style**
* Unified numeric input layout by moving unit labels/suffixes (e.g.,
"seconds", "GB", "%", "connections", "digits", "IOPS", "MB/s", "rows")
to appear after their inputs for a consistent, predictable form
appearance.
* **Accessibility**
* Form controls now expose IDs and ARIA attributes from form context
when available, improving screen-reader descriptions and error
association.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
---------
Co-authored-by: Ivan Vasilov <vasilov.ivan@gmail.com>
## Problem
The Compute and Disk settings page shows all 12+ instance sizes (Nano
through >16XL) in a grid, which requires scrolling on smaller screens
like a 13" MacBook. This makes it impossible to see all three sections
(compute size, disk size, advanced settings) in a single view. Free tier
users have no need to see 8XL, 12XL, or 16XL options.
## Fix
Show only the first 6 compute sizes (Nano through XL) by default,
cutting the grid from 4 rows to 2 rows. Larger sizes (2XL and above,
plus the >16XL contact card) are revealed via a "Show all sizes" toggle
button below the grid. The grid auto-expands if the project is already
running on a large instance size so the selected option is always
visible.
## How to test
- Go to Settings > Compute and Disk on a project
- Verify only 6 cards are visible by default (Nano, Micro, Small,
Medium, Large, XL)
- Verify the Disk size and Advanced settings sections are visible
without scrolling on a 13" screen
- Click "Show all sizes" and verify the remaining sizes appear (2XL
through 16XL plus the >16XL contact card)
- Click "Show fewer sizes" and verify the grid collapses back to 6 cards
- On a project running 4XL or larger, verify the grid opens expanded by
default so the selected card is visible
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **New Features**
* Compute size options now display in a compact view with a toggle to
expand and show all available sizes
* Auto-expands to reveal the currently selected compute size when it's
not visible in the initial view
* Enhanced loading state for improved visual clarity
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
---------
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
This PR fixes some prettier issues:
- Bump and unify all prettier versions to 3.7.3 across teh whole repo
- Bump the SQL prettier plugin
- When running `test:prettier`, check `mdx` files also
- Run the new prettier format on all files
---------
Co-authored-by: Joshen Lim <joshenlimek@gmail.com>
https://linear.app/supabase/issue/FE-2226/fix-incorrect-baseline-and-max-iopsthroughput-values
- Centralized compute/disk limits in
packages/shared-data/compute-disk-limits.ts (MB/s, baseline+max
IOPS/throughput) and re-exported from shared-data.
- Hooked Docs compute/disk table into shared data, converted docs
text/table to MB/s baseline/max.
- Removed duplicated compute IOPS/throughput constants from Studio,
Studio now imports shared data.
- Updated Studio disk schema: compute-aware MAX IOPS/throughput
validation (MB/s), defaults computeSize to ci_micro, and gp3/io2 checks
use compute caps.
- Clarified disk update mutation to send throughput as MB/s despite
backend throughput_mbps field name.
- Added tests for compute-size mappings and IOPS helper logic.
- Added Infra ownership for shared compute/disk data in CODEOWNERS.
- Locked zod version via catalog and added zod dep to shared-data.
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **New Features**
* Added a reusable Compute Disk Limits table and centralized compute
disk limits dataset for dynamic display.
* **Documentation**
* Replaced static per-size tables with a component-driven MB/s view;
clarified baseline vs. burst behavior and updated guides and
troubleshooting.
* **Bug Fixes**
* Validation and UI guidance now honor compute-size limits for IOPS and
throughput.
* **Tests**
* Expanded unit tests for sizing, mappings, and edge cases.
* **Chores**
* Published shared-data exports, added a validation schema, pinned a
dependency, and added ownership entries.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
* Bump the deps, refactor deprecated code.
* Migrate keepPreviousData usage.
* Migrate all uses of InfiniteQuery.
* Fix refetchInterval in queries.
* Migrate all use of isLoading to isPending in mutations.
* Fix accessing location in claim-project.
* Fix a bug in duplicate query keys.
* Migrate all queries to use isPending.
* Revert "Fix accessing location in claim-project."
This reverts commit 2a07df64b5.
* Revert the rss.xml file to master.
* o11y: mirror and sanitize breadcrumbs
Mirror Sentry breadcrumbs as the basis for our own support logging. Also
adds more sanitization to breadcrumbs.
* feat(support form): toggle for attaching dashboard logs
Add a toggle to the support form when the category is "Dashboard bug",
to attach recent dashboard logs. Users can preview the attached logs and
opt out.
* feat(support links): dedicated support link component
Add a new component for support links, which:
- Uses the serializer for support link params to ensure
serialization/deserialization pairs correctly
- Snapshots breadcrumbs so the attached log on the support form will be
cut off at the support link click (otherwise we will get support form
actions cluttering up the log)
* tests(support form): extend timeout on flaky test
* Minor clean up
* fix(support form): allow url to specifically indicate no specified project
* minor nits
* Fix tests
* Fix tests
---------
Co-authored-by: Joshen Lim <joshenlimek@gmail.com>
* Add callout for migration
* Update copy
* Update date
* remove docs button
* Fix conditional - storage upgrade prompt should only show if list v2 is false
* Update Supabase docs URLs to use env variable
Co-authored-by: a <a@alaisteryoung.com>
* Refactor: Use DOCS_URL constant for documentation links
This change centralizes documentation links using a new DOCS_URL constant, improving maintainability and consistency.
Co-authored-by: a <a@alaisteryoung.com>
* Refactor: Use DOCS_URL constant for all documentation links
This change replaces hardcoded documentation URLs with a centralized constant, improving maintainability and consistency.
Co-authored-by: a <a@alaisteryoung.com>
* replace more instances
* ci: Autofix updates from GitHub workflow
* remaining instances
* fix duplicate useRouter
---------
Co-authored-by: Cursor Agent <cursoragent@cursor.com>
Co-authored-by: alaister <10985857+alaister@users.noreply.github.com>
* Refactor Drawer component and add date-fns dependency
Refactored the Drawer component for improved slot-based structure, updated styles, and added 'use client' directive. Added 'date-fns' as a dependency in design-system, updated tsconfig paths for icons, and marked ToggleGroup as a client component.
* nit: add env for svg path
* fix: instructions
* accent color docs and basic tidy
* copy value feature
* improve color contrast
* increase contrast on text-warning in light mode
* update changelog
* replace outdated text-warning utility classes
* remove redundant warning-600
* minor design-system docs updates
* docs updates
* remove unused brand-button class
* update docs
* fix: restore brand default
* update docs brand text color
* low hanging branded text fruit
* Nit refactor and clean up
* re-add Kemal’s README instructions for hot reload
---------
Co-authored-by: Jonathan Summers-Muir <MildTomato@users.noreply.github.com>
Co-authored-by: Joshen Lim <joshenlimek@gmail.com>
* Replace all usage of useProjectContext with useSelectedProjectQuery
* Replace all usage of useSelectedProject with useSelectedProjectQuery
* Replace all usage of useProjectByRef with useProjectByRefQuery
* Replace all usage of useSelectedOrganization with useSelectedOrganizationQuery
* Deprecate useSelectedProject, useSelectedOrganization, and useProjectByRef hooks
* Deprecate ProjecContext
When downgrading from a larger compute size to <large while having provisioned IOPS or throughput, we would leave the disk as-is, even though the configuration is unsupported and then block any disk changes because the instance size is too small, essentially dead-locking the customer on the disk attributes, unless they bump back up to a higher instance size.
PR addresses multiple issues:
- Automatically reset disk attributes to default when customers downgrade to <large instances
- Do not use the baseline compute IOPS/throughput for the fields, as this leads to accidentally over-provisioning by customers
When translation tools go over prices, they wrongfully convert USD into other currencies without changing the price. $25 becomes Rp25 (Rupiah).
By using `translate='no'` we avoid the translation of pricing info. There are some tooltips left that are not covered yet, but this should be the majority already.
Kept the docs Pricing component as simple as possible, just wrapping in a span to avoid translation, not modifying any formatting.
* fix: consistent disk unit
* Update apps/docs/content/guides/platform/manage-your-usage/disk-throughput.mdx
* Update apps/docs/content/guides/platform/manage-your-usage/disk-throughput.mdx
* Update apps/docs/content/guides/platform/manage-your-usage/disk-throughput.mdx
* docs: makes use of singular/plural for metrics that are billed hourly consistent.
* ci: Autofix updates from GitHub workflow
* fix: instance is Mbps
* fix: description Mbps
---------
Co-authored-by: Thomas <31189692+ecktoteckto@users.noreply.github.com>
Co-authored-by: github-tidy-bot <github-tidy-bot@supabase.com>
* Init
* Initial set up for hooking up supavisor and pgbouncer
* Hook up pgbouncer status check after swapping pooler type
* Add check for nano compute for switching to pg bouncer
* Add check for ipv4 addon
* Remove expect error tag
* Add badge to select options for pooler types
* Remove statement mode
* Resolve undefined problem with react hook form
* Fix
* Update UI texts from PgBouncer to Dedicated Pooler
* Feex
* FEEX
* Fix
* Small update to UI
* Smol update
* disable selection of io2 volume type if project region is not supported
* Update apps/studio/components/interfaces/DiskManagement/fields/StorageTypeField.tsx
Co-authored-by: Alaister Young <alaister@users.noreply.github.com>
* Fix import
---------
Co-authored-by: Alaister Young <alaister@users.noreply.github.com>
* Add disk autoscale configuration parameters in compute and disk settings
* TS fix
* Fix
* Fixes based on comments
* Address all feedbacks
* Small tweak
---------
Co-authored-by: Alaister Young <a@alaisteryoung.com>
* Add restrictions for orioledb technical preview
* Add callouts to pgvector and postgis if orioledb
* Restrict restore to new project for orioledb
* Scaffold client side validation for preventing org upgrade if org has oriole db present
* Hook up proper logic for oriole
* Fix
* Remove console log
* Fix type
* Disable version selector if only one version is available
* chore: oriole badges
* UI updates based on requests
* Update copy
* Fix
* Dont open assistant if opt is selected
* Fix
* Fix
* Update badge
* Add feature flag for orioleDB
* Feature flag oriole check in plan update
---------
Co-authored-by: Paul Cioanca <paul.cioanca@supabase.io>
* Fix percentage width for disk space bar
* Add warning in disk size section if 90% + within cooldown
* Show read only warning in compute and disks
* Fix copy
* Add conditional
* inited. added disk config to a new page
* add instances
* move moar
* moved things around. billing badges updated. compute added
* tidy
* new components
* form now dynamically updating itself
* updated compute form. moved warning panels. added collapsible for advanced options
* review dialog now only showing what is relevant
* Update DiskManagementForm.tsx
* compute sizes now a reccomendation
* fix old form
* started adding flags
* removed unused code. fixed issue with IOPS price showing on smaller compute
* moar clearning
* IOPS logic wrong way round
* type fixes
* start adding better error handling
* TIDY
* moved everything to own file
* tidy
* fix hydration issue
* moved some components around
* clean up
* inline errors
* update form message
* Update DiskManagementForm.tsx
* error fields fixed. some formatting issues. nano added as an option
* fix constants
* add some plan restrictions
* moar
* units updated. labels updated
* Update DiskManagement.schema.ts
* fix a ton of type issues
* text udpates
* add panel to suggest switching to io2
* more notice board stuff
* number formatting. moved a file
* Update DiskManagementForm.tsx
* remove console logs
* upgrade comms. more type fixes
* add empty states for the old areas
* more links
* updated some label issues
* hide labels when chart is active
* Update DiskManagement.utils.ts
* Delete next-env.d.ts
* Update DiskManagementForm.tsx
* Update DiskManagement.schema.ts
* text updates
* Update DiskManagement.constants.tsx
* Update next-env.d.ts
* Update next-env.d.ts
* Small clean uop
* Clean up empty files
* Clean up spelling
* Clean up more
* Fix typo in file name
* Clean up import statements
* Update DiskManagementForm.tsx
* fix issues
* Update ProjectLayout.tsx
* Remove unused import
* Fix
* Address nit
* Update database.tsx
* remove supress toast
* Update DiskManagement.schema.ts
* Update database.tsx
* change upgrade comms
* Update DiskManagementPanelForm.tsx
* fixes
* fix button size on old form
* Update DiskManagementForm.tsx
* Update StorageTypeField.tsx
* update labels on compute
* dont show banner when infra is FLY
* update comms. hide disk config for FLY
* Fix TS
* Last round of clean upo
* fix message state
* fix message
* Fix TS
* Update DiskManagement.utils.ts
* fix errors
* Update BillingChangeBadge.tsx
* fixed some label issues
---------
Co-authored-by: Joshen Lim <joshenlimek@gmail.com>
Ivan Vasilov