## Context
Having a custom report block on the project home page with a SQL that
returns a large set of results (e.g > 100k rows) causes a client side
crash with "Maximum call stack size exceeded"
<img width="400" alt="image"
src="https://github.com/user-attachments/assets/e4bb5b73-e114-4687-9d0b-a7bff328167c"
/>
This is happening due to an array spread in `computeYAxisWidth` in
`Math.max` - which am hence opting to use a `reduce` instead to mitigate
the problem.
Am also opting to apply the same autolimit logic in the SQL editor into
the `QueryBlock` here, so that we don't unnecessarily fetch a large
dataset in this UI. Added a UI indicator as well if auto limit has been
applied (So this also overlaps into dashboard scalability too)
<img width="1383" height="465" alt="image"
src="https://github.com/user-attachments/assets/08b66398-f3b8-49ce-b4a4-23c91510bd54"
/>
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **New Features**
* Report query blocks now feature automatic SQL limiting functionality,
which restricts query results to a maximum of 100 rows when enabled
* When active, query result blocks display an informational notice to
users, clearly indicating the row restriction that has been applied
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
## Problem
Our `<Button>` component breaks the default `button` contract by
redefining the `type` prop to set its variant (`primary`, `default`,
etc) instead of the button type (`submit`, `button`, etc).
This is confusing and forces to write more code when using it with
shadcn components that expect/inject the standard button props.
## Solution
- rename the `type` prop to `variant`
- rename the `htmlType` prop to `type`
- propagate the changes where necessary
- format code
## How to test
As this is just prop renaming, if it builds it's ok
---------
Co-authored-by: Ivan Vasilov <vasilov.ivan@gmail.com>
## Context
Just some clean up as I was going through stuff
- `useExecuteSqlQuery` is deprecated and not used at all
- As such `execute-sql-query` is technically irrelevant, the more
relevant file is `execute-sql-mutation`
- Hence opting to consolidate `execute-sql-query` into
`execute-sql-mutation`
- Also removing `ExecuteSqlError` since its just re-exporting the
`ResponseError` type
There's a lot of file changes but its essentially just updating the
importing statements across the files
## 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?
Previously the `overflow-hidden` on the custom report charts container
was cutting off the tooltip within the card bounds. This PR introduces
`PortalChartTooltip` here so the tooltip is able to appear outside of
the bounding area. It is still directionally aware, just relative to the
page viewport as opposed to bounding box.
| Before | After |
|--------|--------|
| <img width="376" height="341" alt="image"
src="https://github.com/user-attachments/assets/4a85901c-6b68-4f02-a13f-d2e261267348"
/> | <img width="466" height="324" alt="Screenshot 2026-06-09 at 11 45
38"
src="https://github.com/user-attachments/assets/0071c670-ef12-4df5-90fc-79df19ce98ca"
/> |
If you want to test quickly but have no data, you can ask Claude to
generate you a dummy custom report chart and then move it around, test
the long tooltips.
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **New Features**
* Tooltips can now render in a dedicated layer to improve visibility and
avoid clipping.
* Option added to enable portal-rendered tooltips in condensed/snippet
views.
* **Bug Fixes**
* Improved tooltip positioning to prevent overflow outside the viewport.
* Corrected percentage label suffixes in chart tooltips.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
## Problem
On 4XL and larger compute sizes, Supabase disk IO is sustained rather
than burst-budget based. Baseline equals max, so there is no extra burst
balance to track (the instance can still be throttled if it hits its
configured IOPS or throughput limit, but that is unrelated to a burst
credit pool). The "Disk IO Burst Balance" / "Disk IO % Remaining" /
"Disk IO % Consumed" charts in custom reports are therefore meaningless
on 4XL+, yet they are currently still offered in the picker and rendered
as empty/misleading charts.
Custom reports persist their layout, so a report created when a project
was on a smaller compute and later upgraded to 4XL+ can still contain a
saved burst chart that we need to handle gracefully.
## Fix
- Add a shared \`hasBurstableIO(infra_compute_size)\` helper in
\`DiskManagement.utils.ts\` (replaces the local
\`BURSTABLE_IO_VARIANTS\` set previously defined inline in
\`database-charts.ts\`).
- Filter \`disk_io_budget\` and \`disk_io_consumption\` out of the
custom report chart picker (\`MetricOptions\`) when the project is on a
non-burstable compute size.
- In \`ReportBlock\`, detect saved burst charts on a 4XL+ project and
render a new \`UnavailableChartBlock\` that explains the chart no longer
applies and can be removed.
- The database observability burst balance chart already gated on
\`hasBurstableIO\`; updated to use the shared helper.
- Infrastructure activity page already has equivalent handling via the
dedicated-IO admonition, so no change needed there.
Linear: FDBKPRI-1404
## Test plan
- [ ] On a project below 4XL, the custom report picker still lists "Disk
IO % Remaining" and "Disk IO % Consumed", and they render normally
- [ ] On a 4XL+ project, neither metric appears in the picker
- [ ] Open an existing custom report that contains a saved burst balance
chart on a 4XL+ project, and confirm the placeholder block renders with
a clear explanation and a remove action
- [ ] Database observability page on 4XL+: the burst balance chart
remains hidden, all other charts render
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **New Features**
* Disk IO metrics are now filtered based on your project’s compute size
so only relevant charts appear.
* Metrics tied to burstable IO are hidden when the current instance does
not support burstable IO; deprecated metrics remain excluded.
* When a disk IO chart isn’t available for your instance, the UI shows
an explanatory unavailable-chart block with text about burst-balance
limits for very large instances and optional actions to remove the
unavailable chart.
<!-- review_stack_entry_start -->
[](https://app.coderabbit.ai/change-stack/supabase/supabase/pull/46327?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 -->
## Problem
The dashboard's timezone picker (#45517) propagates to log timestamps
and the shared TimestampInfo component, but observability and reports
charts still render their X-axis labels, range labels, and tooltip
headers in the browser's local timezone. The result is jarring once a
user picks a non-local timezone: hover a chart and you get one tz, hover
a log row and you get another.
## Fix
Routes all display-side timestamp formatting in the chart layer through
the existing picker-aware helpers (\`useFormatDateTime\` /
\`formatDateTime\`) so chart UI matches the rest of the dashboard.
- **ComposedChart.utils** \`CustomTooltip\` (the hotspot — drives every
observability dashboard tooltip): reads the active timezone via
\`useTimezone\` for both the header label and the formatted timestamp.
- **AreaChart** / **BarChart**: introduce a \`formatChartDate\` helper
that honours each component's existing \`displayDateInUtc\` prop,
otherwise routes through the picker.
- **ChartBlock**: the two recharts \`labelFormatter\` arrows now close
over \`useFormatDateTime\`.
- **ChartHighlightActions**: range labels in the zoom dropdown migrated
to the same hook.
Intentionally untouched (must stay UTC):
- \`ChartHandler\` / \`ChartBlock\` \`startDate\`/\`endDate\` (API range
params, day boundary).
- \`ChartBlock.tsx:166\` explicit \`.utc()\` data-key normalisation.
- \`useFillTimeseriesSorted\` and friends (range math, no display).
## How to test
- Sign in. Open the avatar dropdown, pick a timezone different from your
browser local (e.g. Asia/Tokyo).
- Visit any project, then \`/project/<ref>/reports/database\` (or any
\`/observability/...\` page).
- Hover any chart series — the tooltip header should display the chosen
IANA name and times in that timezone.
- Click-drag a range on a chart to open the zoom dropdown — start/end
labels in the menu should also be in the chosen timezone.
- Switch back to "Auto detect" and confirm everything reverts to
browser-local.
- For an AreaChart/BarChart that uses \`displayDateInUtc\` (e.g. some
legacy reports), confirm those still render in UTC regardless of picker.
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **Refactor**
* Standardized date/time formatting across charts, tooltips, axis
labels, header/footer labels, and highlight range labels in Reports and
chart components.
* Switched to a shared, timezone-aware formatter that respects UTC
display mode or the selected picker/timezone, ensuring consistent,
human-readable timestamps throughout the UI.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
Mark provenance of SQL via the branded types SafeSqlFragment and
UntrustedSqlFragment. Only SafeSqlFragment should be executed;
UntrustedSqlFragments require some kind of implicit user approval (show
on screen + user has to click something) before they are promoted to
SafeSqlFragment.
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **New Features**
* Editor and RLS tester show loading states for inferred/generated SQL
and include a dedicated user SQL editor for safer edits.
* **Refactor**
* Platform-wide SQL handling tightened: snippets and AI-generated SQL
are treated as untrusted/display-only until promoted, improving safety
and consistency.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
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>
This PR preps the monorepo for a migration to Tailwind v4:
- Bump all Tailwind dependencies and libraries to the latest possible
version, while still compatible with Tailwind 3.
- Cleans up obsolete Tailwind 3 specific options and configs.
- Cleans up unused CSS files and fixes the CSS imports.
- Migrates all `important` uses in `@apply` lines to using the `!`
prefix.
- Move `typography.css` to the `config` package and import it from the
apps.
- Migrated all occurrences of `flex-grow`, `flex-shrink`,
`overflow-clip` and `overflow-ellipsis` since they're deprecated and
will be removed in Tailwind 4.
- Make the default theme object typesafe in the `ui` package.
- Migrate all `bg-opacity`, `border-opacity`, `ring-opacity` and
`divider-opacity` to the new format where they're declared as part of
the property color.
- Bump and unify all imports of `postcss` dependency.
## Problem
The Report Blocks section (custom dashboards) has four visual and UX
bugs: tooltip content overflows its container, Y-axis labels with 5+
digits get clipped (e.g. `10000` renders as `0000`), action buttons
become unreachable when a block title is long, and scrolling inside a
scrollable block also scrolls the parent page.
## Fix
- Remove the fixed `w-[200px]` class from `ChartTooltipContent` in
`ChartBlock` so tooltips auto-size to their content instead of
overflowing.
- Compute a dynamic Y-axis width in `ChartBlock` based on the string
length of the maximum data value, replacing the `undefined` default that
caused clipping.
- Add `min-w-0` to the label container and `shrink-0` to the actions
container in `ReportBlockContainer` so the truncation works correctly
and action buttons are never pushed off-screen.
- Add `overscroll-contain` to the scrollable SQL code and results table
divs in `QueryBlock` to stop scroll events from propagating to the page.
## How to test
- Navigate to a custom Report with multiple blocks
- Hover over a chart bar on a block with a long metric name. The tooltip
should be fully visible with no text overflow.
- Find or create a block whose Y-axis values exceed 9999 (e.g. disk
IOPS). The full number should appear on the Y-axis without any leading
digits being clipped.
- Use a block on a read replica so the label appends "of replica",
making it long. The chart-type toggle, log scale toggle, and remove
buttons should all remain visible and clickable.
- Add a SQL snippet block that returns a large table of results. Scroll
within the results table. The page should not scroll while the inner
table is scrolling.
## Before
<img width="1166" height="680" alt="CleanShot 2026-04-07 at 15 36 45@2x"
src="https://github.com/user-attachments/assets/8e7bd3c9-8319-47c9-b2d9-b194d2803809"
/>
## After
<img width="1166" height="680" alt="CleanShot 2026-04-07 at 15 36 15@2x"
src="https://github.com/user-attachments/assets/6ca5873a-cd09-4001-9cd0-932c12b6536e"
/>
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **Style**
* More consistent Y-axis sizing with dynamic widths for better label
fit.
* Improved Y-axis number formatting (K/M suffixes, sensible decimals)
for clearer tick labels.
* Simplified, more flexible chart tooltips (min-width applied; removed
fixed widths).
* Tighter report header layout so labels truncate predictably and
actions keep their size.
* Added overscroll containment to query results and SQL view to reduce
unwanted scrolling.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
---------
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
## Summary
- Adds `min-w-0` to the flex container holding the report block label in
`ReportBlockContainer`, so CSS truncation works correctly and the Remove
button stays visible even with long report names.
## Test plan
- [x] Create a report with a very long name on the Project Overview page
- [x] Verify the Remove (X) button remains visible without needing to
zoom out
- [x] Verify the report name truncates with an ellipsis
Fixes FE-2944
## Demo
<img width="938" height="390" alt="image"
src="https://github.com/user-attachments/assets/17419035-b03d-4f61-a324-8e446685a109"
/>
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **Bug Fixes**
* Improved text truncation in report block titles to prevent layout
overflow in narrow containers.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
- Deleted SQL Snippets leave a hanging block that loads forever in
custom reports, and its not possible to delete them.
- Now you can delete blocks if they get stuck loading
- Also shows correct error state when a block couldn't load because the
sql snippet was removed
## before
- stuck forever
<img width="1296" height="936" alt="CleanShot 2026-02-26 at 13 23 25@2x"
src="https://github.com/user-attachments/assets/bb65cc5f-c2a4-4027-876e-db9682ec6f3c"
/>
## after
- show error state
- allow user to delete snippet
<img width="1388" height="862" alt="CleanShot 2026-02-26 at 13 23 45@2x"
src="https://github.com/user-attachments/assets/c5d6c114-071b-4e4d-a913-25b3c788db95"
/>
---------
Co-authored-by: Joshen Lim <joshenlimek@gmail.com>
* 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.
* refactor: infra queries to use `attributes`
This PR refactors the infrastructure monitoring query code reducing duplication and unifying the API request to always be `attributes`:
• Removed the separate useInfraMonitoringQuery hook and getInfraMonitoring function that handled a single monitoring query
• Consolidated all infrastructure monitoring queries into a unified useInfraMonitoringAttributesQuery hook that handles multi-attribute requests
• Moved interval selection logic from the query layer to the consumer (InfrastructureActivity.tsx), where it can be computed dynamically based on user-selected date ranges
• Simplified query types by removing intermediate InfraMonitoringData and InfraMonitoringVariables types
• Interval is now computed in the component (defaults to 1d, switches to 1h for date ranges ≤48 hours) rather than hardcoded in the query layer
• All queries now use the unified multi-attribute endpoint with explicit parameter passing
* fix: handle single-attribute response format
As part of reducing reliance on our old daily stats pipeline, this PR adds additional deprecated properties. Especially with the revamped reports, these daily stats are barely getting called anymore and it is not worth exposing them to the customers this way.
* update onboarding
* update model and fix part issue
* action orientated assistant
* fix tool
* lock
* remove unused filter
* fix tests
* fix again
* update package
* update container
* fix tests
* refactor(ai assistant): break out message markdown and profile picture
* wip
* refactor(ai assistant): break up message component
* refactor: break ai assistant message down into multiple files
* refactor: simplify ReportBlock state
* fix: styling of draggable report block header
When the drag handle is showing, it overlaps with the block header.
Decrease the opacity of the header so the handle can be seen and the two
can be distinguished.
* fix: minor tweaks to tool ui
* refactor: simplify DisplayBlockRenderer state
* fix: remove double deploy button in edge function block
When the confirm footer is shown, the deploy button on the top right should be
hidden (not just disabled) to avoid confusion.
* refactor, test: message sanitization by opt-in level
Refactor the message sanitization to have more type safety and be more testable.
Add tests to ensure:
- Message sanitization always runs on generate-v4
- Message sanitization correctly works by opt-in level
* Fix conflicts in pnpm lock
* Couple of nits and refactors
* Revert casing for report block snippet
* adjust sanitised prompt
* Fix tests
---------
Co-authored-by: Charis Lam <26616127+charislam@users.noreply.github.com>
Co-authored-by: Joshen Lim <joshenlimek@gmail.com>
* step 1
* use mcp
* system prompt
* clean up
* space
* clean up
* add three state opt in toggle
* clean up
* todo
* hooks
* refactor opt in form and modal
* refinements
* add bedrock
* remove console
* update mcp util
* use bedrock
* remove openai sdk package
* re-add tools
* update complete endpoints
* fix: ai assistant markdown formatting (#35541)
* fix: mcp types and project ref (#35540)
* feat: more flexible aws credential provider (#35538)
* feat: more flexible aws credential provider
* fix: add AWS_REGION to turbo env vars
* change to allowed
* update complete endpoints
* add an additional permission
* refinements
* use claud 4
* legal copy changes
* update other ai functions to use bedrock
* update generate v3 copy
* remove generate sql modal
* fixes for query block
* re-add dragging to reports
* clean up
* add open ai edge function example
* use handle error from fetchers
* remove schema and lean on tools
* copy
* Assistant MCP tests (#36049)
* feat: refactor and test mcp and model logic
* fix: remove get_project tool
* fix: remove additional get_project tool references
* update copy
* Clean up, fixes, refactors
* oops
* Float errors from AI endpoionts as toasts
* Use a env var AWS_BEDROCK_PROFILE for bedrock.
* Rename the env var for AWS bedrock profile.
* feat: support custom aws bedrock env vars
* chore: add comments explaining aws credential chain
* MCP Self Hosted Check (#36185)
support self hosted
* feat: bedrock auth via vercel oidc
* Fix broken unit test
* Feeeex
* Refactor useOrgOptedIntoAi
* Remove useDisallowHipaa hook
* small system prompt change
* readd vercel packages
* fix self hosted
* increase max duration
* try more direct prompt
* max duration 90
* reduce max steps and add loading
* mono font
* backwards compat styling
* Chore/limit number of messages sent to assistant (#36388)
* Limit number of historical messages that get sent to assistant
* Update max chat history to 5
* alignment
* bump mcp server version
* Add feature flag for opt in tags (#36466)
* Add feature flag for opt in tags
* Add one more check
* security section system prompt
* rely on default link and replace image markdown
* Add custom link component to assistant message block (#36527)
* Add custom link component to assistant message block
* Update based on feedback
* Render plain text if URL is deemed unsafe
* fix mcp tools and parse data (#36593)
* Update Admonition for AI Assistant for when opt in is re-enabled (#36663)
* Update Admonition for AI Assistant for when opt in is re-enabled
* Update
* Smol fix
* Fix TS
* Tiny
---------
Co-authored-by: Greg Richardson <greg.nmr@gmail.com>
Co-authored-by: Joshen Lim <joshenlimek@gmail.com>
Co-authored-by: Ivan Vasilov <vasilov.ivan@gmail.com>
Removes a bunch of charts relying on daily data, that should not be relevant for users. Daily stats is a bad aggregation for users in itself, as the data is 24 hours or more delayed and not granular. Data from all removed charts are better looked at through logs (i.e. auth logs), as it can be aggregated and looked at in a more fine-granular way.
* Add refresh to custom reports
* Add refresh button to API and storage reports
* Add refresh button to database report
* Clean up database report
* Clean up custom report
* Add telemetry for add SQL block and dragging SQL block from assistant
* Add UI indication for drag support block in Assistant
* Add UI indication for SQL block support in SQL editor Chart options
* Feat/reports v2/create report (#33022)
* re-add chart label
* fix undefined value
* remove mouse over on chart
---------
Co-authored-by: Saxon Fletcher <saxonafletcher@gmail.com>
* Support dragging blocks from assistant to custom report
* Show toast when new snippet is created from dragging block into custom report
* Only show drag handle for block in assistant panel if in custom reports
* Feature flag dragging Assistant query block to custom report
* Optimize DX for dragging block into report from assistant
* Update import
* Remove max width
* Fix dragging non sql block causing client crash
* Fix chart suffix
* Fix tooltip value for non % values
* Misc fixes
* Support persisting non sql block line or bar view, and fix cancelling not resetting the block state
* Update
* Preserve chart config when dragging block from assistant to report
* Fix
* Standardize block designs
* Clean up
* Add warning threshold colors for infra-monitoring stats
* Address all other comments
* Swap create custom report to use PUT
