## Context
There's some changes for Vector Buckets with the wrappers extension on
version 0.5.7 and above
The "Query from Postgres" CTA (which runs an import foreign schema
command) no longer accepts the bucket name as part of the schema options
(ref
[changelog](https://github.com/supabase/wrappers/releases/tag/v0.5.7)),
and instead takes the bucket name as the remote schema name.
So changes here are to update the SQL for importing foreign schema for
vector buckets if wrappers extension is above 0.5.7
Also - the `::embd` type no longer seems to exist in 0.5.7 as well, so
updated the example code too
## To test
Can probably test on a new US East 1 project on the preview
- [ ] Verify creating a vector bucket table + can query from postgres
(via the Insert vectors CTA after creating a table)
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **Bug Fixes**
* Fixed breadcrumb navigation links for vector bucket pages.
* **Improvements**
* Unified loading state for schema workflows and disabled actions while
loading.
* Enhanced version compatibility checks for foreign schema imports to
support newer extension syntax.
* Adjusted generated SQL for vector data to use the appropriate
embedding syntax based on extension version.
<sub>✏️ Tip: You can customize this high-level summary in your review
settings.</sub>
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
* fix(studio): use paginated buckets query for useSelectedBucket
Part of the effort to migrate all buckets list queries to the paginated
version. This migrates the useSelectedBucket hook, which is used on the
bucket details page to fetch the bucket info.
* fix(studio): flash of stale content when switching buckets
There's a flash of stale content when switching between buckets in the
File Explorer. This happens since the change to useSelectedBucket,
because when the bucket changes, the new data is no longer available
immediately. It must be awaited, and meanwhile `columns` in global
storage state is stale. This commit adds a check for whether the state
is stale by checking the current ID against the stored ID.
---------
Co-authored-by: Ali Waseem <waseema393@gmail.com>
* Validate against empty name in FileExplorerRowEditing.tsx
* Update test
* Fix
* Opt to revert changes if saving empty string
* Nit
* Attempt to write tests with claude
* Trim whitespace in FileExplorerRowEditing when saving
* Refactor to remove usage of useQueryStateWithSelect in auth policies and vector bucket
* Refactor to remove usage of useQueryStateWithSelect in edge functionn secrets
* updated middle click, ctrl click to open new tab
* updated PR review
* updated to create a wrapper for handling navigation
* updated file spelling
* updated to add edge functions button
* updated navigation for command space and enter
* added inset focus class
* Add shimmering-loader CSS to ui-patterns.
* Import the shimmering-loader classes from the ui-patterns component.
* Remove ShimmeringLoader from studio.
* Migrate studio to use ui-patterns/ShimmeringLoader.
* Migrate away from using default import for ShimmeringLoader.
* Fix the css imports in docs and studio.
Iterator.prototype.map() is only newly baseline available, which is
causing problems with older browsers.
Switched to Array.from(<iterator>) since that (and Map as an iterable)
date from 2015/2016
The global storage size validation depends on an unpaginated buckets
query to determine whether it is lower than any individual bucket's
cutoff. This causes a problem for users with tens of thousands of
buckets.
There's a bit of a UX/performance problem here, because in order to
determine whether any bucket's `file_size_limit` exceeds the global
setting, we need to get the max `file_size_limit` of `storage.buckets`
-- however, that column is not indexed.
My workaround is:
- Below a certain threshold (10,000) buckets, the query for max
`file_size_limit` is automatically run on form submit.
- Above that threshold, the user must confirm whether they want to run
the query. They're still allowed to change the storage config without
running it -- this does open a loophole where they can have a global
storage setting lower than an individual bucket's file size limit, but
though this is a potentially confusing situation, it's not strictly an
error.
---------
Co-authored-by: Joshen Lim <joshenlimek@gmail.com>
## Summary of changes
Buckets policies page at /project/_/storage/files/policies now uses the
paginated buckets endpoint to fetch buckets.
## Before
Used a non-paginated query.
## After
To support a paginated query, we now have:
- Infinite scroll loading behavior
- Fixed some of the buckets utilities to account for policy references
to buckets that haven't loaded yet
* 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.
* init request upgrade plan modal
* Consolidate UpgradeToPro components across UI + use UpgradePlanButton, UpgradePlanButton determines whether to show request upgrade CTA
* Allow to pass icon to Admonition
* Tiny fix on upgrade plan button CTA to consider addons
* Hook up upgrade request endpoint
* Update API types
* remove hardcode
* Add request upgrade CTA in plan side panel
* Fix disk compute
* Show request upgrade button for change to large compute
* Nit
The bucket list endpoint now accepts pagination options. Adapting the
storage buckets table UI to paginate in pages of 100 buckets at a time,
with infinite loading/scrolling.
* fix(studio): auto-select required permissions for storage policy operations
when selecting update or delete, automatically select the select
permission since these operations require it for the client library
methods to be correctly highlighted.
when deselecting select, automatically deselect update and delete
to maintain consistency.
fixes#40923
* Helper msg
* wrap
* stamp: improve code style
---------
Co-authored-by: kallebysantos <kalleby_santos@hotmail.com>
* Midway
* Hook up create/delete table, and update all iceberg RQs with mgmt API routes
* Add insert data CTA, update create table logic for decimal and fixed types
* Clean up
* Nit
* Fix test
* Add examples.
* Move the button for creating a vector bucket outside of the dialog.
* Install the wrappers extension if missing when creating a vector bucket.
* Refactor the quick start code for vector buckets.
* Fix a type error.
* chore(studio): vector tables UI improvements (#40598)
* Add examples.
* clearer examples
* clearer dimension samples
* quick start as sheet
* improve structure
* cleanup
* icons and text size
* improve snippet code
* icons
* sql editor link
* cleanup
---------
Co-authored-by: Ivan Vasilov <vasilov.ivan@gmail.com>
* Minor naming fix.
* Fix the link to the Table Editor.
* Smol fix + refactor and clean up
* Nit
* Nit improvement
* Make the SQL option default.
* fix docs link
* fix example code
* remove redundant card
* Check if the project has secret keys or service role keys. Remove the NEXT_PUBLIC from the env vars to avoid exposing them to the FE.
---------
Co-authored-by: Danny White <3104761+dnywh@users.noreply.github.com>
Co-authored-by: Joshen Lim <joshenlimek@gmail.com>
Joshen Lim