Inside this page and when we want to manually change the values, it is loading

Hi Everyone, Earlier there was an option to download SSL certificate for the database. Now, that option is no longer there.

Hi ! I’m making a Coda Pack (extension for Coda). They give these different way of getting auth through their extension :  https://coda.io/packs/build/latest/guides/basics/authentication/ but I have no idea how to manage this with Supabase Auth. Any help would be great. Thank you very much

Is there any way, even if it means manually hacking the node_modules, to see the generated postgres SQL query that's being sent to the database? I've got supabase.from queries being built but it's not returning rows which means my expectations doesn't match up with how I think the queries are being constructed.

I'm trying to come up with a way to block users from logging in / accessing information on my app. Is there a way to add a custom auth role to a user eg (banned vs authenticated) which would carry over to RLS policies that look for anon and authenticated roles? I have a lot of RLS policies set up already and I am trying to come up with a solution that is a single change that makes it so a user can no longer access things an authenticated or anon role auth user could access. Thanks

Hello, I spun up the svelte auth example. I see users in the table. However, I am unable to log in with any of the sample users. From the users list, I send a password reset. I click on the link and it takes me to the original screen with the login button rather than a password reset form. Can anyone recommend a boilerplate repo that is more complete? Thank you

Hello everyone, I've made an app with React making a bucket and the template "Give users access to only their own a top level folder named as uid", so each user can upload and delete their own photos. I was wondering if there was a template too so everyone can upload images to a common bucket and each user can delete their own and only their own images. Like a profile section to delete them and a common place to show them all like Instagram 😄 Thank you!

I'm not sure why it fails but it fails, reason is in the screenshot among with table information, is this not something i can create?

We have separate projects for staging and prod environment. I accidentally upgraded the staging project to Pro. Is there a way to undo that?

Hi guys - about 10% of my users are complaining that magic links don't work for them. It doesn't initialize the session correctly. This only happens to this 10% though. Any thoughts as to why? My logs are attached (so there is an error), but these users are telling me they're opening the link almost as soon as they get it, so it can't be a time out. Not sure what to do...

I hope this isn't too silly, but I totally understand the role of foreign key relationships in terms of joining tables and it being a kind of "pact" so the database engine knows it can trust the data types and values, and not have to deal with ambiguity. So I also then understand why you can't easily do joins without them. But it seems like if you have, say, user IDs on a "profiles" table and other table that also reference that user's ID through a foreign key relationship, that you should be able to independently remove just those records that reference "profiles". Every piece of info I come across says, well, profiles is the source, if you delete that row there you can delete everything with a cascade. I definitely don't want to kill off the source -- I just want to remove some feeder rows that use it, so that the database can be scrubbed of test data on those feeder rows. I know there's a "security definer" for bypassing RLS but it doesn't seem like (for probably good reason for data integrity) there's a bypass for foreign keys. Do I need to reset the relationship backwards? Or is there a mechanism to temporarily release the keys to perform a delete and set them back to what they just were?

I have a postgres function

add_new_query

that reads from the

profile

table and inserts into 2 tables

search_query

and

search_response

. This worked fine until I created a materialized view that reads from

search_query

,

search_response

and 1 other table. The materialized view is refreshed using a trigger function

refresh_search_feed

that is called every time there is an

INSERT

,

UPDATE

or

DELETE

on any of the three tables that used to create the view. When I call the postgres function

add_new_query

via

rpc

, I get the error:

sql
Error occurred in function add_new_query: must be owner of materialized view search_feed

I tried to change the owner of the materialized view with

ALTER materialized view search_feed OWNER to authenticated;

but this throws the following error:

sql
permission denied for schema public

I guess this is happening because the materialized view is created as a table? Is there a workaround?

Are they the same thing, or Edge function is specific to Supabase and likewise with Postgresql function?

Is it possible to get realtime updates when files are uploaded/deleted/etc.?

i am wondering how to call my edge function with my custom domain name and if the supabase.functions.invoke() method is only for db functions only? Anyone has experience with these?

Team, I am trying to develop the hello-world edge function locally: I ran this command: supabase functions serve hello-world --debug The screenshot is what i was shown: error: Module not found "file:///home/deno/functions/hello-world/index.ts". Platform: macOS I installed deno via brew; I was worried it being missing was the problem.

I am trying to install supabase CLI with https://github.com/xulioc/sveltekit-supabase-dashboard.

Larry@ASUS MINGW64 /c/__repos/_sv/supabase/sveltekit-supabase-dashboard (main)
$ npm install supabase --save-dev
npm ERR! code ERESOLVE
npm ERR! ERESOLVE could not resolve
npm ERR!
npm ERR! While resolving: sveltekit-supabase-dashboard@0.1.0
npm ERR! Found: @sveltejs/adapter-auto@1.0.0-next.80
npm ERR! node_modules/@sveltejs/adapter-auto
npm ERR!   dev @sveltejs/adapter-auto@"^1.0.2" from the root project
npm ERR!
npm ERR! Could not resolve dependency:
npm ERR! dev @sveltejs/adapter-auto@"^1.0.2" from the root project
npm ERR!
npm ERR! Conflicting peer dependency: @sveltejs/kit@1.7.0
npm ERR! node_modules/@sveltejs/kit
npm ERR!   peer @sveltejs/kit@"^1.0.0" from @sveltejs/adapter-auto@1.0.3
npm ERR!   node_modules/@sveltejs/adapter-auto
npm ERR!     dev @sveltejs/adapter-auto@"^1.0.2" from the root project
npm ERR!
npm ERR! Fix the upstream dependency conflict, or retry
npm ERR! this command with --force or --legacy-peer-deps
npm ERR! to accept an incorrect (and potentially broken) dependency resolution.
npm ERR!
npm ERR!
npm ERR! For a full report see:
npm ERR! C:\Users\Larry\AppData\Local\npm-cache\_logs\2023-02-16T18_39_53_904Z-eresolve-report.txt

npm ERR! A complete log of this run can be found in:
npm ERR!     C:\Users\Larry\AppData\Local\npm-cache\_logs\2023-02-16T18_39_53_904Z-debug-0.log

Hey Team, We are using NextJS + Supabase for auth, and have a

middleware.js

file to handle some protected routes. We are essentially using the strategy from the Supabase docs here: https://supabase.com/docs/guides/auth/auth-helpers/nextjs#auth-with-nextjs-middleware The issue we are running into is that when running page speed tests the chunk for this file is being listed as 'Reduce unused JS' because of the size of the file. Is there a way to reduce this file size that we are overlooking? Here is our final middleware.js file for reference:

js
import { NextResponse } from 'next/server'
import { createMiddlewareSupabaseClient } from '@supabase/auth-helpers-nextjs'

// This function can be marked `async` if using `await` inside
export async function middleware(request) {
    const res = NextResponse.next()
    // Create authenticated Supabase Client.
    const supabase = createMiddlewareSupabaseClient({ req: request, res })

    // Check if we have a session
    const {
      data: { session },
    } = await supabase.auth.getSession()

    // Auth condition not met, redirect to login page.
    if(!session) {   
        const redirectUrl = request.nextUrl.clone()
        redirectUrl.pathname = '/login'
        redirectUrl.searchParams.set(`redirectedFrom`, request.nextUrl.pathname)

        return NextResponse.redirect(redirectUrl)
    }
  
    return res;
}

// See "Matching Paths" below to learn more
export const config = {
  matcher: [
    /*
     * Match all request paths except for the ones starting with:
     * - api (API routes)
     * - _next/static (static files)
     * - _next/image (image optimization files)
     * - favicon.ico (favicon file)
     */
    '/((?!api|_next/static|_next/image|favicon.ico|login|profile|signup).*)',
  ],

}

I've implemented the base64 method for uploading an image to storage but the file uploaded isn't what I'm expecting. I imagined that I'd end up with a file just as if I'd dragged and dropped it into the Studio; what I'm actually getting is a file with the correct filename, size, extension, and

publicUrl

, but it's not an image. The `publicUrl `just renders a small white square (the browser doesn't register an image there), the downloadable file from Studio isn't recognised as an image file in my OS, and there is no preview image in Studio. How can I get from base64 data to a file in storage that is functional - i.e. I can see the image at it's

publicUrl

? This is my upload function -

payload

is just the

arrayBuffer

data

const { data, error } = await supabase.storage
  .from(bucket)
  .upload(`${path}${filename}`, payload, {
    contentType: "image/png",
    upsert: true
});

Im trying to clone supabase from github and I am having some trouble. I followed the developers guide (https://github.com/supabase/supabase/blob/master/DEVELOPERS.md). My git, node, and npm are all up to date. I fork and then clone the repo. I install turbo (

npm install turbo

) and get a big list of errors (too big to post in this message. I can send in txt file if you need), but from what I can tell its complaining about the installation of

libpg-query

so I run

npm cache clean --force

and then re run

npm install

, but it still gives me the same errors. ok. So I disregard the turbo install errors and try to just run

npm run dev:docs

and I get another big error and it fails. The error pretty much says that a required package

@next/mdx

could not be found. so I check my package.json for it and its in there

"@next/mdx": "^12.0.4"

. I dont know if Im following the wrong guide or missing a crucial step, but when I try to install the dependencies I am just bombarded with errors. Also I tried on a different machine and am getting the same results so this isn't an underlying issue with the packages on my specific machine. Any advice helps, thanks!

Are you able to create a custom oauth provider? I am using nextjs and have been experiencing issues with next-auth by implementing a custom provider without a .well-known. So i am curious if I am able to implement custom providers with supabase?

So I think I'm experiencing a bug (or a brainfart). Under Database -> Functions I only see the "IN ALPHA" warning and none of the functions I have uploaded. I keeep uploading new functions, and still they will not show. I AM able to attach them to Triggers though. I have logged out/ logged in/ refreshed several times over the last hour and still the problem persists. Help!

https://github.com/supabase/workflows Was it canned/postponed? Honestly I don't really need this much functionality, however I could really benefit from long running function support - they don't need to be "on the edge". Basically just want serverless workers.

Hello, I would like to sign up a user and have the user ID returned to me. However, whenever I use an api call, such as

supabase.auth.signUp

or

supabase.auth.signInWithOtp

the return values are empty:

const { data, error } = await supabase.auth.signInWithOtp({
  email,
  options: {
    emailRedirectTo: getPropertyRedirectURL(redirectTo),
    createUser: true
  }
})

console.log("data", data)
// data { user: null, session: null }

The user gets created in the database, but no return value.

Either for self-hosted or cloud, is there any information on how scalable the real-time part is?

The goal is to listen in on Postgres changes with RLS enabled and filtering on specific column values. I see these two ways https://supabase.com/docs/reference/dart/stream but it's not clear what the benefit of the channel name is when listening to Postgres changes and not using presence or broadcast. I'm also unsure how to go about using the

in

filter. My code seems to work when I use the

eq.

filter and changing the channel name doesn't seem to change anything. Also, my

messages

and

chat_rooms

table both have RLS with the correct settings for what I want. This is my existing code:

dart
  void subscribeToChat(String accountUserId, Iterable<int> chatRoomIds) {
    print("Subscribed to rooms and messages");
    _supabase.channel('public:messages:$accountUserId').on(
      RealtimeListenTypes.postgresChanges,
      ChannelFilter(
        event: 'INSERT',
        schema: 'public',
        table: constants.Table.messages.name,
        filter: 'chat_room_id=in.(${chatRoomIds.join(',')})',
      ),
      (payload, [ref]) {
        print('New message received: ${payload.toString()}');
      },
    ).on(
      RealtimeListenTypes.postgresChanges,
      ChannelFilter(
        event: 'INSERT',
        schema: 'public',
        table: constants.Table.chat_rooms.name,
        filter: 'user_id=eq.$accountUserId',
      ),
      (payload, [ref]) {
        print('New room received: ${payload.toString()}');
      },
    ).subscribe();
  }

Are there any security concerns with having your migration files in Github? It's a private repo so it shouldn't be too big of a deal. It wouldn't be the same as storing your service_key right?

Noticing I have a

handle_new_user

function which tries to write to a table that no longer exists...

My service checks the supabase backend several times a day, and I noticed a couple of failures in the logs with these kinds of messages:

2023-02-16T23:58:15.014 app[16e79027] waw [info] "message": "FetchError: request to <my supabase url> failed, reason: write EPROTO 139692654376896:error:14094410:SSL routines:ssl3_read_bytes:sslv3 alert handshake failure:../deps/openssl/openssl/ssl/record/rec_layer_s3.c:1546:SSL alert number 40\n",

Any idea what causes this? this is a regular GET that succeeds many times a day usually.

Hey guys, a quick question. In terms of the lifecycle of supabase client in a front end framework like React, Vue, Angular.. is the state of the client persistent as a singleton instance throughout the life cycle of the front end app, or does it create scoped instance/connection every request made to the database?