Does anyone see an issue with this code? I'm attempting to create a function and trigger but getting a syntax error.

CREATE
or REPLACE FUNCTION create_signature_code()
returns trigger as $$
begin
  UPDATE public.signature_activity
  SET signature_code = '1234'
  WHERE id = new.id
  
  return new;
end;
$$ language plpgsql security definer;

create trigger on_new_signature_activity
  after insert on public.signature_activity
  for each row execute procedure public.create_signature_code();

Failed to run sql query: syntax error at or near "return"

Im trying to delete some items but im getting this error. what could do?

I have some large queries which are being generated by postgrest and are very slow which I am trying to figure out how to optimize. Unfortunately the query is larger than the track_activity_query_size so it gets truncated in the logs and everything. I wondered how I can enable pgsm_enable_query_plan config to have it show me the query plan, or extend track_activity_query_size. This is on hosted supabase

I have a table ( test_load ) and another called ( test_account ) - within in the test_load table is a column called rtms__Carrier__c that is a UUID of the test_account ID field - so im trying to make a foreign key relationship but i keep getting this error : Failed to run sql query: insert or update on table "test_load" violates foreign key constraint "test_load_rtms__Carrier__c_fkey" i know that this should work but im not sure how to fix this error

I have this error in CI (but not locally): https://github.com/hamza1311/firestorm/actions/runs/5014021366/jobs/8987795179#step:9:189 Can anyone help fix this? The supabase type is https://github.com/hamza1311/firestorm/blob/665a8f10dd0cf8db067670905763013925144479/src/lib/supabase.ts#L21

Connection pooling is the default when using the Javascript client correct?

js
  const { data, error } = await supabase
    .from("url_data")
    .select('*, clicks(*)')
    .eq('slug', slug)
    .single()`

i have items clicks with the same url_data id but its returning []

Can't seem to delete a function I made since it does not show up in database functions and its for the protected auth schema. Tried via the SQL editor but it keeps telling me 'Failed to run sql query: could not find a function named ...'

Hi its my first time on this channel, im using signinwithotp function. Is there a way i can distinguish between a new auth user or an already auth user? Thanks

If I use SupaFly to deploy Supabase to Fly.io and then use Auto Scaling, will I be able to use more websockets? Previously, I used Upstash to operate Serverless Redis as much as I needed, But I was wondering if I could auto-scaling the Supabase Realtime Subscription. ( I am currently looking at a distribution tool called SupaFly ) https://github.com/nicholasoxford/SupaFly

If I am only using RSC with the new Next.js, do I still need to use client side context to create a SupabaseProvider for the authentication to work?

Dear everyone Self-Hosting: Can Supabase deploy a product, because I try to deploy in docker local, the url is always localhost:3000 and also deploy to vps ip_vps:3000 has no security attached so please ask Is the supabase only at the research level or can be used for real? thank you guys

Does anybody know if there is a function in Supabase that syncs the form state with the database? The idea is to have it update the db when there is a change of value in the form. I have seen the RLS documentation but it would seem the most suitable function would be using

postgress_change

when subscribed to the channel, but that just updates the form when there is a change in db on not vice versa. Thoughts of an own implementation would be to self implement a debounce rate and a mix of

insert

and

upsert

calls to track the change of the form state before loading it into db

onChange

. Doing this with nextJS. Any feedback regarding the implementation or reusable patterns are also much appreciated.

I followed the instructions on https://supabase.com/docs/guides/getting-started/local-development#enabling-local-logging to a T. I created a Google Cloud account (provided CC too, yay...). Sometimes I'm getting a ton of error messages when starting the containers. Sometimes I don't, but there is no log icon in the Studio menu. And I see a bunch of errors in the vector container log (see attached, something to do with being rate limited by Google). I do not do anything else with BigQuery than launching the Supabase CLI, so I'm not sure why I'm being rate limited. https://cdn.discordapp.com/attachments/1108943696038199327/1108943696193404988/message.txt

Is there a way to see a query log/audit using the local dev Supabase? For example using the pgAudit extension locally?

I'm trying to pull the docker image down for my rpi and it keeps pulling the amd64 image no matter what I try. Anyone else able to pull the correct image for arm based systems? This is what I'm using:

postgres:
    container_name: postgres
    image: supabase/postgres
    platform: linux/arm64
    healthcheck:
      test: pg_isready -U postgres -h localhost
      interval: 5s
      timeout: 5s
      retries: 10
    command:
      - postgres
      - -c
      - config_file=/etc/postgresql/postgresql.conf
      - -c
      - log_min_messages=fatal # prevents Realtime polling queries from appearing in logs
    restart: unless-stopped
    ports:
      # Pass down internal port because it's set dynamically by other services
      - "5432:5432"
    environment:
      # POSTGRES_HOST: /var/run/postgresql
      # PGPORT: ${POSTGRES_PORT}
      # POSTGRES_PORT: ${POSTGRES_PORT}
      # PGPASSWORD: ${POSTGRES_PASSWORD}
      POSTGRES_PASSWORD: postgres
      # PGDATABASE: ${POSTGRES_DB}
      # POSTGRES_DB: ${POSTGRES_DB}
    volumes:
      - ${STORAGE_ROOT}/data/postgres/realtime.sql:/docker-entrypoint-initdb.d/migrations/99-realtime.sql:Z
      # Must be superuser to create event trigger
      - ${STORAGE_ROOT}/data/postgres/webhooks.sql:/docker-entrypoint-initdb.d/init-scripts/98-webhooks.sql:Z
      # Must be superuser to alter reserved role
      - ${STORAGE_ROOT}/data/postgres/roles.sql:/docker-entrypoint-initdb.d/init-scripts/99-roles.sql:Z
      # PGDATA directory is persisted between restarts
      - ${STORAGE_ROOT}/data/postgres/data:/var/lib/postgresql/data:Z

Hi folks, as in the title I would like to create new table dynamically using javascript. So for example each user can create a multiple projects on my page. Each project will contain a lot of data. - so I would like to build the logic that if the user clicks create new project button, it creates new table in supabase with table name I provide in the parameters and certain schema. And ideas how do I do it. I appreciate all the help. thanks

Anyone have an idea wha'ts wrong with my code? I can't seem to make it work

I'm looking to use a FastAPI application to pass-through specific Supabase PostgREST endpoints I'd like to make available to customers. Furthermore, I'd like to add custom endpoints for the business logic specific endpoints. I'm wondering if anyone has a similar setup and how they solved to following items: - How to combine the openAPI docs created by PostgREST with the endpoint documentation of the custom FastAPI endpoints - How does one use the end-users access tokens to also make sure RLS/Auth keep functioning as when directly interacting with PostgREST - How does one use Supabase Auth for the custom FastAPI endpoints - Preferred way of linking FastAPI and Supabase (direct DB sessions or going through PostgREST) Eager to hear any idea's or experiences with similar setups!

If I import types like this: //@deno-types="https://esm.sh/@types/express"; supabase showing error: error: Relative import path "stream/promises" not prefixed with / or ./ or ../ and not in import map from "https://esm.sh/v122/@types/node@18.16.9/stream/promises.d.ts" If you want to use a built-in Node module, add a "node:" prefix (ex. "node:stream/promises"). at https://esm.sh/v122/@types/node@18.16.9/stream/promises.d.ts:42:19

I have come across numerous discussions on Discord and GitHub regarding Supabase's role-based user access. It appears that Supabase has indeed implemented a role system, which can be utilized in their Row-Level Security (RLS) feature. However, ,it seems that the role functionality is not functioning as expected. I am wondering if I may have overlooked any crucial steps or requirements. Could you kindly advise if there is anything I might be missing in order to properly utilize Supabase's role-based user access? First, I created a role named 'webadmin' within the Supabase dashboard, accessible at https://app.supabase.com/project/_/database/roles. Next, I proceeded to create Row-Level Security (RLS) for a specific table. Subsequently, I executed the following SQL command to update the role of the desired user:

UPDATE auth.users
SET role = 'webadmin'
WHERE email = 'akxxxxxx@gmail.com';

Additionally, I granted the 'webadmin' role to the 'authenticator' user by executing the following command: GRANT webadmin TO authenticator; So can i have any advice ?

Hi guys, In our Startup we want to create a small database to be used by tech and non-tech people. Could you help me clear out some doubts regarding databases in Supabase? - is the UI friendly for a non-tech person to make changes into a cell? - 3 different people in the org may be changing different parts of the table. Since different kind of people will be changing the table, it'd make sense to have separate "safe" tables where they can make their changes (independently), and then merge it into production. What's the approach you recommend? - is there a log for changes in a certain table? We use SQL but we're not used to managing a SQL database for envs (prod / stage /etc) bc we've been using firebase since 2019. Thanks in advance! 🙂

Hi, Is it possible to create a cron job from client or edge function ? I tried with database function but I get Invalid SQL query

Hi, im using signInWithOTP and i want to auto login without having to receive a code each time. In the previous login page i used password so i managed to store the details of the user as asyncstorage with react native and re log him automotaicaly. Now i moved to signInWithOTP. Is there a way i can have a similar approach pleasE? const { data, error } = await supabase.auth.signInWithPassword({ email: loginDetails.email, password: loginDetails.password, }); if (data.session) { dispatch(setSessionToken(data.session.access_token)); dispatch(setUserID(data.user.id)); setIsSocketConnected(true); }

So I don't want my frontend developer to know the table name. Is it a good idea to use rpc to perform basic actions like select, insert, update and delete. Will it have any impact on performance?

I know the regular mail template settings are not available from the self-hosted Supabase Studio. But I can't find anywhere what makes those emails either. Is there any way of changing those mail templates whatsoever?

Hey guys, a new Supabase user here. I'm stuck with the following issue: I have a table called

connections

with the following structure:

table connections

id  |  user_id  | data

where

user_id

is a foreign key pointing to

auth.users.id

. Now I want to insert a new record via Javascript to the

connections

table (it's only possible when the user is authenticated):

js
const { data, error, ...rest } = await supabase
  .from("connections")
  .insert({
    data
  })

What I would like to happen is that

user_id

would be inserted automatically based on the currently authenticated user - basically just inserting

auth.uid()

. Unfortunately, I can't figure out how to achieve that without me manually passing the data in the payload like so:

js
const { data, error, ...rest } = await supabase
  .from("connections")
  .insert({
    data,
    user_id: supabase.getSession().session.user.id
  })

which doesn't seem right... I was thinking of creating a trigger that adds the

user_id

after the connection record is created but I don't know how to retrieve the currently authenticated user's ID in a trigger. Any help would be highly appreciated! Thank you!

Hi, i'm currently try to setup supabase (self hosted via kubernetes) on our existing database project. we are using timescale and have a lot of bucket tables. in fact so many that the database -> table dashboard does not load due to a timeout. is there a way to tell supabase that it should exclude certain schemas and/or tables?

How do i get an access token that doesnt expire to auto login a user after closing the app using signInWithOTP please?

Hi folks, since Supabase basically lets you send freeform SQL from a browser, how do I stop a user from trying something silly like a

cross join

between a pair of million row

generate_series

-es and crashing the server due to RAM/disk exhaustion?