Hey folks! I followed the tutorial from https://supabase.com/docs/guides/getting-started/tutorials/with-nextjs#launch and when i run the

npm run dev

command I am greeted with this error Any idea what went wrong? Fresh install, only changes and updates are what is in that document.

Draftbit (front end React app) needs a URL passed through to show images. I can manually add a URL in Draftbit from Cloudflare and it works (shows image). However, I can not figure out how to pass through a URL from Directus/Supabase to do the same (as part of the database table). My stack: Draftbit (frontend) --> Supabase (Postgres) --> Directus ( API layer, local install) Happy to do this manually but ideally I would find a solution to do the following: -- tie the URL to my "article" database/table -- upload article info with image URL -- utilize a CDN for efficient image processing. There are a few posts/searches that talk about errors for getPublicUrl returns but I am looking for a simple, GUI/low code based way to tie my image URL (fromCloudflare or Supabase object storage) and tie it to my Supabase "article" table. Any suggestions would be greatly appreciated. I have found very good, detailed solutions for solving this programmatically but nothing in a low code fashion. For reference, here are what seem to be very good solutions that are more involved. Supabase using their cloud storage (https://github.com/supabase/supabase/discussions/5196) How to manage your content with Directus and Cloudinary -- code included for building a Vue front end for Directus that points to Cloudinary (https://medium.com/birdie-ai/how-to-manage-your-content-with-directus-and-cloudinary-ded6ddd9a4b1)

I have a table

asset

with a foreign key (

asset_type_id

) that links it to table

asset_type

.

I've been experiencing state management issues when saving Queries for the past week. Using a single browser tab, saved Queries are randomly overridden. This is an issue because we're using Queries for scheduling cron jobs that our app's dependent on; however, if a cron job is erased or updated without our team explicitly doing it, it creates issues.

Hi, I am trying to setup a project using the supabase auth-helpers-sveltekit package and keep getting this error. Found this (https://giterhub.com/repository/activity/supabase-community/supabase-auth-helpers) online and I have a hooks.client.ts and hooks.server.ts file setup with $lib/supabase imported in both but it still just gives me this error. Anyone know what the problem might be?

Hey guys, I'm getting started with Superbase and was wondering if there's a way to protect stored files i.e render the files ( image or video ) on the client-side but prevent any download of that file. I see it's possible to protect the file with the permission/auth system of Superbase but after giving access to the user, is there a way to prevent any download. I'm currently thinking about setting up a proxy for the backend but was wondering if there's a better way

Hey guys, given the following two tables

CREATE TABLE IF NOT EXISTS problems
(
    id BIGINT GENERATED ALWAYS AS IDENTITY PRIMARY KEY,
    statement text not null,
    placeholder text not null,
    tests text not null,
    created_at timestamptz DEFAULT CURRENT_TIMESTAMP NOT NULL,
    updated_at timestamptz DEFAULT CURRENT_TIMESTAMP NOT NULL,
    deleted_at timestamptz
);

ALTER TABLE problems enable row level security;

CREATE TABLE IF NOT EXISTS problems_mapping
(
    id BIGINT GENERATED ALWAYS AS IDENTITY PRIMARY KEY,
    problem_id bigint not null,
    course_id bigint not null,
    section_id bigint,
    created_at timestamptz DEFAULT CURRENT_TIMESTAMP NOT NULL,
    updated_at timestamptz DEFAULT CURRENT_TIMESTAMP NOT NULL,
    deleted_at timestamptz,
    unique(problem_id, course_id, section_id),
    FOREIGN KEY(course_id) REFERENCES courses(id) ON DELETE CASCADE,
    FOREIGN KEY(section_id) REFERENCES sections(id) ON DELETE CASCADE,
    FOREIGN KEY(problem_id) REFERENCES problems(id) ON DELETE CASCADE
);

How do I translate the following query

select p.* from problems as p inner join problems_mapping as pm on p.id = pm.problem_id and pm.course_id = 1;

With the javascript library?

Hi, couldn't find any info about this. New to deno as well. My worry is my edge functions are failing and I never find out about it. Typically, you wold use a monitoring solution for this. Doesn't look like supabase has it yet. So what are folks using to monitor their edge functions? Thanks!

Hi, I have a table with a trigger to update updated_at column before every update. I am trying to change a column value for a specific row in this table from the dashboard. I get this error. Failed to run SQ query. moddatetime must be fired for row. So it looks like the trigger is preventing me from update a column. If I disable it, I am able to update column. Is there a way around this? I would like to be able to update a column without constantly disabling and enabling the trigger. Thank you!

using the following import from the dijonmusters/happy-days github repository:

import Stripe from "https://esm.sh/stripe?target=deno";

But getting the following error message when deploying

Bundling create-stripe-customer
Error: Error bundling function: exit status 1
file:///src/index.ts
error: Uncaught (in promise) Error: Relative import path "http" not prefixed with / or ./ or ../
      const ret = new Error(getStringFromWasm0(arg0, arg1));
                  ^
    at __wbg_new_8d2af00bc1e329ee (https://deno.land/x/eszip@v0.30.0/eszip_wasm.generated.js:312:19)
    at <anonymous> (https://deno.land/x/eszip@v0.30.0/eszip_wasm_bg.wasm:1:79439)
    at <anonymous> (https://deno.land/x/eszip@v0.30.0/eszip_wasm_bg.wasm:1:1388039)
    at <anonymous> (https://deno.land/x/eszip@v0.30.0/eszip_wasm_bg.wasm:1:1862894)
    at __wbg_adapter_18 (https://deno.land/x/eszip@v0.30.0/eszip_wasm.generated.js:146:6)
    at real (https://deno.land/x/eszip@v0.30.0/eszip_wasm.generated.js:130:14)

i'm using auth supabase and want to send nice looking emails to users. i found tailwind email templates but to display properly i have to include the tailwind cdn in the message. supabase removes the header from the template and only keeps the body. how can i fix this?

I've been trying to use

signInWithOtp

(on NodeJS) and it returns

AuthRetryableFetchError: {}

. Looking into the auth logs I see the following error:

`
Event Message
running db migrations: error executing migrations/[REDACTED]_modify_users_email_unique_index.up.sql, sql: -- this change is relatively temporary
-- it is meant to keep database consistency guarantees until there is proper
-- introduction of account linking / merging / delinking APIs, at which point
-- rows in the users table will allow duplicates but with programmatic control

alter table only auth.users
  add column if not exists is_sso_user boolean not null default false;

comment on column auth.users.is_sso_user is 'Auth: Set this column to true when the account comes from SSO. These accounts can have duplicate emails.';

create unique index if not exists users_email_partial_key on auth.users (email) where (is_sso_user = false);

comment on index auth.users_email_partial_key is 'Auth: A partial unique index that applies only when is_sso_user is false';

alter table only auth.users
  drop constraint if exists users_email_key;
: ERROR: cannot drop constraint users_email_key on table users because other objects depend on it (SQLSTATE 2BP01)

I have little understanding of SQL so some help of what's going on and what to do would be much appreciated.

Hi, At the moment I can log in and get the user user but as soon as I refresh the page or close the app and reopen these details are gone. But I can see the session in the localstorage in web developer. Thank you!

I have an issue that i cannot resolve. I keep getting an exception for a simple select to my table, which has policy for SELECT to Enable read access for all users. This is my query

dart
supabaseClient
          .from('users')
          .select<PostgrestListResponse>('id')
          .eq('username', userName)
          .limit(1)
          .single();

and the stacktrace

#0      PostgrestBuilder._parseResponse
package:postgrest/src/postgrest_builder.dart:323
#1      PostgrestBuilder._execute
package:postgrest/src/postgrest_builder.dart:198
        <asynchronous suspension>
#2      PostgrestBuilder.then
package:postgrest/src/postgrest_builder.dart:400
        <asynchronous suspension>

{
  "headers": [
    {
      "cf_cache_status": "DYNAMIC",
      "cf_ray": "78756a92c6578ee0-SOF",
      "content_length": null,
      "content_location": null,
      "content_range": null,
      "content_type": "application/vnd.pgrst.object+json; charset=utf-8",
      "date": "Tue, 10 Jan 2023 12:29:06 GMT",
      "sb_gateway_mode": null,
      "sb_gateway_version": "1",
      "transfer_encoding": "chunked",
      "x_kong_proxy_latency": "0",
      "x_kong_upstream_latency": "3"
    }
  ],
  "origin_time": 88,
  "status_code": 406
}

What i am trying to achieve is to get an unread count in threads table and return threads columns which i am getting, the problem i am facing right now is that i cant populate user_id which is a relation to profiles table so i am getting this error. This is the sql function:

CREATE OR REPLACE FUNCTION get_threads_by_user_id(userId uuid)
returns table(id uuid, whom uuid, subject text, user_id uuid, unread_count bigint) AS $$
BEGIN
    RETURN QUERY 
    SELECT threads.id,threads.whom,threads.subject, threads.user_id, COUNT(CASE WHEN threads_message.is_read = false THEN 1 ELSE NULL END) AS unread_count
    FROM threads
    JOIN threads_message ON threads.id = threads_message.thread_id
    WHERE threads.user_id = userId
    GROUP BY threads.id;
END;
$$ LANGUAGE plpgsql;

and this is the supabase call with rpc:

const { data, error } = await supabase
      .rpc('get_threads_by_user_id', {
        userid: userId,
      })
      .select('*, user_id(*)');

Is it possible to download the souce of a function I have already deployed? I stupidly deleted it from my local disk before i pushed it to git 😦

I was setting up an edge function to allow users to invite others via email but hit an error that supabase.auth.api.inviteUserByEmail doesn't exist. I downgraded createClient from https://esm.sh/@supabase/supabase-js@2 to https://esm.sh/@supabase/supabase-js@1.33.1 and it works as expected in the documentation https://supabase.com/docs/reference/javascript/auth-admin-inviteuserbyemail I am using the SERVICE_ROLE_KEY not the ANON_KEY to create the client. Was there some refactoring in the latest release that hasnt been updated in the documentation yet?

Hi, I’m trying to look for a more in depth documentation on storage with examples. Is this the only resource there is? https://supabase.com/docs/guides/storage

is this expected behaviour ??

I'm not sure if function with service role key okay to be called in client side (Flutter) or not, but here is my code (for Deno):

javascript
const supabase = createClient(supabaseURL!, supabaseServiceRole!, {
      auth: {
        autoRefreshToken: false,
        persistSession: false,
      },
      global: { headers: { Authorization: req.headers.get("Authorization")! } },
    });

    const { data: isAdmin }: { data: boolean | null } = await supabase
      .rpc("is_admin")
      .single();

    if (!isAdmin) {
      return new Response(
        JSON.stringify({ message: "User is not Super Admin." }),
        {
          headers: { "Content-Type": "application/json" },
          status: 403,
        }
      );
    }

    const adminAuth = supabase.auth.admin;

    const { data: user } = await adminAuth.inviteUserByEmail(email, {
      data: {
        name: name,
        role: role,
        super_admin: super_admin,
        supporter_id: supporter_id,
      },
    });

Heya, we're getting serious traffic from a specific IP toward the Auth endpoint. Looks like it's from a cloudflare assigned IP. Is this something that should be blocked from Supabase side, or should this be done from us. (I've confiscated the full IP)

"remote_addr":"2a06:************"

I'd like to observe postgres log in real-time by running

docker logs supabase_db_myproject -f

Instead, I see following entries in Postgres container logs :

2023-01-10 10:52:42.450 UTC [47] LOG:  redirecting log output to logging collector process
2023-01-10 10:52:42.450 UTC [47] HINT:  Future log output will appear in directory "/var/log/postgresql".

Executing

SHOW logging_collector

indeed returns "on". So I decided to disable it in the

postgresql.conf

, and to my surprise, found out that logging collector is not actually enabled there (

logging_collector = off

in

/etc/postgres/logging.conf

file). I'm confused. Any help is very much appreciated!

Hello all, I want to select people/profiles around certain location, however, I just want to show people who has verified their email, how can I achieve that? So far I have this:

create or replace function get_people(radius int, point text, get_all_people boolean)
returns setof profile
language plpgsql
as $$
begin
  if get_all_people then
    return query select * from profile where id != auth.uid();
  else
    return query select * from profile where id != auth.uid() and ST_DWithin(location, point::geometry, radius);
  end if;
end;
$$;

I have tried to access auth.confirmation_token and stuff like that but it seems that only auth.uid() is available. Thank you in advance and regards

I had created several function running on user creation thus on auth.users. All the functions were in public schema. I moved couple of functions in auth schema and the error disappeared. But I can't the see those function anymore. How can I see them and or delete them.

Based on some very annoying requirements, I have to do some mildy janky stuff with the auth emails. The answer to this question will greatly affect how my implementation goes.

Hi, I was able to create a function under the storage schema. However I can no longer view or edit it because we don't have the option to change schema in the view page, nor is it listed anywhere other than when you try to assign it to a trigger? Is this intentional? It appears to be a bug.

I see in many tutorials on Supabaase and other websites of people using the

text

datatype in their Supabase tables but I thought only the

varchar

datatype can be indexed and is much faster? Any insight on this?

I have a structured users in the DB , wish to just get first level set of users restricted by RLS, and then using virtual column by postgres function get the further structure In profiles table I have user_id and parent_id for each row. in js client :

.select("name, surname, children, code, id:user_id ")
      .eq("parent", ctx.user.id);

in the db :

create function children(profiles)
  returns jsonb
  language plpgsql
  security definer
  as
  $$
    begin
    return (select array_to_json(array_agg(profile)) from (
      select user_id as id, name, surname, code, from profiles where parent = $1.user_id) profile);
    end;
  $$

I wish just use the function recursively in sub-query but it doesn't work . Any ideas ?

I remember there used to be easy deployment for digital ocean, now I don’t see it anymore? Is there anything similar?

I am trying to use supabase with react native and I am following this tutorial - https://www.supabase.jp/docs/guides/with-expo Code provided in the tutorial doesn't work:

Property 'session' does not exist on type 'SupabaseAuthClient'.

export default function App() {
  const [session, setSession] = useState<Session | null>(null)

  useEffect(() => {
    setSession(supabase.auth.session())

    supabase.auth.onAuthStateChange((_event, session) => {
      setSession(session)
    })
  }, [])

  return (
    <View>
      {session && session.user ? <Account key={session.user.id} session={session} /> : <Auth />}
    </View>
  )
}