There doesn't seem to be a way to check the difference between null and 'null' in the is() filter

[AdamGEmerson]

Describe the bug

It seems that when using .is(), there is no way to query for string literals that say "null".

In the supabase js library, this is the given example for is():

DB

create table
  countries (id int8 primary key, name text);

insert into
  countries (id, name)
values
  (1, 'null'),
  (2, null);

Query

const { data, error } = await supabase
  .from('countries')
  .select()
  .is('name', null)

Response

{
  "data": [
    {
      "id": 1,
      "name": "null"
    }
  ],
  "status": 200,
  "statusText": "OK"
}

This behavior cannot be replicated in the python library because rather than substituting None for the js keyword null, the python library uses the string 'null' to check for nullness in the database.

Replicating this in the python library gives a different result, row 2.

data, count = supabase.table('countries') \
  .select('*') \
  .is_('name', 'null') \
  .execute()

('data', [{'id': 2, 'name': None}])

[AdamGEmerson]

On second look, this might just be an issue with the JS documentation. I suppose the correct way to query for a string "null" would be using .eq() not .is_(). I'll try messing around with JS library to confirm.

[silentworks]

This functionality has been added to the library supabase/postgrest-py#446