feat: Support COUNT of relevant rows

Feature request

Is your feature request related to a problem? Please describe.

I’d like to perform COUNT(*) on relevant rows.

Describe the solution you’d like

A PostgrestTransformBuilder method. PostgREST supports this.

[steve-chavez]

Correct me if I am talking gibberish, but with this method, would you be able to query count value of joined table?

@dshukertjr No, this interface is just for counting the root table.

I guess I would have to send two request to achieve this huh?

Not necessarily, you can always create a VIEW or FUNCTION that has the aggregate to do it in one request.

There’s also another option, that is currently undocumented in postgrest since it’s not complete. Basically:

const { data, error } = await postgrest
    .from('users')
    .select(`
        name,
        tweets (
          count
        )
  `)

That will get the tweets count as you’d expect. However if you add more attributes(like tweets(content,date,count)) it will fail.

Calling other aggregate functions with group by is a future enhancement.

[dshukertjr]

Thanks @soedirgo and @steve-chavez. I will add this feature and create a pull request.