Postgres concatenation of a column for a given group - sql

Question

I currently have a Postgres DB with a SHOWS table that looks like this:
id date venue price is_sold ages time multi_day pit
13 2016-01-02 924 Gilman Street, Berkeley $8 FALSE a/a 7:30pm FALSE FALSE
I have a query that orders everything in this table by date (which is of timestamptz type):
SELECT shows.* FROM shows ORDER BY date
Right now, running this query with the pg-promise library produces an array of objects that look like this:
[
{ id: 1,
date: Thu Oct 20 2016 17:00:00 GMT-0700 (PDT),
venue: 'Regency Ballroom, S.F.',
price: null,
is_sold: false,
ages: 'a/a',
time: null,
multi_day: false,
pit: false },
{ id: 2,
date: Thu Oct 20 2016 17:00:00 GMT-0700 (PDT),
venue: 'Great American Music Hall.',
price: null,
is_sold: false,
ages: 'a/a',
time: null,
multi_day: false,
pit: false } ... ]
I want to return an array that groups this data by date column. Example output:
[
{date: Thu Oct 20 2016 17:00:00 GMT-0700 (PDT),
shows: [1, 2]}, // or however this should be structured; I just want a way to get multiple show ids from a single date.
...
]
which I assume I could then INNER JOIN to get the rest of the show details that I need.
How do I adjust my query to return this sort of grouping?

Answer 1

Just to answer the question, use
SELECT date, string_agg(id, ', ') AS show_ids
FROM shows
GROUP BY date