Skip to content

A platform for community discussion. Free, open, simple.

PERF: Detect visit island boundaries directly in the DAU/MAU query (#41612)

Krzysztof Kotlarekd34814baf7ab
diff --git a/app/models/user_visit.rb b/app/models/user_visit.rb
index ce325fd0e58..c93701edd4d 100644
--- a/app/models/user_visit.rb
+++ b/app/models/user_visit.rb
@@ -16,33 +16,34 @@ class UserVisit < ActiveRecord::Base
def self.count_by_active_users(start_date, end_date)
sql = <<~SQL
- WITH visits_in_mau_range AS (
- SELECT user_id, visited_at AS d
- FROM user_visits
- WHERE visited_at >= :start_date::DATE - 29
- AND visited_at <= :end_date::DATE
- ),
- visit_island_starts AS (
- SELECT user_id, d,
- CASE
- WHEN lag(d) OVER (PARTITION BY user_id ORDER BY d) >= d - 30 THEN 0
- ELSE 1
- END AS new_island
- FROM visits_in_mau_range
- ),
- visit_islands AS (
- SELECT user_id, d, sum(new_island) OVER (PARTITION BY user_id ORDER BY d) AS grp
- FROM visit_island_starts
- ),
- active_ranges AS (
- SELECT min(d) AS start_d, max(d) + 29 AS end_d
- FROM visit_islands
- GROUP BY user_id, grp
+ WITH island_boundaries AS (
+ -- A user's visit days merge into an "active island" when gaps are
+ -- 30 days or less; the user counts toward MAU from each island's
+ -- start until 29 days after its end. A row is an island start when
+ -- the previous visit is too far back, an island end when the next
+ -- visit is too far ahead.
+ SELECT d, prev_d, next_d
+ FROM (
+ SELECT
+ visited_at AS d,
+ lag(visited_at) OVER w AS prev_d,
+ lead(visited_at) OVER w AS next_d
+ FROM user_visits
+ WHERE visited_at >= :start_date::DATE - 29
+ AND visited_at <= :end_date::DATE
+ WINDOW w AS (PARTITION BY user_id ORDER BY visited_at)
+ ) visits
+ WHERE prev_d IS NULL OR prev_d < d - 30
+ OR next_d IS NULL OR next_d > d + 30
),
active_range_events AS (
- SELECT start_d AS day, 1 AS delta FROM active_ranges
+ SELECT d AS day, 1 AS delta
+ FROM island_boundaries
+ WHERE prev_d IS NULL OR prev_d < d - 30
UNION ALL
- SELECT end_d + 1 AS day, -1 AS delta FROM active_ranges
+ SELECT d + 30 AS day, -1 AS delta
+ FROM island_boundaries
+ WHERE next_d IS NULL OR next_d > d + 30
),
daily_active_deltas AS (
SELECT day, sum(delta) AS delta