Skip to content

Commit

Permalink
Merge branch 'aspicer/stickiness' of github.com:PostHog/posthog into …
Browse files Browse the repository at this point in the history 
…aspicer/stickiness
  • Loading branch information
@aspicer
aspicer committed Nov 26, 2024
2 parents e994f37 + f1741ef commit bef27a2
Show file tree
Hide file tree
Showing 5 changed files with 319 additions and 0 deletions.
319 changes: 319 additions & 0 deletions ee/clickhouse/views/test/__snapshots__/test_clickhouse_stickiness.ambr
View file
Original file line number Diff line number Diff line change
Expand Up @@ -362,3 +362,322 @@
OFFSET 0
'''
# ---
# name: TestClickhouseStickiness.test_stickiness_all_time
'''
/* user_id:0 request:_snapshot_ */
SELECT timestamp
from events
WHERE team_id = 99999
AND timestamp > '2015-01-01'
order by timestamp
limit 1
'''
# ---
# name: TestClickhouseStickiness.test_stickiness_all_time.1
'''
/* user_id:0 request:_snapshot_ */
SELECT countDistinct(aggregation_target),
num_intervals
FROM
(SELECT if(notEmpty(pdi.distinct_id), pdi.person_id, e.person_id) AS aggregation_target,
countDistinct(toStartOfDay(toTimeZone(toDateTime(timestamp, 'UTC'), 'UTC'))) as num_intervals
FROM events e
LEFT OUTER JOIN
(SELECT distinct_id,
argMax(person_id, version) as person_id
FROM person_distinct_id2
WHERE team_id = 99999
AND distinct_id IN
(SELECT distinct_id
FROM events
WHERE team_id = 99999
AND event = 'watched movie'
AND toTimeZone(timestamp, 'UTC') >= toDateTime(toStartOfDay(toDateTime('2020-01-01 00:00:00', 'UTC')), 'UTC')
AND toTimeZone(timestamp, 'UTC') <= toDateTime('2020-01-08 23:59:59', 'UTC') )
GROUP BY distinct_id
HAVING argMax(is_deleted, version) = 0) AS pdi ON e.distinct_id = pdi.distinct_id
WHERE team_id = 99999
AND toTimeZone(timestamp, 'UTC') >= toDateTime(toStartOfDay(toDateTime('2020-01-01 00:00:00', 'UTC')), 'UTC')
AND toTimeZone(timestamp, 'UTC') <= toDateTime('2020-01-08 23:59:59', 'UTC')
AND event = 'watched movie'
GROUP BY aggregation_target)
WHERE num_intervals <= 9
GROUP BY num_intervals
ORDER BY num_intervals
'''
# ---
# name: TestClickhouseStickiness.test_stickiness_all_time_with_sampling
'''
/* user_id:0 request:_snapshot_ */
SELECT timestamp
from events
WHERE team_id = 99999
AND timestamp > '2015-01-01'
order by timestamp
limit 1
'''
# ---
# name: TestClickhouseStickiness.test_stickiness_all_time_with_sampling.1
'''
/* user_id:0 request:_snapshot_ */
SELECT countDistinct(aggregation_target),
num_intervals
FROM
(SELECT if(notEmpty(pdi.distinct_id), pdi.person_id, e.person_id) AS aggregation_target,
countDistinct(toStartOfDay(toTimeZone(toDateTime(timestamp, 'UTC'), 'UTC'))) as num_intervals
FROM events e SAMPLE 1.0
LEFT OUTER JOIN
(SELECT distinct_id,
argMax(person_id, version) as person_id
FROM person_distinct_id2
WHERE team_id = 99999
AND distinct_id IN
(SELECT distinct_id
FROM events
WHERE team_id = 99999
AND event = 'watched movie'
AND toTimeZone(timestamp, 'UTC') >= toDateTime(toStartOfDay(toDateTime('2020-01-01 00:00:00', 'UTC')), 'UTC')
AND toTimeZone(timestamp, 'UTC') <= toDateTime('2020-01-08 23:59:59', 'UTC') )
GROUP BY distinct_id
HAVING argMax(is_deleted, version) = 0) AS pdi ON e.distinct_id = pdi.distinct_id
WHERE team_id = 99999
AND toTimeZone(timestamp, 'UTC') >= toDateTime(toStartOfDay(toDateTime('2020-01-01 00:00:00', 'UTC')), 'UTC')
AND toTimeZone(timestamp, 'UTC') <= toDateTime('2020-01-08 23:59:59', 'UTC')
AND event = 'watched movie'
GROUP BY aggregation_target)
WHERE num_intervals <= 9
GROUP BY num_intervals
ORDER BY num_intervals
'''
# ---
# name: TestClickhouseStickiness.test_stickiness_hours
'''
/* user_id:0 request:_snapshot_ */
SELECT countDistinct(aggregation_target),
num_intervals
FROM
(SELECT if(notEmpty(pdi.distinct_id), pdi.person_id, e.person_id) AS aggregation_target,
countDistinct(toStartOfHour(toTimeZone(toDateTime(timestamp, 'UTC'), 'UTC'))) as num_intervals
FROM events e
LEFT OUTER JOIN
(SELECT distinct_id,
argMax(person_id, version) as person_id
FROM person_distinct_id2
WHERE team_id = 99999
AND distinct_id IN
(SELECT distinct_id
FROM events
WHERE team_id = 99999
AND event = 'watched movie'
AND toTimeZone(timestamp, 'UTC') >= toDateTime(toStartOfHour(toDateTime('2020-01-01 12:00:00', 'UTC')), 'UTC')
AND toTimeZone(timestamp, 'UTC') <= toDateTime('2020-01-01 20:00:00', 'UTC') )
GROUP BY distinct_id
HAVING argMax(is_deleted, version) = 0) AS pdi ON e.distinct_id = pdi.distinct_id
WHERE team_id = 99999
AND toTimeZone(timestamp, 'UTC') >= toDateTime(toStartOfHour(toDateTime('2020-01-01 12:00:00', 'UTC')), 'UTC')
AND toTimeZone(timestamp, 'UTC') <= toDateTime('2020-01-01 20:00:00', 'UTC')
AND event = 'watched movie'
GROUP BY aggregation_target)
WHERE num_intervals <= 10
GROUP BY num_intervals
ORDER BY num_intervals
'''
# ---
# name: TestClickhouseStickiness.test_stickiness_people_endpoint
'''
/* user_id:0 request:_snapshot_ */
SELECT DISTINCT aggregation_target AS actor_id
FROM
(SELECT if(notEmpty(pdi.distinct_id), pdi.person_id, e.person_id) AS aggregation_target,
countDistinct(toStartOfDay(toTimeZone(toDateTime(timestamp, 'UTC'), 'UTC'))) as num_intervals
FROM events e
LEFT OUTER JOIN
(SELECT distinct_id,
argMax(person_id, version) as person_id
FROM person_distinct_id2
WHERE team_id = 99999
AND distinct_id IN
(SELECT distinct_id
FROM events
WHERE team_id = 99999
AND ((event = 'watched movie'))
AND toTimeZone(timestamp, 'UTC') >= toDateTime(toStartOfDay(toDateTime('2020-01-01 00:00:00', 'UTC')), 'UTC')
AND toTimeZone(timestamp, 'UTC') <= toDateTime('2020-01-08 23:59:59', 'UTC') )
GROUP BY distinct_id
HAVING argMax(is_deleted, version) = 0) AS pdi ON e.distinct_id = pdi.distinct_id
WHERE team_id = 99999
AND toTimeZone(timestamp, 'UTC') >= toDateTime(toStartOfDay(toDateTime('2020-01-01 00:00:00', 'UTC')), 'UTC')
AND toTimeZone(timestamp, 'UTC') <= toDateTime('2020-01-08 23:59:59', 'UTC')
AND ((event = 'watched movie'))
GROUP BY aggregation_target)
WHERE num_intervals = 1
LIMIT 100
OFFSET 0
'''
# ---
# name: TestClickhouseStickiness.test_stickiness_people_paginated
'''
/* user_id:0 request:_snapshot_ */
SELECT DISTINCT aggregation_target AS actor_id
FROM
(SELECT if(notEmpty(pdi.distinct_id), pdi.person_id, e.person_id) AS aggregation_target,
countDistinct(toStartOfDay(toTimeZone(toDateTime(timestamp, 'UTC'), 'UTC'))) as num_intervals
FROM events e
LEFT OUTER JOIN
(SELECT distinct_id,
argMax(person_id, version) as person_id
FROM person_distinct_id2
WHERE team_id = 99999
AND distinct_id IN
(SELECT distinct_id
FROM events
WHERE team_id = 99999
AND ((event = 'watched movie'))
AND toTimeZone(timestamp, 'UTC') >= toDateTime(toStartOfDay(toDateTime('2020-01-01 00:00:00', 'UTC')), 'UTC')
AND toTimeZone(timestamp, 'UTC') <= toDateTime('2020-01-08 23:59:59', 'UTC') )
GROUP BY distinct_id
HAVING argMax(is_deleted, version) = 0) AS pdi ON e.distinct_id = pdi.distinct_id
WHERE team_id = 99999
AND toTimeZone(timestamp, 'UTC') >= toDateTime(toStartOfDay(toDateTime('2020-01-01 00:00:00', 'UTC')), 'UTC')
AND toTimeZone(timestamp, 'UTC') <= toDateTime('2020-01-08 23:59:59', 'UTC')
AND ((event = 'watched movie'))
GROUP BY aggregation_target)
WHERE num_intervals = 1
LIMIT 100
OFFSET 0
'''
# ---
# name: TestClickhouseStickiness.test_stickiness_people_paginated.1
'''
/* user_id:0 request:_snapshot_ */
SELECT DISTINCT aggregation_target AS actor_id
FROM
(SELECT if(notEmpty(pdi.distinct_id), pdi.person_id, e.person_id) AS aggregation_target,
countDistinct(toStartOfDay(toTimeZone(toDateTime(timestamp, 'UTC'), 'UTC'))) as num_intervals
FROM events e
LEFT OUTER JOIN
(SELECT distinct_id,
argMax(person_id, version) as person_id
FROM person_distinct_id2
WHERE team_id = 99999
AND distinct_id IN
(SELECT distinct_id
FROM events
WHERE team_id = 99999
AND ((event = 'watched movie'))
AND toTimeZone(timestamp, 'UTC') >= toDateTime(toStartOfDay(toDateTime('2020-01-01 00:00:00', 'UTC')), 'UTC')
AND toTimeZone(timestamp, 'UTC') <= toDateTime('2020-01-08 23:59:59', 'UTC') )
GROUP BY distinct_id
HAVING argMax(is_deleted, version) = 0) AS pdi ON e.distinct_id = pdi.distinct_id
WHERE team_id = 99999
AND toTimeZone(timestamp, 'UTC') >= toDateTime(toStartOfDay(toDateTime('2020-01-01 00:00:00', 'UTC')), 'UTC')
AND toTimeZone(timestamp, 'UTC') <= toDateTime('2020-01-08 23:59:59', 'UTC')
AND ((event = 'watched movie'))
GROUP BY aggregation_target)
WHERE num_intervals = 1
LIMIT 100
OFFSET 100
'''
# ---
# name: TestClickhouseStickiness.test_stickiness_with_person_on_events_v2
'''

SELECT DISTINCT person_id
FROM events
WHERE team_id = 99999
AND distinct_id = 'person2'
'''
# ---
# name: TestClickhouseStickiness.test_stickiness_with_person_on_events_v2.1
'''
/* user_id:0 request:_snapshot_ */
SELECT countDistinct(aggregation_target),
num_intervals
FROM
(SELECT if(notEmpty(overrides.distinct_id), overrides.person_id, e.person_id) AS aggregation_target,
countDistinct(toStartOfDay(toTimeZone(toDateTime(timestamp, 'UTC'), 'UTC'))) as num_intervals
FROM events e
LEFT OUTER JOIN
(SELECT argMax(person_distinct_id_overrides.person_id, person_distinct_id_overrides.version) AS person_id,
person_distinct_id_overrides.distinct_id AS distinct_id
FROM person_distinct_id_overrides
WHERE equals(person_distinct_id_overrides.team_id, 99999)
GROUP BY person_distinct_id_overrides.distinct_id
HAVING ifNull(equals(argMax(person_distinct_id_overrides.is_deleted, person_distinct_id_overrides.version), 0), 0)) AS overrides ON e.distinct_id = overrides.distinct_id
WHERE team_id = 99999
AND toTimeZone(timestamp, 'UTC') >= toDateTime(toStartOfDay(toDateTime('2020-01-01 00:00:00', 'UTC')), 'UTC')
AND toTimeZone(timestamp, 'UTC') <= toDateTime('2020-01-08 23:59:59', 'UTC')
AND event = 'watched movie'
AND notEmpty(e.person_id)
GROUP BY aggregation_target)
WHERE num_intervals <= 9
GROUP BY num_intervals
ORDER BY num_intervals
'''
# ---
# name: TestClickhouseStickiness.test_timezones
'''

SELECT countDistinct(aggregation_target),
num_intervals
FROM
(SELECT if(notEmpty(pdi.distinct_id), pdi.person_id, e.person_id) AS aggregation_target,
countDistinct(toStartOfDay(toTimeZone(toDateTime(timestamp, 'UTC'), 'UTC'))) as num_intervals
FROM events e
LEFT OUTER JOIN
(SELECT distinct_id,
argMax(person_id, version) as person_id
FROM person_distinct_id2
WHERE team_id = 99999
AND distinct_id IN
(SELECT distinct_id
FROM events
WHERE team_id = 99999
AND event = '$pageview'
AND toTimeZone(timestamp, 'UTC') >= toDateTime(toStartOfDay(toDateTime('2021-05-01 00:00:00', 'UTC')), 'UTC')
AND toTimeZone(timestamp, 'UTC') <= toDateTime('2021-05-15 23:59:59', 'UTC') )
GROUP BY distinct_id
HAVING argMax(is_deleted, version) = 0) AS pdi ON e.distinct_id = pdi.distinct_id
WHERE team_id = 99999
AND toTimeZone(timestamp, 'UTC') >= toDateTime(toStartOfDay(toDateTime('2021-05-01 00:00:00', 'UTC')), 'UTC')
AND toTimeZone(timestamp, 'UTC') <= toDateTime('2021-05-15 23:59:59', 'UTC')
AND event = '$pageview'
GROUP BY aggregation_target)
WHERE num_intervals <= 16
GROUP BY num_intervals
ORDER BY num_intervals
'''
# ---
# name: TestClickhouseStickiness.test_timezones.1
'''

SELECT countDistinct(aggregation_target),
num_intervals
FROM
(SELECT if(notEmpty(pdi.distinct_id), pdi.person_id, e.person_id) AS aggregation_target,
countDistinct(toStartOfDay(toTimeZone(toDateTime(timestamp, 'UTC'), 'US/Pacific'))) as num_intervals
FROM events e
LEFT OUTER JOIN
(SELECT distinct_id,
argMax(person_id, version) as person_id
FROM person_distinct_id2
WHERE team_id = 99999
AND distinct_id IN
(SELECT distinct_id
FROM events
WHERE team_id = 99999
AND event = '$pageview'
AND toTimeZone(timestamp, 'US/Pacific') >= toDateTime(toStartOfDay(toDateTime('2021-05-01 00:00:00', 'US/Pacific')), 'US/Pacific')
AND toTimeZone(timestamp, 'US/Pacific') <= toDateTime('2021-05-15 23:59:59', 'US/Pacific') )
GROUP BY distinct_id
HAVING argMax(is_deleted, version) = 0) AS pdi ON e.distinct_id = pdi.distinct_id
WHERE team_id = 99999
AND toTimeZone(timestamp, 'US/Pacific') >= toDateTime(toStartOfDay(toDateTime('2021-05-01 00:00:00', 'US/Pacific')), 'US/Pacific')
AND toTimeZone(timestamp, 'US/Pacific') <= toDateTime('2021-05-15 23:59:59', 'US/Pacific')
AND event = '$pageview'
GROUP BY aggregation_target)
WHERE num_intervals <= 16
GROUP BY num_intervals
ORDER BY num_intervals
'''
# ---
Binary file modified frontend/__snapshots__/scenes-app-insights--stickiness-edit--dark--webkit.png
View file
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file modified frontend/__snapshots__/scenes-app-insights--stickiness-edit--dark.png
View file
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file modified frontend/__snapshots__/scenes-app-insights--stickiness-edit--light--webkit.png
View file
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file modified frontend/__snapshots__/scenes-app-insights--stickiness-edit--light.png
View file
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.

0 comments on commit bef27a2

Please sign in to comment.