fix(insights): Fix interval for total value queries (#23037)

2024-11-21 13:39:22 +01:00 · 2024-06-19 10:11:38 +01:00 · 2024-06-19 10:11:38 +01:00 · 96f60f80db
commit 96f60f80db
parent 837fe038ed
4 changed files with 84 additions and 64 deletions
--- a/livestream/docker-compose.yml
+++ b/livestream/docker-compose.yml
@ -1,62 +1,62 @@
-version: "3.8"
+version: '3.8'

 services:
-  postgres:
-    image: postgres:16-alpine
-    restart: always
-    ports:
-      - "5432:5432"
-    environment:
-      POSTGRES_USER: postgres
-      POSTGRES_PASSWORD: postgres
-      POSTGRES_DB: liveevents
-    healthcheck:
-      test: ["CMD-SHELL", "pg_isready -U postgres"]
-      interval: 5s
-      timeout: 5s
+    postgres:
+        image: postgres:16-alpine
+        restart: always
+        ports:
+            - '5432:5432'
+        environment:
+            POSTGRES_USER: postgres
+            POSTGRES_PASSWORD: postgres
+            POSTGRES_DB: liveevents
+        healthcheck:
+            test: ['CMD-SHELL', 'pg_isready -U postgres']
+            interval: 5s
+            timeout: 5s

-  redis:
-    image: redis:alpine
-    restart: always
-    ports:
-      - "6379:6379"
+    redis:
+        image: redis:alpine
+        restart: always
+        ports:
+            - '6379:6379'

-  redpanda:
-    image: vectorized/redpanda:v23.2.17
-    command:
-      - redpanda start
-      - --smp 1
-      - --overprovisioned
-      - --node-id 0
-      - --kafka-addr PLAINTEXT://0.0.0.0:29092,OUTSIDE://0.0.0.0:9092
-      - --advertise-kafka-addr PLAINTEXT://redpanda:29092,OUTSIDE://localhost:9092
-      - --pandaproxy-addr 0.0.0.0:8082
-      - --advertise-pandaproxy-addr localhost:8082
-    ports:
-      - 8081:8081
-      - 8082:8082
-      - 9092:9092
-      - 29092:29092
+    redpanda:
+        image: vectorized/redpanda:v23.2.17
+        command:
+            - redpanda start
+            - --smp 1
+            - --overprovisioned
+            - --node-id 0
+            - --kafka-addr PLAINTEXT://0.0.0.0:29092,OUTSIDE://0.0.0.0:9092
+            - --advertise-kafka-addr PLAINTEXT://redpanda:29092,OUTSIDE://localhost:9092
+            - --pandaproxy-addr 0.0.0.0:8082
+            - --advertise-pandaproxy-addr localhost:8082
+        ports:
+            - 8081:8081
+            - 8082:8082
+            - 9092:9092
+            - 29092:29092

-  console:
-    image: docker.redpanda.com/redpandadata/console:v2.3.8
-    restart: on-failure
-    entrypoint: /bin/sh
-    command: -c "echo \"$$CONSOLE_CONFIG_FILE\" > /tmp/config.yml; /app/console"
-    environment:
-      CONFIG_FILEPATH: /tmp/config.yml
-      CONSOLE_CONFIG_FILE: |
-        kafka:
-          brokers: ["redpanda:29092"]
-          schemaRegistry:
-            enabled: true
-            urls: ["http://redpanda:8081"]
-        connect:
-          enabled: true
-          clusters:
-            - name: datagen
-              url: http://connect:8083
-    ports:
-      - "8088:8088"
-    depends_on:
-      - redpanda
+    console:
+        image: docker.redpanda.com/redpandadata/console:v2.3.8
+        restart: on-failure
+        entrypoint: /bin/sh
+        command: -c "echo \"$$CONSOLE_CONFIG_FILE\" > /tmp/config.yml; /app/console"
+        environment:
+            CONFIG_FILEPATH: /tmp/config.yml
+            CONSOLE_CONFIG_FILE: |
+                kafka:
+                  brokers: ["redpanda:29092"]
+                  schemaRegistry:
+                    enabled: true
+                    urls: ["http://redpanda:8081"]
+                connect:
+                  enabled: true
+                  clusters:
+                    - name: datagen
+                      url: http://connect:8083
+        ports:
+            - '8088:8088'
+        depends_on:
+            - redpanda
--- a/posthog/hogql_queries/insights/trends/test/snapshots/test_trends.ambr
+++ b/posthog/hogql_queries/insights/trends/test/snapshots/test_trends.ambr
@ -3390,10 +3390,10 @@
             (SELECT dateDiff('second', min(sessions.min_timestamp), max(sessions.max_timestamp)) AS `$session_duration`,
                     sessions.session_id AS session_id
              FROM sessions
-              WHERE and(equals(sessions.team_id, 2), ifNull(greaterOrEquals(plus(toTimeZone(sessions.min_timestamp, 'UTC'), toIntervalDay(3)), toStartOfWeek(assumeNotNull(parseDateTime64BestEffortOrNull('2019-12-28 00:00:00', 6, 'UTC')), 0)), 0), ifNull(lessOrEquals(minus(toTimeZone(sessions.min_timestamp, 'UTC'), toIntervalDay(3)), assumeNotNull(parseDateTime64BestEffortOrNull('2020-01-04 23:59:59', 6, 'UTC'))), 0))
+              WHERE and(equals(sessions.team_id, 2), ifNull(greaterOrEquals(plus(toTimeZone(sessions.min_timestamp, 'UTC'), toIntervalDay(3)), toStartOfDay(assumeNotNull(parseDateTime64BestEffortOrNull('2019-12-28 00:00:00', 6, 'UTC')))), 0), ifNull(lessOrEquals(minus(toTimeZone(sessions.min_timestamp, 'UTC'), toIntervalDay(3)), assumeNotNull(parseDateTime64BestEffortOrNull('2020-01-04 23:59:59', 6, 'UTC'))), 0))
              GROUP BY sessions.session_id,
                       sessions.session_id) AS e__session ON equals(e.`$session_id`, e__session.session_id)
-           WHERE and(equals(e.team_id, 2), greaterOrEquals(toTimeZone(e.timestamp, 'UTC'), toStartOfWeek(assumeNotNull(parseDateTime64BestEffortOrNull('2019-12-28 00:00:00', 6, 'UTC')), 0)), lessOrEquals(toTimeZone(e.timestamp, 'UTC'), assumeNotNull(parseDateTime64BestEffortOrNull('2020-01-04 23:59:59', 6, 'UTC'))), equals(e.event, 'sign up'), true)
+           WHERE and(equals(e.team_id, 2), greaterOrEquals(toTimeZone(e.timestamp, 'UTC'), toStartOfDay(assumeNotNull(parseDateTime64BestEffortOrNull('2019-12-28 00:00:00', 6, 'UTC')))), lessOrEquals(toTimeZone(e.timestamp, 'UTC'), assumeNotNull(parseDateTime64BestEffortOrNull('2020-01-04 23:59:59', 6, 'UTC'))), equals(e.event, 'sign up'), true)
           GROUP BY e.`$session_id`,
                    breakdown_value
           ORDER BY 1 DESC, breakdown_value DESC)
@ -3867,7 +3867,7 @@
             (SELECT dateDiff('second', min(sessions.min_timestamp), max(sessions.max_timestamp)) AS `$session_duration`,
                     sessions.session_id AS session_id
              FROM sessions
-              WHERE and(equals(sessions.team_id, 2), ifNull(greaterOrEquals(plus(toTimeZone(sessions.min_timestamp, 'UTC'), toIntervalDay(3)), toStartOfWeek(assumeNotNull(parseDateTime64BestEffortOrNull('2019-12-28 00:00:00', 6, 'UTC')), 0)), 0), ifNull(lessOrEquals(minus(toTimeZone(sessions.min_timestamp, 'UTC'), toIntervalDay(3)), assumeNotNull(parseDateTime64BestEffortOrNull('2020-01-04 23:59:59', 6, 'UTC'))), 0))
+              WHERE and(equals(sessions.team_id, 2), ifNull(greaterOrEquals(plus(toTimeZone(sessions.min_timestamp, 'UTC'), toIntervalDay(3)), toStartOfDay(assumeNotNull(parseDateTime64BestEffortOrNull('2019-12-28 00:00:00', 6, 'UTC')))), 0), ifNull(lessOrEquals(minus(toTimeZone(sessions.min_timestamp, 'UTC'), toIntervalDay(3)), assumeNotNull(parseDateTime64BestEffortOrNull('2020-01-04 23:59:59', 6, 'UTC'))), 0))
              GROUP BY sessions.session_id,
                       sessions.session_id) AS e__session ON equals(e.`$session_id`, e__session.session_id)
           INNER JOIN
@ -3888,7 +3888,7 @@
                                                                WHERE equals(person.team_id, 2)
                                                                GROUP BY person.id
                                                                HAVING and(ifNull(equals(argMax(person.is_deleted, person.version), 0), 0), ifNull(less(argMax(person.created_at, person.version), plus(now64(6, 'UTC'), toIntervalDay(1))), 0)))), 0)) SETTINGS optimize_aggregation_in_order=1) AS e__pdi__person ON equals(e__pdi.e__pdi___person_id, e__pdi__person.id)
-           WHERE and(equals(e.team_id, 2), greaterOrEquals(toTimeZone(e.timestamp, 'UTC'), toStartOfWeek(assumeNotNull(parseDateTime64BestEffortOrNull('2019-12-28 00:00:00', 6, 'UTC')), 0)), lessOrEquals(toTimeZone(e.timestamp, 'UTC'), assumeNotNull(parseDateTime64BestEffortOrNull('2020-01-04 23:59:59', 6, 'UTC'))), equals(e.event, 'sign up'), true)
+           WHERE and(equals(e.team_id, 2), greaterOrEquals(toTimeZone(e.timestamp, 'UTC'), toStartOfDay(assumeNotNull(parseDateTime64BestEffortOrNull('2019-12-28 00:00:00', 6, 'UTC')))), lessOrEquals(toTimeZone(e.timestamp, 'UTC'), assumeNotNull(parseDateTime64BestEffortOrNull('2020-01-04 23:59:59', 6, 'UTC'))), equals(e.event, 'sign up'), true)
           GROUP BY e.`$session_id`,
                    breakdown_value
           ORDER BY 1 DESC, breakdown_value DESC)
@ -3944,10 +3944,10 @@
       (SELECT dateDiff('second', min(sessions.min_timestamp), max(sessions.max_timestamp)) AS `$session_duration`,
               sessions.session_id AS session_id
        FROM sessions
-        WHERE and(equals(sessions.team_id, 2), ifNull(greaterOrEquals(plus(toTimeZone(sessions.min_timestamp, 'UTC'), toIntervalDay(3)), toStartOfWeek(assumeNotNull(parseDateTime64BestEffortOrNull('2019-12-28 00:00:00', 6, 'UTC')), 0)), 0), ifNull(lessOrEquals(minus(toTimeZone(sessions.min_timestamp, 'UTC'), toIntervalDay(3)), assumeNotNull(parseDateTime64BestEffortOrNull('2020-01-04 23:59:59', 6, 'UTC'))), 0))
+        WHERE and(equals(sessions.team_id, 2), ifNull(greaterOrEquals(plus(toTimeZone(sessions.min_timestamp, 'UTC'), toIntervalDay(3)), toStartOfDay(assumeNotNull(parseDateTime64BestEffortOrNull('2019-12-28 00:00:00', 6, 'UTC')))), 0), ifNull(lessOrEquals(minus(toTimeZone(sessions.min_timestamp, 'UTC'), toIntervalDay(3)), assumeNotNull(parseDateTime64BestEffortOrNull('2020-01-04 23:59:59', 6, 'UTC'))), 0))
        GROUP BY sessions.session_id,
                 sessions.session_id) AS e__session ON equals(e.`$session_id`, e__session.session_id)
-     WHERE and(equals(e.team_id, 2), greaterOrEquals(toTimeZone(e.timestamp, 'UTC'), toStartOfWeek(assumeNotNull(parseDateTime64BestEffortOrNull('2019-12-28 00:00:00', 6, 'UTC')), 0)), lessOrEquals(toTimeZone(e.timestamp, 'UTC'), assumeNotNull(parseDateTime64BestEffortOrNull('2020-01-04 23:59:59', 6, 'UTC'))), equals(e.event, 'sign up'))
+     WHERE and(equals(e.team_id, 2), greaterOrEquals(toTimeZone(e.timestamp, 'UTC'), toStartOfDay(assumeNotNull(parseDateTime64BestEffortOrNull('2019-12-28 00:00:00', 6, 'UTC')))), lessOrEquals(toTimeZone(e.timestamp, 'UTC'), assumeNotNull(parseDateTime64BestEffortOrNull('2020-01-04 23:59:59', 6, 'UTC'))), equals(e.event, 'sign up'))
     GROUP BY e.`$session_id`
     ORDER BY 1 DESC)
  LIMIT 50000 SETTINGS readonly=2,
--- a/posthog/hogql_queries/insights/trends/test/test_trends_query_runner.py
+++ b/posthog/hogql_queries/insights/trends/test/test_trends_query_runner.py
@ -1429,6 +1429,24 @@ class TestTrendsQueryRunner(ClickhouseTestMixin, APIBaseTest):
        assert response.results[0]["count"] == 0
        assert response.results[0]["aggregated_value"] == 10

+    def test_trends_display_aggregate_interval(self):
+        self._create_test_events()
+
+        response = self._run_trends_query(
+            "2020-01-09",
+            "2020-01-20",
+            IntervalType.MONTH,  # E.g. UI sets interval to month, but we need the total value across all days
+            [EventsNode(event="$pageview")],
+            TrendsFilter(display=ChartDisplayType.BOLD_NUMBER),
+            None,
+        )
+
+        assert len(response.results) == 1
+        assert response.results[0]["data"] == []
+        assert response.results[0]["days"] == []
+        assert response.results[0]["count"] == 0
+        assert response.results[0]["aggregated_value"] == 10
+
    def test_trends_display_cumulative(self):
        self._create_test_events()

--- a/posthog/hogql_queries/insights/trends/trends_query_runner.py
+++ b/posthog/hogql_queries/insights/trends/trends_query_runner.py
@ -66,6 +66,7 @@ from posthog.schema import (
    HogQLQueryModifiers,
    DataWarehouseEventsModifier,
    BreakdownType,
+    IntervalType,
 )
 from posthog.warehouse.models import DataWarehouseTable
 from posthog.utils import format_label_date, multisort
@ -574,10 +575,11 @@ class TrendsQueryRunner(QueryRunner):

    @cached_property
    def query_date_range(self):
+        interval = IntervalType.DAY if self._trends_display.is_total_value() else self.query.interval
        return QueryDateRange(
            date_range=self.query.dateRange,
            team=self.team,
-            interval=self.query.interval,
+            interval=interval,
            now=datetime.now(),
        )