Merge pull request #6931 from ZeaLoVe/add-mini-ontime

ci: add mini ontime dataset to stateful test

Author: mergify[bot]

.github/actions/test_stateful_standalone_linux/action.yml

@@ -41,7 +41,15 @@ runs:
     - name: Run Stateful Tests with Standalone mode (ubuntu-latest only)
       shell: bash
       run: |
-        bash ./scripts/ci/ci-run-stateful-tests-standalone-s3.sh
+        ./scripts/ci/ci-run-stateful-tests-standalone-s3.sh
+
+    - name: Upload .stdout files
+      if: failure()
+      uses: actions/upload-artifact@v3
+      with:
+        name: test-stateful-standalone-linux-stdouts
+        path: |
+           ./tests/suites/1_stateful/*/*.stdout
 
     - name: Upload failure
       if: failure()

tests/suites/1_stateful/04_mini_dataset/04_0000_mini_ontime.result

@@ -0,0 +1,46 @@
+1	16773
+2	16651
+7	15551
+3	13671
+5	13159
+4	13153
+6	11041
+1	4279
+2	3637
+7	3571
+3	2788
+5	2191
+4	1880
+6	1813
+ORD	2036
+DTW	1817
+DFW	1719
+MSP	1506
+ATL	877
+LGA	816
+CVG	603
+BOS	556
+MEM	509
+RDU	371
+FL	221.4065829978688
+OH	208.2369273484498
+NW	207.8941134667772
+MQ	205.4136942462831
+HA	54.34782608695652
+FL	17308.666824532324
+HA	5713.526570048309
+MQ	12316.754450296687
+NW	11370.336373327018
+OH	12599.259602036094
+2006	12.16979169791698
+99999.0
+San Diego, CA	Los Angeles, CA	622
+Los Angeles, CA	San Diego, CA	619
+Kahului, HI	Honolulu, HI	605
+Honolulu, HI	Kahului, HI	592
+Chicago, IL	Minneapolis, MN	586
+Minneapolis, MN	Chicago, IL	565
+New York, NY	Boston, MA	523
+Boston, MA	New York, NY	522
+New York, NY	Raleigh/Durham, NC	488
+Lihue, HI	Honolulu, HI	486

tests/suites/1_stateful/04_mini_dataset/04_0000_mini_ontime.sh

@@ -0,0 +1,31 @@
+#!/usr/bin/env bash
+
+CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
+. "$CURDIR"/../../../shell_env.sh
+
+echo "drop table if exists ontime_mini;" | $MYSQL_CLIENT_CONNECT
+## Create table
+cat $CURDIR/../ddl/ontime.sql | sed 's/ontime/ontime_mini/g' | $MYSQL_CLIENT_CONNECT
+
+ontime_statements=(
+  ## Load data
+  "COPY INTO ontime_mini FROM 'https://repo.databend.rs/dataset/stateful/ontime_2006_100000.csv' FILE_FORMAT = ( type = 'CSV' field_delimiter = ',' record_delimiter = '\n' skip_header = 1 );"
+  ## run test
+  "SELECT DayOfWeek, count(*) AS c FROM ontime_mini WHERE (Year >= 2000) AND (Year <= 2008) GROUP BY DayOfWeek ORDER BY c DESC;"
+  "SELECT DayOfWeek, count(*) AS c FROM ontime_mini WHERE (DepDelay > 10) AND (Year >= 2000) AND (Year <= 2008) GROUP BY DayOfWeek ORDER BY c DESC;"
+  "SELECT Origin, count(*) AS c FROM ontime_mini WHERE (DepDelay > 10) AND (Year >= 2000) AND (Year <= 2008) GROUP BY Origin ORDER BY c DESC LIMIT 10;"
+  "SELECT IATA_CODE_Reporting_Airline AS Carrier, count() FROM ontime_mini WHERE (DepDelay > 10) AND (Year = 2007) GROUP BY Carrier ORDER BY count() DESC;"
+  "SELECT IATA_CODE_Reporting_Airline AS Carrier, avg(CAST(DepDelay > 10, Int8)) * 1000 AS c3 FROM ontime_mini WHERE Year = 2007 GROUP BY Carrier ORDER BY c3 DESC;"
+  "SELECT IATA_CODE_Reporting_Airline AS Carrier, avg(CAST(DepDelay > 10, Int8)) * 1000 AS c3 FROM ontime_mini WHERE (Year >= 2000) AND (Year <= 2008) GROUP BY Carrier ORDER BY c3 DESC;"
+  "SELECT IATA_CODE_Reporting_Airline AS Carrier, avg(DepDelay) * 1000 AS c3 FROM ontime_mini WHERE (Year >= 2000) AND (Year <= 2008) GROUP BY Carrier ORDER BY Carrier;"
+  "SELECT Year, avg(DepDelay) FROM ontime_mini GROUP BY Year;"
+  "SELECT avg(c1) FROM ( SELECT Year, Month, count(*) AS c1 FROM ontime_mini GROUP BY Year, Month ) AS a;"
+  "SELECT OriginCityName, DestCityName, count(*) AS c FROM ontime_mini GROUP BY OriginCityName, DestCityName ORDER BY c DESC LIMIT 10;"
+)
+
+for i in "${ontime_statements[@]}"; do
+  echo "set enable_planner_v2 = 1; $i" | $MYSQL_CLIENT_CONNECT
+done
+
+## Clean table
+echo "drop table if exists ontime_mini all;" | $MYSQL_CLIENT_CONNECT