ci(tests): add mini dataset to stateful test

.github/actions/test_stateful_standalone_linux/action.yml

@@ -9,6 +9,13 @@ inputs:
     description: ""
     required: true
     default: "x86_64-unknown-linux-gnu"
+  aws_access_key_id:
+    description: "aws access key id"
+    required: true
+  aws_secret_access_key:
+    description: "aws secret access key"
+    required: true
+
 runs:
   using: "composite"
   steps:
@@ -40,8 +47,10 @@ runs:
 
     - name: Run Stateful Tests with Standalone mode (ubuntu-latest only)
       shell: bash
-      run: |
-        bash ./scripts/ci/ci-run-stateful-tests-standalone-s3.sh
+      run: bash ./scripts/ci/ci-run-stateful-tests-standalone-s3.sh
+      env:
+        REPO_AWS_ACCESS_KEY_ID: ${{ inputs.aws_access_key_id }}
+        REPO_AWS_SECRET_ACCESS_KEY: ${{ inputs.aws_secret_access_key }}
 
     - name: Upload failure
       if: failure()

.github/workflows/dev-linux.yml

@@ -124,6 +124,9 @@ jobs:
     steps:
       - uses: actions/checkout@v3
       - uses: ./.github/actions/test_stateful_standalone_linux
+        with:
+          aws_access_key_id: ${{ secrets.REPO_ACCESS_KEY_ID }}
+          aws_secret_access_key: ${{ secrets.REPO_SECRET_ACCESS_KEY }}
 
   test_fuzz_standalone_linux:
     runs-on: [self-hosted, X64, Linux, development]

tests/suites/1_stateful/04_mini_dataset/00_0000_mini_ontime.result

@@ -0,0 +1,56 @@
+count()
+99999
+DayOfWeek       c
+1       16773
+2       16651
+7       15551
+3       13671
+5       13159
+4       13153
+6       11041
+DayOfWeek       c
+1       4279
+2       3637
+7       3571
+3       2788
+5       2191
+4       1880
+6       1813
+Origin  c
+ORD     2036
+DTW     1817
+DFW     1719
+MSP     1506
+ATL     877
+LGA     816
+CVG     603
+BOS     556
+MEM     509
+RDU     371
+Carrier c3
+FL      221.4065829978688
+OH      208.2369273484498
+NW      207.8941134667772
+MQ      205.4136942462831
+HA      54.34782608695652
+Carrier c3
+HA      5713.526570048309
+MQ      12316.754450296687
+FL      17308.666824532324
+OH      12599.259602036094
+NW      11370.336373327018
+Year    avg(DepDelay)
+2006    12.16979169791698
+avg(c1)
+99999.0
+OriginCityName  DestCityName    c
+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/00_0000_mini_ontime.sh

@@ -0,0 +1,26 @@
+#!/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' | $MYSQL_CLIENT_CONNECT
+
+## Load data
+echo "COPY INTO ontime_mini FROM 's3://repo.databend.rs/dataset/stateful/ontime_2006_100000.csv' credentials=(aws_key_id='$REPO_AWS_ACCESS_KEY_ID' aws_secret_key='$REPO_AWS_SECRET_ACCESS_KEY') FILE_FORMAT = ( type = 'CSV' field_delimiter = ',' record_delimiter = '\n' skip_header = 1 );" | $MYSQL_CLIENT_CONNECT
+
+## Run test
+echo 'SELECT DayOfWeek, count(*) AS c FROM ontime_mini WHERE (Year >= 2000) AND (Year <= 2008) GROUP BY DayOfWeek ORDER BY c DESC;' |$MYSQL_CLIENT_CONNECT
+echo 'SELECT DayOfWeek, count(*) AS c FROM ontime_mini WHERE (DepDelay > 10) AND (Year >= 2000) AND (Year <= 2008) GROUP BY DayOfWeek ORDER BY c DESC;' |$MYSQL_CLIENT_CONNECT
+echo 'SELECT Origin, count(*) AS c FROM ontime_mini WHERE (DepDelay > 10) AND (Year >= 2000) AND (Year <= 2008) GROUP BY Origin ORDER BY c DESC LIMIT 10;' |$MYSQL_CLIENT_CONNECT
+echo 'SELECT IATA_CODE_Reporting_Airline AS Carrier, count() FROM ontime_mini WHERE (DepDelay > 10) AND (Year = 2007) GROUP BY Carrier ORDER BY count() DESC;' |$MYSQL_CLIENT_CONNECT
+echo 'SELECT IATA_CODE_Reporting_Airline AS Carrier, avg(CAST(DepDelay > 10, Int8)) * 1000 AS c3 FROM ontime_mini WHERE Year = 2007 GROUP BY Carrier ORDER BY c3 DESC;' |$MYSQL_CLIENT_CONNECT
+echo '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;' |$MYSQL_CLIENT_CONNECT
+echo 'SELECT IATA_CODE_Reporting_Airline AS Carrier, avg(DepDelay) * 1000 AS c3 FROM ontime_mini WHERE (Year >= 2000) AND (Year <= 2008) GROUP BY Carrier;' |$MYSQL_CLIENT_CONNECT
+echo 'SELECT Year, avg(DepDelay) FROM ontime_mini GROUP BY Year;' |$MYSQL_CLIENT_CONNECT
+echo 'SELECT avg(c1) FROM ( SELECT Year, Month, count(*) AS c1 FROM ontime_mini GROUP BY Year, Month ) AS a;' |$MYSQL_CLIENT_CONNECT
+echo 'SELECT OriginCityName, DestCityName, count(*) AS c FROM ontime_mini GROUP BY OriginCityName, DestCityName ORDER BY c DESC LIMIT 10;' |$MYSQL_CLIENT_CONNECT
+
+## Clean table
+echo "drop table ontime_mini all;" | $MYSQL_CLIENT_CONNECT