[Enhancement] enhancement for regression-test framework

be/src/vec/columns/column_vector.h

@@ -21,6 +21,7 @@
 #pragma once
 
 #include <cmath>
+#include <type_traits>
 
 #include "olap/uint24.h"
 #include "vec/columns/column.h"
@@ -188,7 +189,8 @@ class ColumnVector final : public COWHelper<ColumnVectorHelper, ColumnVector<T>>
         for (int i = 0; i < num; i++) {
             const char* cur_ptr = data_ptr + value_size * i;
             uint64_t value = *reinterpret_cast<const uint64_t*>(cur_ptr);
-            vectorized::VecDateTimeValue date(value);
+            vectorized::VecDateTimeValue date;
+            date.from_olap_datetime(value);
             this->insert_data(reinterpret_cast<char*>(&date), 0);
         }
     }

regression-test/data/aggregate/aggregate.out

@@ -1,75 +1,76 @@
 -- This file is automatically generated. You should know what you did if you want to edit this
--- !aggregate --
+-- !aggregate_min_max --
 TESTING	AGAIN
 
--- !aggregate --
+-- !aggregate_avg --
 3309.4	50.12646
 
--- !aggregate --
+-- !aggregate_avg2 --
 3566.3333333333335	50.65743333333333
 
--- !aggregate --
+-- !aggregate_multi_count --
 15	15	14	14	14	15
 
--- !aggregate --
+-- !aggregate_multi_distinct_count --
 9	9	10	8	7	2
 
--- !aggregate --
+-- !aggregate_multi_max --
 8996	99.8777	testing	1999-01-08	2010-01-02T05:09:06
 
--- !aggregate --
+-- !aggregate_multi_min --
 12	0.245	AGAIN	1922-04-02	1989-01-08T04:05:06
 
--- !aggregate --
+-- !aggregate_mix --
 14	99.8777	3309.4
 
--- !aggregate --
+-- !aggregate_stddev1 --
 3037.0783723835643	35.210273031541995
 
--- !aggregate --
-3145.0576394647455	35.21027303154199
+-- !aggregate_stddev2 --
+3145.057639464745	35.21027303154199
 
--- !aggregate --
+-- !aggregate_stddev3 --
 3037.0783723835643	40.09062957713574
 
--- !aggregate --
+-- !aggregate_stddev4 --
 3143.674778708147	36.44609513116206
 
--- !aggregate --
+-- !aggregate_stddev5 --
 3335.837376132116	36.446095131162046
 
--- !aggregate --
+-- !aggregate_stddev6 --
 3143.6747787081463	42.52253405404598
 
--- !aggregate --
+-- !aggregate_sum --
 49641	751.8969000000001
 
--- !aggregate --
+-- !aggregate_distinct_sum --
 32097	455.9169
 
--- !aggregate --
+-- !aggregate_variance --
 9223845.040000001	1239.7633269557336
 
--- !aggregate --
-9891387.555555556	1239.7633269557332
+-- !aggregate_variance2 --
+9891387.555555554	1239.7633269557332
 
--- !aggregate --
+-- !aggregate_variance3 --
 9223845.04	1607.2585798911111
 
--- !aggregate --
+-- !aggregate_variance4 --
 9882691.114285717	1328.3178503097147
 
--- !aggregate --
-1.1127811E7	1328.317850309714
+-- !aggregate_variance5 --
+1.1127810999999998E7	1328.317850309714
 
--- !aggregate --
+-- !aggregate_variance6 --
 9882691.114285713	1808.1659023775
 
--- !aggregate --
+-- !aggregate_variance7 --
 9223845.040000001	1239.7633269557336
 
--- !aggregate --
-9891387.555555556	1239.7633269557332
+-- !aggregate_variance8 --
+9891387.555555554	1239.7633269557332
 
--- !aggregate --
+-- !aggregate_variance9 --
 9223845.04	1607.2585798911111
+

regression-test/data/demo/select_union_all_action.out

@@ -8,3 +8,4 @@
 0	abc
 1	123
 2	\N
+

regression-test/data/demo/thread_action.out

@@ -1,7 +1,7 @@
 -- This file is automatically generated. You should know what you did if you want to edit this
--- !diffrent_tag1 --
+-- !diffrent_tag2 --
 100
 
--- !diffrent_tag2 --
+-- !diffrent_tag1 --
 100
 

regression-test/data/empty_table/sql/avg_decimal.sql.generated.out

@@ -0,0 +1,4 @@
+-- This file is automatically generated. You should know what you did if you want to edit this
+-- !avg_decimal --
+\N
+

regression-test/framework/src/main/groovy/org/apache/doris/regression/suite/Suite.groovy

@@ -257,7 +257,7 @@ abstract class Suite extends Script implements GroovyInterceptable {
         logger.info("Execute tag: ${tag}, sql: ${sql}".toString())
 
         if (context.config.generateOutputFile || context.config.forceGenerateOutputFile) {
-            def result = JdbcUtils.executorToStringList(context.getConnection(), sql)
+            def result = JdbcUtils.executeToStringList(context.getConnection(), sql)
             if (order) {
                 result = sortByToString(result)
             }
@@ -281,13 +281,10 @@ abstract class Suite extends Script implements GroovyInterceptable {
             }
 
             OutputUtils.TagBlockIterator expectCsvResults = context.getOutputIterator().next()
-            List<List<Object>> realResults = JdbcUtils.executorToStringList(context.getConnection(), sql)
-            if (order) {
-                realResults = sortByToString(realResults)
-            }
+
             String errorMsg = null
             try {
-                errorMsg = OutputUtils.checkOutput(expectCsvResults, realResults.iterator(), "Check tag '${tag}' failed")
+                errorMsg = OutputUtils.checkOutput(expectCsvResults, context.getConnection(), sql, order, "Check tag '${tag}' failed")
             } catch (Throwable t) {
                 List excelContentList = [context.file.getName(), tag, sql.trim(), t]
                 context.recorder.reportDiffResult(excelContentList)

regression-test/framework/src/main/groovy/org/apache/doris/regression/util/JdbcUtils.groovy

@@ -34,7 +34,7 @@ class JdbcUtils {
         }
     }
 
-    static List<List<Object>> executorToStringList(Connection conn, String sql) {
+    static List<List<Object>> executeToStringList(Connection conn, String sql) {
         conn.prepareStatement(sql).withCloseable { stmt ->
             boolean hasResultSet = stmt.execute()
             if (!hasResultSet) {

regression-test/framework/src/main/groovy/org/apache/doris/regression/util/OutputUtils.groovy

@@ -24,8 +24,20 @@ import org.apache.commons.csv.CSVPrinter
 import org.apache.commons.csv.CSVRecord
 import org.apache.commons.io.LineIterator
 
+import java.sql.Connection
+import java.sql.ResultSet
+
+import static org.apache.doris.regression.util.DataUtils.sortByToString
+
 @CompileStatic
 class OutputUtils {
+    static String toCsvString(Object cell) {
+        StringWriter writer = new StringWriter()
+        def printer = new CSVPrinter(new PrintWriter(writer), CSVFormat.MYSQL)
+        printer.print(cell)
+        return writer.toString()
+    }
+
     static String toCsvString(List<Object> row) {
         StringWriter writer = new StringWriter()
         def printer = new CSVPrinter(new PrintWriter(writer), CSVFormat.MYSQL)
@@ -35,22 +47,65 @@ class OutputUtils {
         return writer.toString()
     }
 
-    static String checkOutput(Iterator<List<String>> expect, Iterator<List<Object>> real, String info) {
-        while (true) {
-            if (expect.hasNext() && !real.hasNext()) {
-                return "${info}, result mismatch, real line is empty, but expect is ${expect.next()}"
-            }
-            if (!expect.hasNext() && real.hasNext()) {
-                return "${info}, result mismatch, expect line is empty, but real is ${toCsvString(real.next())}"
-            }
-            if (!expect.hasNext() && !real.hasNext()) {
-                break
-            }
+    static String checkOutput(Iterator<List<String>> expectIterator, Connection conn, String sql, boolean order, String info) {
+        conn.prepareStatement(sql).withCloseable { stmt ->
+            boolean hasResultSet = stmt.execute()
+            if (!hasResultSet) {
+                if(expectIterator.hasNext()) {
+                    return "${info}, result mismatch, real row number is 0, but expect more than it"
+                }
+            } else {
+                def columnCount = stmt.resultSet.metaData.columnCount
+
+                List<List<Object>> rows = new ArrayList<>()
+                while(stmt.resultSet.next()) {
+                    def row = new ArrayList<>()
+                    for (int i = 1; i <= columnCount; ++i) {
+                        row.add(stmt.resultSet.getObject(i))
+                    }
+                    rows.add(row)
+                }
+                if (order) {
+                    rows = sortByToString(rows)
+                }
 
-            def expectCsvString = toCsvString(expect.next() as List<Object>)
-            def realCsvString = toCsvString(real.next())
-            if (!expectCsvString.equals(realCsvString)) {
-                return "${info}, result mismatch.\nExpect line is: ${expectCsvString}\nBut real is   : ${realCsvString}"
+                for(int i = 1; i <= rows.size(); i++) {
+                    if(expectIterator.hasNext()) {
+                        List<String> expect_row = expectIterator.next()
+                        for(int j = 1; j <= columnCount; j++) {
+                            String expect_cell = toCsvString(expect_row[j-1])
+                            String real_cell = toCsvString(rows[i-1][j-1])
+                            String tp = stmt.resultSet.metaData.getColumnTypeName(j)
+                            if(tp == "FLOAT" || tp == "DOUBLE") {
+                                double expect_double = Double.parseDouble(expect_cell)
+                                double real_double = Double.parseDouble(real_cell)
+
+                                double real_relative_error = Math.abs(expect_double - real_double) / real_double
+                                double expect_relative_error = 1e-10
+
+                                if(expect_relative_error < real_relative_error) {
+                                    return "${info}, ${tp} result mismatch.\nExpect cell is: ${expect_cell}\nBut real is: ${real_cell}\nrelative error is: ${real_relative_error}, bigger than ${expect_relative_error}"
+                                }
+                            } else if(tp == "DATE" || tp =="DATETIME") {
+                                expect_cell = expect_cell.replace("T", " ")
+                                real_cell = real_cell.replace("T", " ")
+
+                                if(!expect_cell.equals(real_cell)) {
+                                    return "${info}, ${tp} result mismatch.\nExpect cell is: ${expect_cell}\nBut real is: ${real_cell}"
+                                }
+                            } else {
+                                if(!expect_cell.equals(real_cell)) {
+                                    return "${info}, ${tp} result mismatch.\nExpect cell is: ${expect_cell}\nBut real is: ${real_cell}"
+                                }
+                            }
+                        }
+                    } else {
+                        return "${info}, result mismatch, real row number is ${rows.size()}, but expect less than it"
+                    }
+                }
+                if(expectIterator.hasNext()) {
+                    return "${info}, result mismatch, real row number is ${rows.size()}, but expect less than it"
+                }
             }
         }
     }

regression-test/suites/aggregate/aggregate.groovy

@@ -23,7 +23,7 @@ def tableName = "datetype"
 
 sql """ DROP TABLE IF EXISTS ${tableName} """
 sql """
-    CREATE TABLE IF NOT EXISTS ${tableName} (
+    CREATE TABLE ${tableName} (
         c_bigint bigint,
         c_double double,
         c_string string,
@@ -74,28 +74,38 @@ streamLoad {
     }
 }
 
-qt_aggregate """ select max(upper(c_string)), min(upper(c_string)) from ${tableName} """
-qt_aggregate """ select avg(c_bigint), avg(c_double) from ${tableName} """
-qt_aggregate """ select avg(distinct c_bigint), avg(distinct c_double) from ${tableName} """
-qt_aggregate """ select count(c_bigint),count(c_double),count(c_string),count(c_date),count(c_timestamp),count(c_boolean) from ${tableName} """
-qt_aggregate """ select count(distinct c_bigint),count(distinct c_double),count(distinct c_string),count(distinct c_date),count(distinct c_timestamp),count(distinct c_boolean) from ${tableName} """
-qt_aggregate """ select max(c_bigint), max(c_double),max(c_string), max(c_date), max(c_timestamp) from ${tableName} """
-qt_aggregate """ select min(c_bigint), min(c_double), min(c_string), min(c_date), min(c_timestamp) from ${tableName} """
-qt_aggregate """ select count(c_string), max(c_double), avg(c_bigint) from ${tableName} """
-qt_aggregate """ select stddev_pop(c_bigint), stddev_pop(c_double) from ${tableName} """
-qt_aggregate """ select stddev_pop(distinct c_bigint), stddev_pop(c_double) from ${tableName} """
-qt_aggregate """ select stddev_pop(c_bigint), stddev_pop(distinct c_double) from ${tableName} """
-qt_aggregate """ select stddev_samp(c_bigint), stddev_samp(c_double) from ${tableName} """
-qt_aggregate """ select stddev_samp(distinct c_bigint), stddev_samp(c_double) from ${tableName} """
-qt_aggregate """ select stddev_samp(c_bigint), stddev_samp(distinct c_double) from ${tableName} """
-qt_aggregate """ select sum(c_bigint), sum(c_double) from ${tableName} """
-qt_aggregate """ select sum(distinct c_bigint), sum(distinct c_double) from ${tableName} """
-qt_aggregate """ select var_pop(c_bigint), var_pop(c_double) from ${tableName} """
-qt_aggregate """ select var_pop(distinct c_bigint), var_pop(c_double) from ${tableName} """
-qt_aggregate """ select var_pop(c_bigint), var_pop(distinct c_double) from ${tableName} """
-qt_aggregate """ select var_samp(c_bigint), var_samp(c_double) from ${tableName} """
-qt_aggregate """ select var_samp(distinct c_bigint), var_samp(c_double) from ${tableName} """
-qt_aggregate """ select var_samp(c_bigint), var_samp(distinct c_double) from ${tableName} """
-qt_aggregate """ select variance(c_bigint), variance(c_double) from ${tableName}  """
-qt_aggregate """ select variance(distinct c_bigint), variance(c_double) from ${tableName}  """
-qt_aggregate """ select variance(c_bigint), variance(distinct c_double) from ${tableName}  """
+qt_aggregate_min_max """ select max(upper(c_string)), min(upper(c_string)) from ${tableName} """
+
+qt_aggregate_avg """ select avg(c_bigint), avg(c_double) from ${tableName} """
+qt_aggregate_avg2 """ select avg(distinct c_bigint), avg(distinct c_double) from ${tableName} """
+
+qt_aggregate_multi_count """ select count(c_bigint),count(c_double),count(c_string),count(c_date),count(c_timestamp),count(c_boolean) from ${tableName} """
+
+qt_aggregate_multi_distinct_count """ select count(distinct c_bigint),count(distinct c_double),count(distinct c_string),count(distinct c_date),count(distinct c_timestamp),count(distinct c_boolean) from ${tableName} """
+
+qt_aggregate_multi_max """ select max(c_bigint), max(c_double),max(c_string), max(c_date), max(c_timestamp) from ${tableName} """
+
+qt_aggregate_multi_min """ select min(c_bigint), min(c_double), min(c_string), min(c_date), min(c_timestamp) from ${tableName} """
+
+qt_aggregate_mix """ select count(c_string), max(c_double), avg(c_bigint) from ${tableName} """
+
+qt_aggregate_stddev1 """ select stddev_pop(c_bigint), stddev_pop(c_double) from ${tableName} """
+qt_aggregate_stddev2 """ select stddev_pop(distinct c_bigint), stddev_pop(c_double) from ${tableName} """
+qt_aggregate_stddev3 """ select stddev_pop(c_bigint), stddev_pop(distinct c_double) from ${tableName} """
+qt_aggregate_stddev4 """ select stddev_samp(c_bigint), stddev_samp(c_double) from ${tableName} """
+qt_aggregate_stddev5 """ select stddev_samp(distinct c_bigint), stddev_samp(c_double) from ${tableName} """
+qt_aggregate_stddev6 """ select stddev_samp(c_bigint), stddev_samp(distinct c_double) from ${tableName} """
+
+qt_aggregate_sum """ select sum(c_bigint), sum(c_double) from ${tableName} """
+
+qt_aggregate_distinct_sum """ select sum(distinct c_bigint), sum(distinct c_double) from ${tableName} """
+
+qt_aggregate_variance """ select var_pop(c_bigint), var_pop(c_double) from ${tableName} """
+qt_aggregate_variance2 """ select var_pop(distinct c_bigint), var_pop(c_double) from ${tableName} """
+qt_aggregate_variance3 """ select var_pop(c_bigint), var_pop(distinct c_double) from ${tableName} """
+qt_aggregate_variance4 """ select var_samp(c_bigint), var_samp(c_double) from ${tableName} """
+qt_aggregate_variance5 """ select var_samp(distinct c_bigint), var_samp(c_double) from ${tableName} """
+qt_aggregate_variance6 """ select var_samp(c_bigint), var_samp(distinct c_double) from ${tableName} """
+qt_aggregate_variance7 """ select variance(c_bigint), variance(c_double) from ${tableName}  """
+qt_aggregate_variance8 """ select variance(distinct c_bigint), variance(c_double) from ${tableName}  """
+qt_aggregate_variance9 """ select variance(c_bigint), variance(distinct c_double) from ${tableName}  """

run-regression-test.sh

@@ -133,6 +133,7 @@ RUN_JAR=${OUTPUT_DIR}/lib/regression-test-*.jar
 if [ ${CLEAN} -eq 1 ]; then
     rm -rf ${REGRESSION_TEST_BUILD_DIR}
     rm -rf ${OUTPUT_DIR}
+    find ${DORIS_HOME}/regression-test/suites/ -name "*.generated.groovy"| xargs -r rm
 fi
 
 if [ ${RUN} -ne 1 ]; then