branch-3.0: [fix](nereids) fix generate column need add cast in stream load #49167

fe/fe-core/src/main/java/org/apache/doris/nereids/trees/plans/commands/info/CreateTableInfo.java

@@ -58,6 +58,7 @@
 import org.apache.doris.nereids.properties.PhysicalProperties;
 import org.apache.doris.nereids.rules.analysis.ExpressionAnalyzer;
 import org.apache.doris.nereids.rules.expression.ExpressionRewriteContext;
+import org.apache.doris.nereids.trees.expressions.Cast;
 import org.apache.doris.nereids.trees.expressions.Expression;
 import org.apache.doris.nereids.trees.expressions.Slot;
 import org.apache.doris.nereids.trees.expressions.SlotReference;
@@ -71,6 +72,7 @@
 import org.apache.doris.nereids.trees.expressions.visitor.DefaultExpressionRewriter;
 import org.apache.doris.nereids.trees.plans.logical.LogicalEmptyRelation;
 import org.apache.doris.nereids.types.DataType;
+import org.apache.doris.nereids.types.coercion.CharacterType;
 import org.apache.doris.nereids.util.TypeCoercionUtils;
 import org.apache.doris.nereids.util.Utils;
 import org.apache.doris.qe.ConnectContext;
@@ -940,19 +942,28 @@ private void generatedColumnCheck(ConnectContext ctx) {
             Expression boundSlotExpression = SlotReplacer.INSTANCE.replace(parsedExpression, columnToSlotReference);
             Scope scope = new Scope(slots);
             ExpressionAnalyzer analyzer = new ExpressionAnalyzer(null, scope, cascadesContext, false, false);
-            Expression expr;
+            Expression expression;
             try {
-                expr = analyzer.analyze(boundSlotExpression, new ExpressionRewriteContext(cascadesContext));
+                expression = analyzer.analyze(boundSlotExpression, new ExpressionRewriteContext(cascadesContext));
             } catch (AnalysisException e) {
                 throw new AnalysisException("In generated column '" + column.getName() + "', "
                         + Utils.convertFirstChar(e.getMessage()));
             }
-            checkExpressionInGeneratedColumn(expr, column, nameToColumnDefinition);
-            TypeCoercionUtils.checkCanCastTo(expr.getDataType(), column.getType());
+            checkExpressionInGeneratedColumn(expression, column, nameToColumnDefinition);
+            TypeCoercionUtils.checkCanCastTo(expression.getDataType(), column.getType());
             ExpressionToExpr translator = new ExpressionToExpr(i, translateMap);
-            Expr e = expr.accept(translator, planTranslatorContext);
-            info.get().setExpr(e);
-            exprAndnames.add(new GeneratedColumnUtil.ExprAndname(e.clone(), column.getName()));
+            Expr expr = expression.accept(translator, planTranslatorContext);
+            info.get().setExpr(expr);
+            // Casting slot to its own type is because when loading data(stream load and other load),
+            // the slots reading from files are string type. So we need to cast it to its own type to avoid error.
+            Expression expressionForLoad = expression.rewriteDownShortCircuit(e -> {
+                if (e instanceof SlotReference && !(e.getDataType() instanceof CharacterType)) {
+                    return new Cast(e, e.getDataType());
+                }
+                return e;
+            });
+            Expr exprForLoad = expressionForLoad.accept(translator, planTranslatorContext);
+            exprAndnames.add(new GeneratedColumnUtil.ExprAndname(exprForLoad.clone(), column.getName()));
         }
 
         // for alter drop column

regression-test/data/ddl_p0/test_create_table_generated_column/objects3.csv

@@ -0,0 +1,8 @@
+5113756983984045899
+5113756983984045898
+3423432435553321
+55545645564565
+5113756983984045895
+4663423423432
+5113756983984045894
+5113756983984045897

regression-test/data/ddl_p0/test_create_table_generated_column/stream_load_and_mysql_load.out

@@ -69,3 +69,13 @@
 1	2	3.0
 2	3	5.0
 
+-- !test_load_cast --
+1	3423432435553321
+2	4663423423432
+4	5113756983984045894
+5	55545645564565
+5	5113756983984045895
+7	5113756983984045897
+8	5113756983984045898
+9	5113756983984045899
+

regression-test/suites/ddl_p0/test_create_table_generated_column/stream_load_and_mysql_load.groovy

@@ -141,4 +141,40 @@ suite("test_generated_column_stream_mysql_load") {
     """
     sql "sync"
     qt_specify_value_for_gencol "select * from test_gen_col_common_steam_mysql_load order by 1,2,3"
+
+    sql """
+    drop table if exists objects3;
+    """
+    sql """
+    CREATE TABLE `objects3` (
+      `OBJECTIDHASH` tinyint AS ((`OBJECTID` % 10)) NULL,
+      `OBJECTID` bigint NOT NULL,
+    ) ENGINE=OLAP
+    UNIQUE KEY(`OBJECTIDHASH`, `OBJECTID`)
+    DISTRIBUTED BY HASH( `OBJECTIDHASH`) BUCKETS 128
+    PROPERTIES (
+        "file_cache_ttl_seconds" = "0",
+        "is_being_synced" = "false",
+        "storage_medium" = "hdd",
+        "storage_format" = "V2",
+        "inverted_index_storage_format" = "V2",
+        "compression" = "ZSTD",
+        "enable_unique_key_merge_on_write" = "true",
+        "light_schema_change" = "true",
+        "disable_auto_compaction" = "false",
+        "enable_single_replica_compaction" = "false",
+        "group_commit_interval_ms" = "5000",
+        "group_commit_data_bytes" = "134217728",
+        "enable_mow_light_delete" = "false"
+    );
+    """
+    streamLoad {
+        table 'objects3'
+        set 'column_separator', ','
+        file 'objects3.csv'
+        set 'columns', 'OBJECTID'
+        time 10000 // limit inflight 10s
+    }
+    sql "sync"
+    qt_test_load_cast "select * from objects3 order by OBJECTIDHASH,OBJECTID"
 }