[opt](partial update) Allow to only specify key columns in partial update

be/src/olap/rowset/segment_v2/segment_writer.cpp

@@ -535,7 +535,7 @@ Status SegmentWriter::probe_key_for_mow(
 // 3. set columns to data convertor and then write all columns
 Status SegmentWriter::append_block_with_partial_content(const vectorized::Block* block,
                                                         size_t row_pos, size_t num_rows) {
-    if (block->columns() <= _tablet_schema->num_key_columns() ||
+    if (block->columns() < _tablet_schema->num_key_columns() ||
         block->columns() >= _tablet_schema->num_columns()) {
         return Status::InvalidArgument(
                 fmt::format("illegal partial update block columns: {}, num key columns: {}, total "

be/src/olap/rowset/segment_v2/vertical_segment_writer.cpp

@@ -555,7 +555,7 @@ Status VerticalSegmentWriter::batch_block(const vectorized::Block* block, size_t
         _opts.rowset_ctx->partial_update_info->is_partial_update &&
         _opts.write_type == DataWriteType::TYPE_DIRECT &&
         !_opts.rowset_ctx->is_transient_rowset_writer) {
-        if (block->columns() <= _tablet_schema->num_key_columns() ||
+        if (block->columns() < _tablet_schema->num_key_columns() ||
             block->columns() >= _tablet_schema->num_columns()) {
             return Status::InvalidArgument(fmt::format(
                     "illegal partial update block columns: {}, num key columns: {}, total "

regression-test/data/unique_with_mow_p0/partial_update/schema_change/load_with_key_column.csv

@@ -1 +1 @@
-1, 1
+2, 2

regression-test/data/unique_with_mow_p0/partial_update/test_partial_update_only_keys.out

@@ -0,0 +1,59 @@
+-- This file is automatically generated. You should know what you did if you want to edit this
+-- !sql --
+0	0	0	0
+1	1	1	1
+2	2	2	2
+
+-- !sql --
+0	0	0	0
+1	1	1	1
+2	2	2	2
+4	\N	\N	\N
+5	\N	\N	\N
+6	\N	\N	\N
+
+-- !sql --
+0	0	0	0
+1	1	1	1
+2	2	2	2
+4	\N	\N	\N
+5	\N	\N	\N
+6	\N	\N	\N
+
+-- !sql --
+0	0	0	0
+1	1	1	1
+2	2	2	2
+4	\N	\N	\N
+5	\N	\N	\N
+6	\N	\N	\N
+
+-- !sql --
+0	0	0	0
+1	1	1	1
+2	2	2	2
+
+-- !sql --
+0	0	0	0
+1	1	1	1
+2	2	2	2
+4	\N	\N	\N
+5	\N	\N	\N
+6	\N	\N	\N
+
+-- !sql --
+0	0	0	0
+1	1	1	1
+2	2	2	2
+4	\N	\N	\N
+5	\N	\N	\N
+6	\N	\N	\N
+
+-- !sql --
+0	0	0	0
+1	1	1	1
+2	2	2	2
+4	\N	\N	\N
+5	\N	\N	\N
+6	\N	\N	\N
+

regression-test/suites/unique_with_mow_p0/partial_update/test_partial_update_insert_light_schema_change.groovy

@@ -264,11 +264,11 @@ suite("test_partial_update_insert_light_schema_change", "p0") {
             sql "sync"
 
             // test insert data with all key column, should fail because
-            // it don't have any value columns
+            // it inserts a new row in strict mode
             sql "set enable_unique_key_partial_update=true;"
             test {
                 sql "insert into ${tableName}(c0,c1) values(1, 1);"
-                exception "INTERNAL_ERROR"
+                exception "Insert has filtered data in strict mode"
             }
             sql "insert into ${tableName}(c0,c1,c2) values(1,0,10);"
             sql "set enable_unique_key_partial_update=false;"

regression-test/suites/unique_with_mow_p0/partial_update/test_partial_update_insert_schema_change.groovy

@@ -249,11 +249,11 @@ suite("test_partial_update_insert_schema_change", "p0") {
     sql "sync"
 
     // test insert data with all key column, should fail because
-    // it don't have any value columns
+    // it inserts a new row in strict mode
     sql "set enable_unique_key_partial_update=true;"
     test {
         sql "insert into ${tableName}(c0,c1) values(1, 1);"
-        exception "INTERNAL_ERROR"
+        exception "Insert has filtered data in strict mode"
     }
     sql "insert into ${tableName}(c0,c1,c2) values(1,0,10);"
     sql "set enable_unique_key_partial_update=false;"

regression-test/suites/unique_with_mow_p0/partial_update/test_partial_update_only_keys.groovy

@@ -0,0 +1,62 @@
+
+// Licensed to the Apache Software Foundation (ASF) under one
+// or more contributor license agreements.  See the NOTICE file
+// distributed with this work for additional information
+// regarding copyright ownership.  The ASF licenses this file
+// to you under the Apache License, Version 2.0 (the
+// "License"); you may not use this file except in compliance
+// with the License.  You may obtain a copy of the License at
+//
+//   http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing,
+// software distributed under the License is distributed on an
+// "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+// KIND, either express or implied.  See the License for the
+// specific language governing permissions and limitations
+// under the License.
+
+suite("test_partial_update_only_keys", "p0") {
+
+    String db = context.config.getDbNameByFile(context.file)
+    sql "select 1;" // to create database
+
+    for (def use_row_store : [false, true]) {
+        logger.info("current params: use_row_store: ${use_row_store}")
+
+        connect(user = context.config.jdbcUser, password = context.config.jdbcPassword, url = context.config.jdbcUrl) {
+            sql "use ${db};"
+            def tableName = "test_primary_key_partial_update"
+            sql """ DROP TABLE IF EXISTS ${tableName} force"""
+            sql """ CREATE TABLE ${tableName} (
+                    `k` BIGINT NOT NULL,
+                    `c1` int,
+                    `c2` int,
+                    `c3` int)
+                    UNIQUE KEY(`k`) DISTRIBUTED BY HASH(`k`) BUCKETS 1
+                    PROPERTIES(
+                        "replication_num" = "1",
+                        "enable_unique_key_merge_on_write" = "true",
+                        "store_row_column" = "${use_row_store}"); """
+            sql """insert into ${tableName} select number,number,number,number from numbers("number"="3");"""
+            qt_sql """select * from ${tableName} order by k;"""
+            // new rows will be appended
+            sql "set enable_unique_key_partial_update=true;"
+            sql "set enable_insert_strict=false;"
+            sql "sync"
+            sql "insert into ${tableName}(k) values(0),(1),(4),(5),(6);"
+            qt_sql """select * from ${tableName} order by k;"""
+
+            // fail if has new rows
+            sql "set enable_insert_strict=true;"
+            sql "sync"
+            sql "insert into ${tableName}(k) values(0),(1),(4),(5),(6);"
+            qt_sql """select * from ${tableName} order by k;"""
+            test {
+                sql "insert into ${tableName}(k) values(0),(1),(10),(11);"
+                exception "Insert has filtered data in strict mode"
+            }
+            qt_sql """select * from ${tableName} order by k;"""
+        }
+    }
+}

regression-test/suites/unique_with_mow_p0/partial_update/test_partial_update_schema_change.groovy

@@ -415,12 +415,13 @@ suite("test_partial_update_schema_change", "p0") {
     });
 
     // test load data with all key column, should fail because
-    // it don't have any value columns
+    // it inserts a new row in strict mode
     streamLoad {
         table "${tableName}"
 
         set 'column_separator', ','
         set 'partial_columns', 'true'
+        set 'strict_mode', 'true'
         set 'columns', 'c0, c1'
 
         file 'schema_change/load_with_key_column.csv'
@@ -434,7 +435,7 @@ suite("test_partial_update_schema_change", "p0") {
             def json = parseJson(result)
             assertEquals("fail", json.Status.toLowerCase())
             assertEquals(1, json.NumberTotalRows)
-            assertEquals(0, json.NumberFilteredRows)
+            assertEquals(1, json.NumberFilteredRows)
             assertEquals(0, json.NumberUnselectedRows)
         }
     }
@@ -1001,6 +1002,7 @@ suite("test_partial_update_schema_change", "p0") {
 
         set 'column_separator', ','
         set 'partial_columns', 'true'
+        set 'strict_mode', 'true'
         set 'columns', 'c0, c1'
 
         file 'schema_change/load_with_key_column.csv'
@@ -1014,7 +1016,7 @@ suite("test_partial_update_schema_change", "p0") {
             def json = parseJson(result)
             assertEquals("fail", json.Status.toLowerCase())
             assertEquals(1, json.NumberTotalRows)
-            assertEquals(0, json.NumberFilteredRows)
+            assertEquals(1, json.NumberFilteredRows)
             assertEquals(0, json.NumberUnselectedRows)
         }
     }

regression-test/suites/unique_with_mow_p0/partial_update/test_partial_update_schema_change_row_store.groovy

@@ -419,12 +419,13 @@ suite("test_partial_update_row_store_schema_change", "p0") {
     });
 
     // test load data with all key column, should fail because
-    // it don't have any value columns
+    // it inserts a new row in strict mode
     streamLoad {
         table "${tableName}"
 
         set 'column_separator', ','
         set 'partial_columns', 'true'
+        set 'strict_mode', 'true'
         set 'columns', 'c0, c1'
 
         file 'schema_change/load_with_key_column.csv'
@@ -438,7 +439,7 @@ suite("test_partial_update_row_store_schema_change", "p0") {
             def json = parseJson(result)
             assertEquals("fail", json.Status.toLowerCase())
             assertEquals(1, json.NumberTotalRows)
-            assertEquals(0, json.NumberFilteredRows)
+            assertEquals(1, json.NumberFilteredRows)
             assertEquals(0, json.NumberUnselectedRows)
         }
     }
@@ -1014,6 +1015,7 @@ suite("test_partial_update_row_store_schema_change", "p0") {
 
         set 'column_separator', ','
         set 'partial_columns', 'true'
+        set 'strict_mode', 'true'
         set 'columns', 'c0, c1'
 
         file 'schema_change/load_with_key_column.csv'
@@ -1027,7 +1029,7 @@ suite("test_partial_update_row_store_schema_change", "p0") {
             def json = parseJson(result)
             assertEquals("fail", json.Status.toLowerCase())
             assertEquals(1, json.NumberTotalRows)
-            assertEquals(0, json.NumberFilteredRows)
+            assertEquals(1, json.NumberFilteredRows)
             assertEquals(0, json.NumberUnselectedRows)
         }
     }