[Fix](segment iterator) fix shrink non-char column coredump (apache#3…

…6275)

If we execute a delete predicate in an inverted index table and then
query something from it, it will cause a core dump in
shrink_char_type_column_suffix_zero. This occurs because the delete
predicate column ID in _char_type_idx is incorrectly shifted to the
result column inserted by the inverted index result.

coredump stack like:
```
COULD NOT CREATE A LOGGINGFILE 20240604-145331!F20240604 14:53:31.991016 24178 column.h:134] Cannot get_shrinked_column() column Const(UInt8)
*** Check failure stack trace: ***
F20240604 14:53:31.991436 24158 column.h:134] Cannot get_shrinked_column() column Const(UInt8)
*** Check failure stack trace: ***
F20240604 14:53:31.991436 24158 column.h:134] Cannot get_shrinked_column() column Const(UInt8)F20240604 14:53:31.991645 24160 column.h:134] Cannot get_shrinked_column() column Const(UInt8)

    at /home/zcp/repo_center/doris_release/doris/be/src/vec/core/block.cpp:1126
    at /home/zcp/repo_center/doris_release/doris/be/src/olap/rowset/segment_v2/segment_iterator.cpp:2408
    at /home/zcp/repo_center/doris_release/doris/be/src/olap/rowset/segment_v2/segment_iterator.cpp:2090
    at /home/zcp/repo_center/doris_release/doris/be/src/olap/rowset/beta_rowset_reader.cpp:342
    at /home/zcp/repo_center/doris_release/doris/be/src/vec/olap/vcollect_iterator.h:256
    at /home/zcp/repo_center/doris_release/doris/be/src/vec/olap/vcollect_iterator.cpp:514
    at /home/zcp/repo_center/doris_release/doris/be/src/vec/olap/vcollect_iterator.cpp:493
    at /home/zcp/repo_center/doris_release/doris/be/src/vec/olap/vcollect_iterator.cpp:692
    at /home/zcp/repo_center/doris_release/doris/be/src/vec/olap/vcollect_iterator.cpp:186
    at /home/zcp/repo_center/doris_release/doris/be/src/vec/olap/block_reader.cpp:156
    at /home/zcp/repo_center/doris_release/doris/be/src/vec/olap/block_reader.cpp:228
--Type <RET> for more, q to quit, c to continue without paging--c
    at /home/zcp/repo_center/doris_release/doris/be/src/vec/exec/scan/new_olap_scanner.cpp:227
```

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

@@ -310,9 +310,6 @@ Status SegmentIterator::_init_impl(const StorageReadOptions& opts) {
     }
 
     RETURN_IF_ERROR(init_iterators());
-    if (_char_type_idx.empty() && _char_type_idx_no_0.empty()) {
-        _vec_init_char_column_id();
-    }
 
     if (opts.output_columns != nullptr) {
         _output_columns = *(opts.output_columns);
@@ -1676,15 +1673,22 @@ bool SegmentIterator::_has_char_type(const Field& column_desc) {
     }
 };
 
-void SegmentIterator::_vec_init_char_column_id() {
+void SegmentIterator::_vec_init_char_column_id(vectorized::Block* block) {
     for (size_t i = 0; i < _schema->num_column_ids(); i++) {
         auto cid = _schema->column_id(i);
         const Field* column_desc = _schema->column(cid);
 
-        if (_has_char_type(*column_desc)) {
-            _char_type_idx.emplace_back(i);
-            if (i != 0) {
-                _char_type_idx_no_0.emplace_back(i);
+        // The additional deleted filter condition will be in the materialized column at the end of the block.
+        // After _output_column_by_sel_idx, it will be erased, so we do not need to shrink it.
+        if (i < block->columns()) {
+            if (_has_char_type(*column_desc)) {
+                _char_type_idx.emplace_back(i);
+                if (i != 0) {
+                    _char_type_idx_no_0.emplace_back(i);
+                }
+            }
+            if (column_desc->type() == FieldType::OLAP_FIELD_TYPE_CHAR) {
+                _is_char_type[cid] = true;
             }
         }
     }
@@ -2037,6 +2041,9 @@ Status SegmentIterator::_next_batch_internal(vectorized::Block* block) {
             _block_rowids.resize(_opts.block_row_max);
         }
         _current_return_columns.resize(_schema->columns().size());
+        if (_char_type_idx.empty() && _char_type_idx_no_0.empty()) {
+            _vec_init_char_column_id(block);
+        }
         for (size_t i = 0; i < _schema->num_column_ids(); i++) {
             auto cid = _schema->column_id(i);
             auto column_desc = _schema->column(cid);

be/src/olap/rowset/segment_v2/segment_iterator.h

@@ -200,7 +200,7 @@ class SegmentIterator : public RowwiseIterator {
     // TODO: Fix Me
     // CHAR type in storage layer padding the 0 in length. But query engine need ignore the padding 0.
     // so segment iterator need to shrink char column before output it. only use in vec query engine.
-    void _vec_init_char_column_id();
+    void _vec_init_char_column_id(vectorized::Block* block);
     bool _has_char_type(const Field& column_desc);
 
     uint32_t segment_id() const { return _segment->id(); }

regression-test/data/inverted_index_p0/test_delete.out

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

regression-test/suites/inverted_index_p0/test_delete.groovy

@@ -0,0 +1,60 @@
+// 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_delete"){
+    // prepare test table
+
+    def timeout = 60000
+    def delta_time = 1000
+    def alter_res = "null"
+    def useTime = 0
+
+    def indexTblName = "test_delete"
+
+    sql "DROP TABLE IF EXISTS ${indexTblName}"
+    // create 1 replica table
+    sql """
+	CREATE TABLE `${indexTblName}` (
+      `a` int NULL COMMENT '',
+      `b` varchar(60) NOT NULL COMMENT '',
+      `c` char(10) NULL COMMENT '',
+      INDEX index_b(b) USING INVERTED  COMMENT '',
+      INDEX index_c(c) USING INVERTED  COMMENT ''
+    ) ENGINE=OLAP
+    DUPLICATE KEY(`a`)
+    COMMENT 'OLAP'
+    DISTRIBUTED BY HASH(`a`) BUCKETS 1
+    PROPERTIES (
+    "replication_allocation" = "tag.location.default: 1",
+    "is_being_synced" = "false",
+    "storage_format" = "V2",
+    "light_schema_change" = "true",
+    "disable_auto_compaction" = "false",
+    "enable_single_replica_compaction" = "false"
+    );
+    """
+
+    sql """ INSERT INTO `${indexTblName}`(`a`, `b`, `c`) VALUES ('1', '6afef581285b6608bf80d5a4e46cf839', 'aaa'), ('2', '48a33ec3453a28bce84b8f96fe161956', 'bbb'),
+                                                                ('3', '021603e7dcfe65d44af0efd0e5aee154', 'ccc'), ('4', 'ee27ee1da291e46403c408e220bed6e1', 'ddd'),
+                                                                ('5', 'a648a447b8f71522f11632eba4b4adde', 'eee'), ('6', 'a9fb5c985c90bf05f3bee5ca3ae95260', 'fff'),
+                                                                ('7', '0974e7a82e30d1af83205e474fadd0a2', 'ggg'); """
+
+
+    sql """ DELETE FROM ${indexTblName} WHERE c IN ('aaa','ccc'); """
+
+    qt_sql """ SELECT count(1) as cnt FROM ${indexTblName} WHERE a BETWEEN 1 AND 6 AND b IN ('48a33ec3453a28bce84b8f96fe161956', 'a9fb5c985c90bf05f3bee5ca3ae95260'); """
+}