branch-2.1: [fix](be) fix parquet file reader not updating page index when processing it #52228 (#52782)

Cherry-picked from #52228

Co-authored-by: SWEI <z_swii@outlook.com>
Co-authored-by: zengsiwei <zengsiwei@kuaishou.com>
Co-authored-by: suxiaogang223 <suxiaogang223@icloud.com>

be/src/common/config.cpp

@@ -1345,7 +1345,7 @@ DEFINE_mInt64(compaction_batch_size, "-1");
 // If set to false, the parquet reader will not use page index to filter data.
 // This is only for debug purpose, in case sometimes the page index
 // filter wrong data.
-DEFINE_mBool(enable_parquet_page_index, "false");
+DEFINE_mBool(enable_parquet_page_index, "true");
 
 DEFINE_mBool(ignore_not_found_file_in_external_table, "true");
 

be/src/vec/exec/format/parquet/vparquet_reader.cpp

@@ -860,7 +860,7 @@ Status ParquetReader::_process_page_index(const tparquet::RowGroup& row_group,
             // use the union row range
             skipped_row_ranges.emplace_back(skipped_row_range);
         }
-        _col_offsets.emplace(parquet_col_id, offset_index);
+        _col_offsets[parquet_col_id] = offset_index;
     }
     if (skipped_row_ranges.empty()) {
         read_whole_row_group();

regression-test/data/external_table_p0/hive/test_hive_parquet_skip_page.out

@@ -1,4 +1,8 @@
 -- This file is automatically generated. You should know what you did if you want to edit this
+-- !small_2rowgroup --
+1024
+4049
+
 -- !q01 --
 1	2132	4633	4	28.00	28955.64	0.09	0.06	N	O	1996-04-21	1996-03-30	1996-05-16	NONE	AIR	lites. fluffily even de
 1	15635	638	6	32.00	49620.16	0.07	0.02	N	O	1996-01-30	1996-02-07	1996-02-03	DELIVER IN PERSON	MAIL	arefully slyly ex

regression-test/suites/external_table_p0/hive/test_hive_parquet_skip_page.groovy

@@ -94,11 +94,21 @@ suite("test_hive_parquet_skip_page", "p0,external,hive,external_docker,external_
         return;
     }
 
+    String externalEnvIp = context.config.otherConfigs.get("externalEnvIp")
+    def hdfs_port = context.config.otherConfigs.get("hive2HdfsPort")
+    def defaultFS = "hdfs://${externalEnvIp}:${hdfs_port}"
+    def hdfsUserName = "doris"
+    def uri = "${defaultFS}" + "/user/doris/preinstalled_data/parquet/small_2rowgroup.parquet"
+    qt_small_2rowgroup """ select * from HDFS(
+                "uri" = "${uri}",
+                "hadoop.username" = "${hdfsUserName}",
+                "format" = "parquet") where a = 1024 or a = 4049
+                order by a;"""
+
     for (String hivePrefix : ["hive2", "hive3"]) {
         try {
             String hms_port = context.config.otherConfigs.get(hivePrefix + "HmsPort")
             String catalog_name = "${hivePrefix}_test_parquet_skip_page"
-            String externalEnvIp = context.config.otherConfigs.get("externalEnvIp")
 
             sql """drop catalog if exists ${catalog_name}"""
             sql """create catalog if not exists ${catalog_name} properties (