From ac610cbe7d212594f1cd2a03898363c34ea5e87c Mon Sep 17 00:00:00 2001
From: zzzxl <yangsiyu@selectdb.com>
Date: Mon, 23 Dec 2024 10:24:01 +0800
Subject: [PATCH] [fix](inverted index) Fix the issue with incorrect seek
 results in DICT_COMPRESS (#45738)

Related PR: https://github.com/apache/doris/pull/44414

Problem Summary:
In inverted index version 3 mode, using dictionary compression may
lead to incorrect results after a seek operation.
---
 be/src/clucene                                |  2 +-
 .../test_inverted_index_v3.out                | 12 +++++++++
 .../test_inverted_index_v3.groovy             | 26 +++++++++++++++++++
 3 files changed, 39 insertions(+), 1 deletion(-)

diff --git a/be/src/clucene b/be/src/clucene
index a506dbb6c523aa..2204eaec46a68e 160000
--- a/be/src/clucene
+++ b/be/src/clucene
@@ -1 +1 @@
-Subproject commit a506dbb6c523aa65044eb1c527a066d236172543
+Subproject commit 2204eaec46a68e5e9a1876b7021f24839ecb2cf0
diff --git a/regression-test/data/inverted_index_p0/test_inverted_index_v3.out b/regression-test/data/inverted_index_p0/test_inverted_index_v3.out
index 9dc20f3e0e0a85..53f4eb7ae0a667 100644
--- a/regression-test/data/inverted_index_p0/test_inverted_index_v3.out
+++ b/regression-test/data/inverted_index_p0/test_inverted_index_v3.out
@@ -23,3 +23,15 @@
 -- !sql --
 105
 
+-- !sql --
+238
+
+-- !sql --
+104
+
+-- !sql --
+104
+
+-- !sql --
+105
+
diff --git a/regression-test/suites/inverted_index_p0/test_inverted_index_v3.groovy b/regression-test/suites/inverted_index_p0/test_inverted_index_v3.groovy
index ea7dd0b595f504..82389d84e3cd67 100644
--- a/regression-test/suites/inverted_index_p0/test_inverted_index_v3.groovy
+++ b/regression-test/suites/inverted_index_p0/test_inverted_index_v3.groovy
@@ -19,9 +19,11 @@
 suite("test_inverted_index_v3", "p0"){
     def indexTbName1 = "test_inverted_index_v3_1"
     def indexTbName2 = "test_inverted_index_v3_2"
+    def indexTbName3 = "test_inverted_index_v3_3"
 
     sql "DROP TABLE IF EXISTS ${indexTbName1}"
     sql "DROP TABLE IF EXISTS ${indexTbName2}"
+    sql "DROP TABLE IF EXISTS ${indexTbName3}"
 
     sql """
       CREATE TABLE ${indexTbName1} (
@@ -59,6 +61,24 @@ suite("test_inverted_index_v3", "p0"){
       );
     """
 
+    sql """
+      CREATE TABLE ${indexTbName3} (
+      `@timestamp` int(11) NULL COMMENT "",
+      `clientip` varchar(20) NULL COMMENT "",
+      `request` text NULL COMMENT "",
+      `status` int(11) NULL COMMENT "",
+      `size` int(11) NULL COMMENT "",
+      INDEX request_idx (`request`) USING INVERTED PROPERTIES("parser" = "english", "support_phrase" = "true", "dict_compression" = "true") COMMENT ''
+      ) ENGINE=OLAP
+      DUPLICATE KEY(`@timestamp`)
+      COMMENT "OLAP"
+      DISTRIBUTED BY RANDOM BUCKETS 1
+      PROPERTIES (
+      "replication_allocation" = "tag.location.default: 1",
+      "inverted_index_storage_format" = "V3"
+      );
+    """
+
     def load_httplogs_data = {table_name, label, read_flag, format_flag, file_name, ignore_failure=false,
                         expected_succ_rows = -1, load_to_single_tablet = 'true' ->
         
@@ -99,6 +119,7 @@ suite("test_inverted_index_v3", "p0"){
     try {
       load_httplogs_data.call(indexTbName1, indexTbName1, 'true', 'json', 'documents-1000.json')
       load_httplogs_data.call(indexTbName2, indexTbName2, 'true', 'json', 'documents-1000.json')
+      load_httplogs_data.call(indexTbName3, indexTbName3, 'true', 'json', 'documents-1000.json')
       
       sql "sync"
 
@@ -112,6 +133,11 @@ suite("test_inverted_index_v3", "p0"){
       qt_sql """ select count() from ${indexTbName2} where request match_phrase 'hm bg'; """
       qt_sql """ select count() from ${indexTbName2} where request match_phrase_prefix 'hm bg'; """
 
+      qt_sql """ select count() from ${indexTbName3} where request match_any 'hm bg'; """
+      qt_sql """ select count() from ${indexTbName3} where request match_all 'hm bg'; """
+      qt_sql """ select count() from ${indexTbName3} where request match_phrase 'hm bg'; """
+      qt_sql """ select count() from ${indexTbName3} where request match_phrase_prefix 'hm bg'; """
+
     } finally {
     }
 }
\ No newline at end of file