(doris) branch master updated: [fix](invert index) fix error handling for match_regexp resulting in an empty match. (#29233)

[jianliangqi]

This is an automated email from the ASF dual-hosted git repository.

jianliangqi pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/doris.git


The following commit(s) were added to refs/heads/master by this push:
     new c8a0d3e03cc [fix](invert index) fix error handling for match_regexp 
resulting in an empty match. (#29233)
c8a0d3e03cc is described below

commit c8a0d3e03ccd13af5f18fa2af1a2fb41ecae2790
Author: zzzxl <33418555+zzzxl1...@users.noreply.github.com>
AuthorDate: Thu Dec 28 19:58:41 2023 +0800

    [fix](invert index) fix error handling for match_regexp resulting in an 
empty match. (#29233)
---
 .../rowset/segment_v2/inverted_index/query/regexp_query.cpp    | 10 ++++++++--
 .../data/inverted_index_p0/test_index_match_regexp.out         |  3 +++
 .../suites/inverted_index_p0/test_index_match_regexp.groovy    |  2 ++
 3 files changed, 13 insertions(+), 2 deletions(-)

diff --git 
a/be/src/olap/rowset/segment_v2/inverted_index/query/regexp_query.cpp 
b/be/src/olap/rowset/segment_v2/inverted_index/query/regexp_query.cpp
index 83c5401bac0..ee959e1da73 100644
--- a/be/src/olap/rowset/segment_v2/inverted_index/query/regexp_query.cpp
+++ b/be/src/olap/rowset/segment_v2/inverted_index/query/regexp_query.cpp
@@ -70,10 +70,12 @@ void RegexpQuery::add(const std::wstring& field_name, const 
std::string& pattern
             }
 
             if (is_match) {
-                terms.emplace_back(std::move(input));
-                if (++count >= _max_expansions) {
+                if (_max_expansions > 0 && count >= _max_expansions) {
                     break;
                 }
+
+                terms.emplace_back(std::move(input));
+                count++;
             }
 
             _CLDECDELETE(term);
@@ -88,6 +90,10 @@ void RegexpQuery::add(const std::wstring& field_name, const 
std::string& pattern
         hs_free_database(database);
     })
 
+    if (terms.empty()) {
+        return;
+    }
+
     query.add(field_name, terms);
 }
 
diff --git a/regression-test/data/inverted_index_p0/test_index_match_regexp.out 
b/regression-test/data/inverted_index_p0/test_index_match_regexp.out
index eab27de65ee..f9a9caf6d74 100644
--- a/regression-test/data/inverted_index_p0/test_index_match_regexp.out
+++ b/regression-test/data/inverted_index_p0/test_index_match_regexp.out
@@ -14,3 +14,6 @@
 -- !sql --
 38
 
+-- !sql --
+0
+
diff --git 
a/regression-test/suites/inverted_index_p0/test_index_match_regexp.groovy 
b/regression-test/suites/inverted_index_p0/test_index_match_regexp.groovy
index 7567102d596..3458f086740 100644
--- a/regression-test/suites/inverted_index_p0/test_index_match_regexp.groovy
+++ b/regression-test/suites/inverted_index_p0/test_index_match_regexp.groovy
@@ -85,6 +85,8 @@ suite("test_index_match_regexp", "p0"){
         qt_sql """ select count() from test_index_match_regexp where request 
match_regexp 's\$'; """
         qt_sql """ select count() from test_index_match_regexp where request 
match_regexp 'er\$'; """
         qt_sql """ select count() from test_index_match_regexp where request 
match_regexp '.*tickets.*'; """
+        qt_sql """ select count() from test_index_match_regexp where request 
match_regexp 'nonexistence'; """
+
     } finally {
         //try_sql("DROP TABLE IF EXISTS ${testTable}")
     }


---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@doris.apache.org
For additional commands, e-mail: commits-h...@doris.apache.org