From a746054456be40daae6b0111f49d9e5ec20ddf4d Mon Sep 17 00:00:00 2001
From: Pxl <xl@selectdb.com>
Date: Fri, 7 Feb 2025 18:57:04 +0800
Subject: [PATCH] [Bug](function) fix wrong result when percentile's second
 argument is 1 (#47586)

### What problem does this PR solve?
fix wrong result when percentile's second argument is 1
related with https://github.com/apache/doris/pull/34382
### Check List (For Author)

- Test <!-- At least one of them must be included. -->
    - [x] Regression test
    - [ ] Unit Test
    - [ ] Manual test (add detailed scripts or steps below)
    - [ ] No need to test or manual test. Explain why:
- [ ] This is a refactor/code format and no logic has been changed.
        - [ ] Previous test can cover this change.
        - [ ] No code files have been changed.
        - [ ] Other reason <!-- Add your reason?  -->

- Behavior changed:
    - [x] No.
    - [ ] Yes. <!-- Explain the behavior change -->

- Does this need documentation?
    - [x] No.
- [ ] Yes. <!-- Add document PR link here. eg:
https://github.com/apache/doris-website/pull/1214 -->

### Check List (For Reviewer who merge this PR)

- [ ] Confirm the release note
- [ ] Confirm test cases
- [ ] Confirm document
- [ ] Add branch pick label <!-- Add branch pick label that this PR
should merge into -->
---
 be/src/util/counts.h                          |  8 +++++---
 .../test_aggregate_percentile_no_cast.out     |  9 +++++++++
 .../test_aggregate_percentile_no_cast.groovy  | 19 +++++++++++++++++++
 3 files changed, 33 insertions(+), 3 deletions(-)

diff --git a/be/src/util/counts.h b/be/src/util/counts.h
index 9f45eb4d42633a..43b815cf8cabe0 100644
--- a/be/src/util/counts.h
+++ b/be/src/util/counts.h
@@ -87,13 +87,15 @@ class Counts {
                 // get val in aggregate_function_percentile_approx.h
                 return 0.0;
             }
-            if (quantile == 1 || _nums.size() == 1) {
-                return _nums.back();
-            }
+
             if (UNLIKELY(!std::is_sorted(_nums.begin(), _nums.end()))) {
                 pdqsort(_nums.begin(), _nums.end());
             }
 
+            if (quantile == 1 || _nums.size() == 1) {
+                return _nums.back();
+            }
+
             double u = (_nums.size() - 1) * quantile;
             auto index = static_cast<uint32_t>(u);
             return _nums[index] +
diff --git a/regression-test/data/nereids_p0/sql_functions/aggregate_functions/test_aggregate_percentile_no_cast.out b/regression-test/data/nereids_p0/sql_functions/aggregate_functions/test_aggregate_percentile_no_cast.out
index 296589ed425c72..53cbce2c30bb85 100644
--- a/regression-test/data/nereids_p0/sql_functions/aggregate_functions/test_aggregate_percentile_no_cast.out
+++ b/regression-test/data/nereids_p0/sql_functions/aggregate_functions/test_aggregate_percentile_no_cast.out
@@ -37,3 +37,12 @@
 -- !select --
 [19.5, 24.25, 101.325]
 
+-- !test --
+-128	127.0	126.93099999999995	-128.0	-127.9238
+-1	127.0	126.94250000000004	-1.0	-0.9995
+0	127.0	126.95400000000001	0.0	4.0E-4
+1	127.0	126.96549999999998	1.0	1.0003
+2	127.0	126.977	2.0	2.002
+12	127.0	126.9885	12.0	12.0115
+127	127.0	127.0	127.0	127.0
+
diff --git a/regression-test/suites/nereids_p0/sql_functions/aggregate_functions/test_aggregate_percentile_no_cast.groovy b/regression-test/suites/nereids_p0/sql_functions/aggregate_functions/test_aggregate_percentile_no_cast.groovy
index a7f382abe2175e..1a9a67ebbafd65 100644
--- a/regression-test/suites/nereids_p0/sql_functions/aggregate_functions/test_aggregate_percentile_no_cast.groovy
+++ b/regression-test/suites/nereids_p0/sql_functions/aggregate_functions/test_aggregate_percentile_no_cast.groovy
@@ -111,4 +111,23 @@ suite("test_aggregate_percentile_no_cast") {
     }
     sql "INSERT INTO percentile_test_db2 values(1,10.1), (2,8.2), (2,114.3) ,(3,10.4) ,(5,29.5) ,(6,101.6)"
     qt_select "select percentile_array(level,[0.5,0.55,0.805])from percentile_test_db2;"
+
+    sql "DROP TABLE IF EXISTS TINYINTDATA_NOT_EMPTY_NOT_NULLABLE"
+    sql """
+    CREATE TABLE TINYINTDATA_NOT_EMPTY_NOT_NULLABLE(id INT, data TINYINT  NOT NULL) DISTRIBUTED BY HASH(id) BUCKETS 1 PROPERTIES ('replication_num' = '1');
+    """
+    sql "DROP TABLE IF EXISTS TEMPDATA"
+    sql """
+    CREATE TABLE IF NOT EXISTS TEMPDATA(id INT, data INT) DISTRIBUTED BY HASH(id) BUCKETS 1 PROPERTIES ('replication_num' = '1');
+    """
+    sql """
+    INSERT INTO TINYINTDATA_NOT_EMPTY_NOT_NULLABLE values (0, -1),(1, 0),(2, 1),(3, 2),(4, 12),(5, -128),(6, 127);
+    """
+    sql """
+    INSERT INTO TEMPDATA values(1, 1); 
+    """
+    qt_test """
+    SELECT ARG0,PERCENTILE(NULLABLE(ARG0),1) OVER(ORDER BY t.ARG0 ROWS BETWEEN  CURRENT ROW  AND  UNBOUNDED FOLLOWING ) as a,PERCENTILE(NULLABLE(ARG0),0.9999) OVER(ORDER BY t.ARG0 ROWS BETWEEN  CURRENT ROW  AND  UNBOUNDED FOLLOWING ) as b,PERCENTILE(NULLABLE(ARG0),0) OVER(ORDER BY t.ARG0 ROWS BETWEEN  CURRENT ROW  AND  UNBOUNDED FOLLOWING ) as c,PERCENTILE(NULLABLE(ARG0),0.0001) OVER(ORDER BY t.ARG0 ROWS BETWEEN  CURRENT ROW  AND  UNBOUNDED FOLLOWING ) as d FROM (SELECT TEMPDATA . data, TABLE0.ARG0 FROM TEMPDATA CROSS JOIN (SELECT data AS ARG0
+ FROM TINYINTDATA_NOT_EMPTY_NOT_NULLABLE ) AS TABLE0) t  GROUP BY ARG0 order by ARG0;
+    """
 }