[arrow] branch main updated: GH-37110: [C++] Expression: SmallestTypeFor lost tz for Scalar (#37135)

[apitrou]

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

apitrou pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/arrow.git


The following commit(s) were added to refs/heads/main by this push:
     new baf3eb9d62 GH-37110: [C++] Expression: SmallestTypeFor lost tz for 
Scalar (#37135)
baf3eb9d62 is described below

commit baf3eb9d62bea7b76c2d18f40a79a1fbe13a860e
Author: mwish <maplewish...@gmail.com>
AuthorDate: Mon Aug 14 23:44:44 2023 +0800

    GH-37110: [C++] Expression: SmallestTypeFor lost tz for Scalar (#37135)
    
    
    
    ### Rationale for this change
    
    This patch ( https://github.com/apache/arrow/pull/15180 ) adds a 
`SmallestTypeFor` to handling expression type. However, it lost timezone when 
handling.
    
    ### What changes are included in this PR?
    
    Add `timezone` in `SmallestTypeFor`
    
    ### Are these changes tested?
    
    Currently not
    
    ### Are there any user-facing changes?
    
    Yeah it's a bugfix
    
    * Closes: #37110
    
    Authored-by: mwish <maplewish...@gmail.com>
    Signed-off-by: Antoine Pitrou <anto...@python.org>
---
 cpp/src/arrow/compute/expression.cc      | 12 ++++++------
 cpp/src/arrow/compute/expression_test.cc | 13 +++++++++++++
 2 files changed, 19 insertions(+), 6 deletions(-)

diff --git a/cpp/src/arrow/compute/expression.cc 
b/cpp/src/arrow/compute/expression.cc
index 247f4e5408..b47e0a3552 100644
--- a/cpp/src/arrow/compute/expression.cc
+++ b/cpp/src/arrow/compute/expression.cc
@@ -480,26 +480,26 @@ TypeHolder SmallestTypeFor(const arrow::Datum& value) {
           return value.type();
         case TimeUnit::MILLI:
           if (ts % 1000 == 0) {
-            return timestamp(TimeUnit::SECOND);
+            return timestamp(TimeUnit::SECOND, ts_type->timezone());
           }
           return value.type();
         case TimeUnit::MICRO:
           if (ts % 1000000 == 0) {
-            return timestamp(TimeUnit::SECOND);
+            return timestamp(TimeUnit::SECOND, ts_type->timezone());
           }
           if (ts % 1000 == 0) {
-            return timestamp(TimeUnit::MILLI);
+            return timestamp(TimeUnit::MILLI, ts_type->timezone());
           }
           return value.type();
         case TimeUnit::NANO:
           if (ts % 1000000000 == 0) {
-            return timestamp(TimeUnit::SECOND);
+            return timestamp(TimeUnit::SECOND, ts_type->timezone());
           }
           if (ts % 1000000 == 0) {
-            return timestamp(TimeUnit::MILLI);
+            return timestamp(TimeUnit::MILLI, ts_type->timezone());
           }
           if (ts % 1000 == 0) {
-            return timestamp(TimeUnit::MICRO);
+            return timestamp(TimeUnit::MICRO, ts_type->timezone());
           }
           return value.type();
         default:
diff --git a/cpp/src/arrow/compute/expression_test.cc 
b/cpp/src/arrow/compute/expression_test.cc
index f90e01a2f8..b852f6f6b0 100644
--- a/cpp/src/arrow/compute/expression_test.cc
+++ b/cpp/src/arrow/compute/expression_test.cc
@@ -63,6 +63,7 @@ const std::shared_ptr<Schema> kBoringSchema = schema({
     field("ts_ns", timestamp(TimeUnit::NANO)),
     field("ts_s", timestamp(TimeUnit::SECOND)),
     field("binary", binary()),
+    field("ts_s_utc", timestamp(TimeUnit::SECOND, "UTC")),
 });
 
 #define EXPECT_OK ARROW_EXPECT_OK
@@ -630,6 +631,18 @@ TEST(Expression, BindWithImplicitCasts) {
                       literal(std::make_shared<TimestampScalar>(0, 
TimeUnit::NANO))),
                   cmp(field_ref("ts_s"),
                       literal(std::make_shared<TimestampScalar>(0, 
TimeUnit::SECOND))));
+    // GH-37110
+    ExpectBindsTo(
+        cmp(field_ref("ts_s_utc"),
+            literal(std::make_shared<TimestampScalar>(0, TimeUnit::NANO, 
"UTC"))),
+        cmp(field_ref("ts_s_utc"),
+            literal(std::make_shared<TimestampScalar>(0, TimeUnit::SECOND, 
"UTC"))));
+    ExpectBindsTo(
+        cmp(field_ref("ts_s_utc"),
+            literal(std::make_shared<TimestampScalar>(123000, TimeUnit::NANO, 
"UTC"))),
+        cmp(field_ref("ts_s_utc"),
+            literal(std::make_shared<TimestampScalar>(123, TimeUnit::MICRO, 
"UTC"))));
+
     ExpectBindsTo(
         cmp(field_ref("binary"), 
literal(std::make_shared<LargeBinaryScalar>("foo"))),
         cmp(field_ref("binary"), 
literal(std::make_shared<BinaryScalar>("foo"))));