(arrow-rs) branch main updated: remove T: ParquetValueType bound on ValueStatistics (#8824)

[alamb]

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

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


The following commit(s) were added to refs/heads/main by this push:
     new 9d7dde2722 remove T: ParquetValueType bound on ValueStatistics (#8824)
9d7dde2722 is described below

commit 9d7dde2722a8392996a1f5a978e989b2fea4ed0e
Author: Patrick Marks <pma...@gmail.com>
AuthorDate: Fri Nov 14 18:12:06 2025 +0100

    remove T: ParquetValueType bound on ValueStatistics (#8824)
    
    The old bound was T: ParquetValueType which isn't public, so it was
    impossible to call the methods on ValueStatistics from a generic
    function in another crate, forcing the use of macros. Split out a few
    functions that require T: AsBytes into a separate impl.
    
    # Which issue does this PR close?
    Closes #8823
    
    # Rationale for this change
    See issue for more details
    
    # Are these changes tested?
    Added a test that call the `ValueStatistics::min_opt` method from a
    generic function without a bound to a private trait.
    
    # Are there any user-facing changes?
    The methods were already public and documented, they just couldn't
    actually be called from external crates.
    
    ---------
    
    Co-authored-by: Ed Seidl <etse...@live.com>
---
 parquet/src/file/statistics.rs | 46 ++++++++++++++++++++++++++++++++----------
 1 file changed, 35 insertions(+), 11 deletions(-)

diff --git a/parquet/src/file/statistics.rs b/parquet/src/file/statistics.rs
index 889e5ea66b..a813e82d13 100644
--- a/parquet/src/file/statistics.rs
+++ b/parquet/src/file/statistics.rs
@@ -515,7 +515,7 @@ pub struct ValueStatistics<T> {
     is_min_max_backwards_compatible: bool,
 }
 
-impl<T: ParquetValueType> ValueStatistics<T> {
+impl<T> ValueStatistics<T> {
     /// Creates new typed statistics.
     pub fn new(
         min: Option<T>,
@@ -580,16 +580,6 @@ impl<T: ParquetValueType> ValueStatistics<T> {
         self.max.as_ref()
     }
 
-    /// Returns min value as bytes of the statistics, if min value is known.
-    pub fn min_bytes_opt(&self) -> Option<&[u8]> {
-        self.min_opt().map(AsBytes::as_bytes)
-    }
-
-    /// Returns max value as bytes of the statistics, if max value is known.
-    pub fn max_bytes_opt(&self) -> Option<&[u8]> {
-        self.max_opt().map(AsBytes::as_bytes)
-    }
-
     /// Whether or not min and max values are set.
     /// Normally both min/max values will be set to `Some(value)` or `None`.
     pub(crate) fn _internal_has_min_max_set(&self) -> bool {
@@ -636,6 +626,18 @@ impl<T: ParquetValueType> ValueStatistics<T> {
     }
 }
 
+impl<T: AsBytes> ValueStatistics<T> {
+    /// Returns min value as bytes of the statistics, if min value is known.
+    pub fn min_bytes_opt(&self) -> Option<&[u8]> {
+        self.min_opt().map(AsBytes::as_bytes)
+    }
+
+    /// Returns max value as bytes of the statistics, if max value is known.
+    pub fn max_bytes_opt(&self) -> Option<&[u8]> {
+        self.max_opt().map(AsBytes::as_bytes)
+    }
+}
+
 impl<T: ParquetValueType> fmt::Display for ValueStatistics<T> {
     fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
         write!(f, "{{")?;
@@ -1121,4 +1123,26 @@ mod tests {
             "Parquet error: Incorrect Int96 max statistics"
         );
     }
+
+    // Ensures that we can call ValueStatistics::min_opt from a
+    // generic function without reyling on a bound to a private trait.
+    fn generic_statistics_handler<T: std::fmt::Display>(stats: 
ValueStatistics<T>) -> String {
+        match stats.min_opt() {
+            Some(s) => format!("min: {}", s),
+            None => "min: NA".to_string(),
+        }
+    }
+
+    #[test]
+    fn test_generic_access() {
+        let stats = Statistics::int32(Some(12), Some(45), None, Some(11), 
false);
+
+        match stats {
+            Statistics::Int32(v) => {
+                let stats_string = generic_statistics_handler(v);
+                assert_eq!(&stats_string, "min: 12");
+            }
+            _ => unreachable!(),
+        }
+    }
 }