[WIP] fix: respect inexact flags in row group metadata

Cargo.toml

@@ -88,19 +88,19 @@ ahash = { version = "0.8", default-features = false, features = [
     "runtime-rng",
 ] }
 apache-avro = { version = "0.17", default-features = false }
-arrow = { version = "55.1.0", features = [
+arrow = { git = "https://github.com/apache/arrow-rs.git", rev = "1029974bc0f03b8adb089bfa8cc8f0ee96701866", features = [
     "prettyprint",
     "chrono-tz",
 ] }
-arrow-buffer = { version = "55.0.0", default-features = false }
-arrow-flight = { version = "55.1.0", features = [
+arrow-buffer = { git = "https://github.com/apache/arrow-rs.git", rev = "1029974bc0f03b8adb089bfa8cc8f0ee96701866", default-features = false }
+arrow-flight = { git = "https://github.com/apache/arrow-rs.git", rev = "1029974bc0f03b8adb089bfa8cc8f0ee96701866", features = [
     "flight-sql-experimental",
 ] }
-arrow-ipc = { version = "55.0.0", default-features = false, features = [
+arrow-ipc = { git = "https://github.com/apache/arrow-rs.git", rev = "1029974bc0f03b8adb089bfa8cc8f0ee96701866", default-features = false, features = [
     "lz4",
 ] }
-arrow-ord = { version = "55.0.0", default-features = false }
-arrow-schema = { version = "55.0.0", default-features = false }
+arrow-ord = { git = "https://github.com/apache/arrow-rs.git", rev = "1029974bc0f03b8adb089bfa8cc8f0ee96701866", default-features = false }
+arrow-schema = { git = "https://github.com/apache/arrow-rs.git", rev = "1029974bc0f03b8adb089bfa8cc8f0ee96701866", default-features = false }
 async-trait = "0.1.88"
 bigdecimal = "0.4.8"
 bytes = "1.10"
@@ -151,7 +151,7 @@ itertools = "0.14"
 log = "^0.4"
 object_store = { version = "0.12.0", default-features = false }
 parking_lot = "0.12"
-parquet = { version = "55.1.0", default-features = false, features = [
+parquet = { git = "https://github.com/apache/arrow-rs.git", rev = "1029974bc0f03b8adb089bfa8cc8f0ee96701866", default-features = false, features = [
     "arrow",
     "async",
     "object_store",

datafusion/core/tests/parquet/file_statistics.rs

@@ -28,7 +28,7 @@ use datafusion::execution::context::SessionState;
 use datafusion::execution::session_state::SessionStateBuilder;
 use datafusion::prelude::SessionContext;
 use datafusion_common::stats::Precision;
-use datafusion_common::DFSchema;
+use datafusion_common::{DFSchema, ScalarValue};
 use datafusion_execution::cache::cache_manager::CacheManagerConfig;
 use datafusion_execution::cache::cache_unit::{
     DefaultFileStatisticsCache, DefaultListFilesCache,
@@ -99,6 +99,35 @@ async fn check_stats_precision_with_filter_pushdown() {
     );
 }
 
+#[tokio::test]
+async fn check_stats_inexact_for_truncated_values() {
+    let filename = "long_strings.parquet";
+    let table_path = ListingTableUrl::parse(filename).unwrap();
+
+    let opt =
+        ListingOptions::new(Arc::new(ParquetFormat::default())).with_collect_stat(true);
+    let table = get_listing_table(&table_path, None, &opt).await;
+
+    let (_, _, state) = get_cache_runtime_state();
+
+    let exec = table.scan(&state, None, &[], None).await.unwrap();
+    assert_eq!(
+        exec.partition_statistics(None).unwrap().num_rows,
+        Precision::Exact(2),
+        "Stats without filter should be exact"
+    );
+
+    let col_stats = &exec.partition_statistics(None).unwrap().column_statistics[0];
+    assert_eq!(
+        col_stats.min_value,
+        Precision::Inexact(ScalarValue::Utf8View(Some("A".repeat(4096))))
+    );
+    assert_eq!(
+        col_stats.max_value,
+        Precision::Inexact(ScalarValue::Utf8View(Some("Z".repeat(4095) + "[")))
+    );
+}
+
 #[tokio::test]
 async fn load_table_stats_with_session_level_cache() {
     let testdata = datafusion::test_util::parquet_test_data();