From 3a2c8ab177b6819691d7868b8e24ef6027bd2299 Mon Sep 17 00:00:00 2001
From: David Peter <sharkdp@users.noreply.github.com>
Date: Mon, 5 Oct 2026 14:21:14 +0200
Subject: [PATCH] Unit safe statistics functions

---
 src/benchmark/measurement.rs | 43 +++++---------------
 src/benchmark/mod.rs         |  7 ++--
 src/quantity/mod.rs          | 78 ++++++++++++++++++++++++++++++++++--
 3 files changed, 88 insertions(+), 40 deletions(-)

diff --git a/src/benchmark/measurement.rs b/src/benchmark/measurement.rs
index 5906a24..4740eb4 100644
--- a/src/benchmark/measurement.rs
+++ b/src/benchmark/measurement.rs
@@ -64,21 +64,18 @@ impl Measurements {
         self.measurements.push(measurement);
     }
 
-    pub fn wall_clock_times(&self) -> Vec<Time> {
-        self.measurements
-            .iter()
-            .map(|m| m.time_wall_clock)
-            .collect()
+    pub fn wall_clock_times(&self) -> impl Iterator<Item = Time> + '_ {
+        self.measurements.iter().map(|m| m.time_wall_clock)
     }
 
     /// The average wall clock time
     pub fn time_wall_clock_mean(&self) -> Time {
-        mean(&self.wall_clock_times())
+        mean(self.wall_clock_times())
     }
 
     /// The standard deviation of all wall clock times. Not available if only one run has been performed
     pub fn stddev(&self) -> Option<Time> {
-        let times = self.wall_clock_times();
+        let times: Vec<_> = self.wall_clock_times().collect(); // TODO: Avoid collecting
 
         if times.len() < 2 {
             None
@@ -89,54 +86,36 @@ impl Measurements {
 
     /// The median wall clock time
     pub fn median(&self) -> Time {
-        median(&self.wall_clock_times())
+        median(&self.wall_clock_times().collect::<Vec<_>>())
     }
 
     /// The minimum wall clock time
     pub fn min(&self) -> Time {
-        min(&self.wall_clock_times())
+        min(self.wall_clock_times())
     }
 
     /// The maximum wall clock time
     pub fn max(&self) -> Time {
-        max(&self.wall_clock_times())
+        max(self.wall_clock_times())
     }
 
     /// Compute modified Z-scores for the wall clock times
     pub fn modified_zscores(&self) -> Vec<f64> {
-        modified_zscores(&self.wall_clock_times())
+        modified_zscores(&self.wall_clock_times().collect::<Vec<_>>())
     }
 
     /// The average user time
     pub fn time_user_mean(&self) -> Time {
-        mean(
-            &self
-                .measurements
-                .iter()
-                .map(|m| m.time_user)
-                .collect::<Vec<_>>(),
-        )
+        mean(self.measurements.iter().map(|m| m.time_user))
     }
 
     /// The average system time
     pub fn time_system_mean(&self) -> Time {
-        mean(
-            &self
-                .measurements
-                .iter()
-                .map(|m| m.time_system)
-                .collect::<Vec<_>>(),
-        )
+        mean(self.measurements.iter().map(|m| m.time_system))
     }
 
     pub fn peak_memory_usage_mean(&self) -> Information {
-        mean(
-            &self
-                .measurements
-                .iter()
-                .map(|m| m.peak_memory_usage)
-                .collect::<Vec<_>>(),
-        )
+        mean(self.measurements.iter().map(|m| m.peak_memory_usage))
     }
 
     pub fn peak_memory_usage_mean(&self) -> Information {
diff --git a/src/benchmark/mod.rs b/src/benchmark/mod.rs
index b0cd3aa..14bbbc3 100644
--- a/src/benchmark/mod.rs
+++ b/src/benchmark/mod.rs
@@ -22,7 +22,7 @@ use crate::output::progress_bar::{
 };
 use crate::output::warnings::{OutlierWarningOptions, Warnings};
 use crate::parameter::ParameterNameAndValue;
-use crate::quantity::{self, const_time_from_seconds, Time, Quantity};
+use crate::quantity::{self, const_time_from_seconds, Quantity, Time};
 use benchmark_result::BenchmarkResult;
 
 use anyhow::{anyhow, Result};
@@ -421,8 +421,7 @@ impl<'a> Benchmark<'a> {
         if matches!(self.options.executor_kind, ExecutorKind::Shell(_))
             && measurements
                 .wall_clock_times()
-                .iter()
-                .any(|&t| t < MIN_EXECUTION_TIME)
+                .any(|t| t < MIN_EXECUTION_TIME)
         {
             warnings.push(Warnings::FastExecutionTime);
         }
@@ -448,7 +447,7 @@ impl<'a> Benchmark<'a> {
 
         if scores[0] > OUTLIER_THRESHOLD {
             warnings.push(Warnings::SlowInitialRun(
-                measurements.wall_clock_times()[0],
+                measurements.wall_clock_times().next().unwrap(),
                 outlier_warning_options,
             ));
         } else if scores.iter().any(|&s| s.abs() > OUTLIER_THRESHOLD) {
diff --git a/src/quantity/mod.rs b/src/quantity/mod.rs
index cdac2e9..a2f9816 100644
--- a/src/quantity/mod.rs
+++ b/src/quantity/mod.rs
@@ -1,9 +1,12 @@
 use core::f64;
 use std::marker::PhantomData;
+use std::ops::AddAssign;
+use std::ops::Div;
 
 use serde::ser::SerializeStruct;
 use serde::Serializer;
 
+use uom::num_traits;
 use uom::si;
 pub use uom::si::information::{byte, gibibyte, kibibyte, mebibyte, tebibyte};
 pub use uom::si::ratio::ratio;
@@ -173,10 +176,40 @@ macro_rules! quantity_fn {
     };
 }
 
-quantity_fn!(mean, values, statistical::mean(&values));
+/// A max function that assumes no NaNs and at least one element
+pub fn max<V: PartialOrd>(values: impl IntoIterator<Item = V>) -> V {
+    values
+        .into_iter()
+        .max_by(|a, b| a.partial_cmp(b).expect("No NaN values"))
+        .expect("'max' requires at least one element")
+}
+
+/// A min function that assumes no NaNs and at least one element
+pub fn min<V: PartialOrd>(values: impl IntoIterator<Item = V>) -> V {
+    values
+        .into_iter()
+        .min_by(|a, b| a.partial_cmp(b).expect("No NaN values"))
+        .expect("'min' requires at least one element")
+}
+
+/// A mean function that assumes at least one element
+pub fn mean<Q, P>(values: impl IntoIterator<Item = Q>) -> Q
+where
+    Q: AddAssign + num_traits::Zero + Div<Ratio, Output = P>,
+    P: Into<Q>,
+{
+    let mut sum = Q::zero();
+    let mut count = 0;
+    for value in values {
+        sum += value;
+        count += 1;
+    }
+
+    let count = Ratio::new::<ratio>(count as f64);
+    (sum / count).into()
+}
+
 quantity_fn!(median, values, statistical::median(&values));
-quantity_fn!(min, values, crate::util::min_max::min(&values));
-quantity_fn!(max, values, crate::util::min_max::max(&values));
 quantity_fn!(standard_deviation, values, {
     let mean_value = statistical::mean(&values);
     statistical::standard_deviation(&values, Some(mean_value))
@@ -187,6 +220,15 @@ pub fn modified_zscores<Q: UnsafeRawValue>(values: &[Q]) -> Vec<f64> {
     crate::outlier_detection::modified_zscores(&values)
 }
 
+#[test]
+fn test_max() {
+    assert_eq!(1.0, max([1.0]));
+    assert_eq!(-1.0, max([-1.0]));
+    assert_eq!(-1.0, max([-2.0, -1.0]));
+    assert_eq!(1.0, max([-1.0, 1.0]));
+    assert_eq!(1.0, max([-1.0, 1.0, 0.0]));
+}
+
 #[test]
 fn test_time() {
     let time = Time::new::<millisecond>(123.4);
@@ -234,7 +276,7 @@ fn test_mean() {
         Time::new::<millisecond>(123.4),
         Time::new::<millisecond>(234.5),
     ];
-    let result = mean(&values);
+    let result = mean(values.into_iter());
     assert_eq!(result.format(TimeUnit::MilliSecond), "178.9 ms");
 }
 
@@ -271,3 +313,31 @@ fn test_format_duration_unit_with_unit() {
     let out = Time::new::<second>(1.3).format(TimeUnit::MicroSecond);
     assert_eq!("1300000.0 µs", out);
 }
+
+#[test]
+fn statistics() {
+    let values = vec![
+        Time::new::<second>(1.0),
+        Time::new::<second>(2.0),
+        Time::new::<second>(3.0),
+    ];
+
+    assert_eq!(
+        mean(values.iter().copied()).format(TimeUnit::Second),
+        "2.000 s"
+    );
+    assert_eq!(median(&values).format(TimeUnit::Second), "2.000 s");
+    assert_eq!(min(&values).format(TimeUnit::Second), "1.000 s");
+    assert_eq!(max(&values).format(TimeUnit::Second), "3.000 s");
+    assert_eq!(
+        standard_deviation(&values).format(TimeUnit::Second),
+        "1.000 s"
+    );
+
+    let values = vec![
+        Information::new::<byte>(1.0),
+        Information::new::<byte>(2.0),
+        Information::new::<byte>(3.0),
+    ];
+    mean(values.iter().copied()).format(InformationUnit::Byte);
+}