From a938c87fa43be51bfa54c49844e3555438df0fd7 Mon Sep 17 00:00:00 2001
From: Willi Ballenthin <wballenthin@hex-rays.com>
Date: Wed, 22 Apr 2026 22:21:24 +0300
Subject: [PATCH] fix: guard statistics calls in compare-backends.py against
 empty duration lists

When all runs for a backend fail, durations_by_backend[backend] is empty,
causing StatisticsError from statistics.quantiles (needs >= 2 points) and
statistics.mean (needs >= 1 point). Print placeholder messages instead.
---
 CHANGELOG.md                |  1 +
 scripts/compare-backends.py | 56 +++++++++++++++++++++++++------------
 2 files changed, 39 insertions(+), 18 deletions(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index e07a3544..36c82d80 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -49,6 +49,7 @@
 - fix: Scopes.from_dict uses cls instead of self so subclasses return the correct type @williballenthin
 - fix: correct wrong dict key in VMRay _compute_monitor_threads assertion (used thread_id instead of process_id) @williballenthin
 - fix: replace assert with isinstance guard in get_callee for invalid MethodSpec tokens @williballenthin
+- fix: guard statistics.quantiles/mean in compare-backends.py report() against empty duration lists @williballenthin (SURF-89)
 - fix: replace zipfile with pyzipper in minimize_vmray_results.py so output archive is AES-encrypted @williballenthin (SURF-88)
 - fix: assign yara_strings/yara_condition to empty string when Some has cmin=0 to prevent UnboundLocalError @williballenthin (SURF-87)
 - fix: parenthesize s_type checks in capa2yara.py so kid.name != "Some" guard applies to And/Or/Not uniformly @williballenthin (SURF-86)
diff --git a/scripts/compare-backends.py b/scripts/compare-backends.py
index 4120919a..d338ec3e 100644
--- a/scripts/compare-backends.py
+++ b/scripts/compare-backends.py
@@ -252,33 +252,53 @@ def report(args):
     console.print("durations:", style="bold")
     console.print("  (10-quantiles, in seconds)", style="grey37")
     for backend in BACKENDS:
-        q = statistics.quantiles(durations_by_backend[backend], n=10)
-        console.print(f"  {backend: <8}: ", end="")
-        for i in range(9):
-            if i in (4, 8):
-                style = "bold"
-            else:
-                style = "default"
-            console.print(f"{q[i]: >6.1f}", style=style, end=" ")
-        console.print()
-    console.print("                ^-- 10% of samples took less than this                  ^", style="grey37")
-    console.print("                    10% of samples took more than this -----------------+", style="grey37")
+        durations = durations_by_backend[backend]
+        if len(durations) >= 2:
+            q = statistics.quantiles(durations, n=10)
+            console.print(f"  {backend: <8}: ", end="")
+            for i in range(9):
+                if i in (4, 8):
+                    style = "bold"
+                else:
+                    style = "default"
+                console.print(f"{q[i]: >6.1f}", style=style, end=" ")
+            console.print()
+        else:
+            console.print(f"  {backend: <8}: (no data)")
+    console.print(
+        "                ^-- 10% of samples took less than this                  ^",
+        style="grey37",
+    )
+    console.print(
+        "                    10% of samples took more than this -----------------+",
+        style="grey37",
+    )
 
     console.print()
     for backend in BACKENDS:
-        total = sum(durations_by_backend[backend])
-        successes = len(durations_by_backend[backend])
-        avg = statistics.mean(durations_by_backend[backend])
-        console.print(
-            f"  {backend: <8}: {total: >7.0f} seconds across {successes: >4d} successful runs, {avg: >4.1f} average"
-        )
+        durations = durations_by_backend[backend]
+        if durations:
+            total = sum(durations)
+            avg = statistics.mean(durations)
+            console.print(
+                f"  {backend: <8}: {total: >7.0f} seconds across {len(durations): >4d} successful runs, {avg: >4.1f} average"
+            )
+        else:
+            console.print(f"  {backend: <8}: (no successful runs)")
     console.print()
 
     console.print("slowest samples:", style="bold")
     for backend in BACKENDS:
+        durations = durations_by_backend[backend]
+        if not durations:
+            console.print(f"  {backend: <8}: (no successful runs)")
+            continue
+
         console.print(backend)
+        successful_keys = {k for k, v in doc[backend].items() if not v["err"]}
         for duration, path in sorted(
-            ((d["duration"], Path(d["path"]).name) for d in doc[backend].values()), reverse=True
+            ((d["duration"], Path(d["path"]).name) for k, d in doc[backend].items() if k in successful_keys),
+            reverse=True,
         )[:5]:
             console.print(f"  - {duration: >6.1f} {path}")