From 2da2f498a2365a6afff4f4d8fb7380edff947ffb Mon Sep 17 00:00:00 2001
From: Ana Maria Martinez Gomez <anamaria.martinezgom@mandiant.com>
Date: Tue, 9 Feb 2021 18:38:27 +0100
Subject: [PATCH] Add script to compare vivisect Python 2 vs 3

Compare the performance of vivisect Python 2 vs 3 by counting the number
of feature of each type extracted for every binary in `tests/data`.
Render the ones that perform bad (under a threshold - 98) and the total
performance. Render also the running time per binary for both Python 2 and 3.

From this result, it seems that vivisect behaves properly with Python3.
---
 scripts/vivisect-py2-vs-py3.sh | 69 ++++++++++++++++++++++++++++++++++
 1 file changed, 69 insertions(+)
 create mode 100755 scripts/vivisect-py2-vs-py3.sh

diff --git a/scripts/vivisect-py2-vs-py3.sh b/scripts/vivisect-py2-vs-py3.sh
new file mode 100755
index 00000000..20e7c817
--- /dev/null
+++ b/scripts/vivisect-py2-vs-py3.sh
@@ -0,0 +1,69 @@
+#!/usr/bin/env bash
+
+int() {
+  int=$(bc <<< "scale=0; ($1 + 0.5)/1")
+}
+
+export TIMEFORMAT='%3R'
+threshold_time=90
+threshold_py3_time=60 # Do not warn if it doesn't take at least 1 minute to run
+rm tests/data/*.viv 2>/dev/null
+mkdir results
+for file in tests/data/*
+do
+  file=$(printf %q "$file") # Handle names with white spaces
+  file_name=$(basename $file)
+  echo $file_name
+
+  rm "$file.viv" 2>/dev/null
+  py3_time=$(sh -c "time python3 scripts/show-features.py $file >> results/p3-$file_name.out 2>/dev/null" 2>&1)
+  rm "$file.viv" 2>/dev/null
+  py2_time=$(sh -c "time python2 scripts/show-features.py $file >> results/p2-$file_name.out 2>/dev/null" 2>&1)
+
+  int $py3_time
+  if (($int > $threshold_py3_time))
+  then
+    percentage=$(bc <<< "scale=3; $py2_time/$py3_time*100 + 0.5")
+    int $percentage
+    if (($int < $threshold_py3_time))
+    then
+      echo -n "  SLOWER ($percentage): "
+    fi
+  fi
+  echo "  PY2($py2_time) PY3($py3_time)"
+done
+
+threshold_features=98
+counter=0
+average=0
+results_for() {
+  py3=$(cat "results/p3-$file_name.out" | grep "$1" | wc -l)
+  py2=$(cat "results/p2-$file_name.out" | grep "$1" | wc -l)
+  if (($py2 > 0))
+  then
+    percentage=$(bc <<< "scale=2; 100*$py3/$py2")
+    average=$(bc <<< "scale=2; $percentage + $average")
+    count=$(($count + 1))
+    int $percentage
+    if (($int < $threshold_features))
+    then
+      echo -e "$1: py2($py2) py3($py3) $percentage% - $file_name"
+    fi
+  fi
+}
+
+rm tests/data/*.viv 2>/dev/null
+echo -e '\nRESULTS:'
+for file in tests/data/*
+do
+  file_name=$(basename $file)
+  if test -f "results/p2-$file_name.out"; then
+    results_for 'insn'
+    results_for 'file'
+    results_for 'func'
+    results_for 'bb'
+  fi
+done
+
+average=$(bc <<< "scale=2; $average/$count")
+echo "TOTAL: $average"