Optimize out N*M suffix matching in Builder.py.

1 files changed, 1 insertions, 1 deletions

diff --git a/bin/objcounts.py b/bin/objcounts.py
index 1b72837a..db02aab4 100644
--- a/bin/objcounts.py
+++ b/bin/objcounts.py
@@ -35,7 +35,7 @@ Compare the --debug=object counts from two build logs.
 
 def fetch_counts(fname):
     contents = open(fname).read()
-    m = re.search('\nObject counts:\n(.*)\n[^\s]', contents, re.S)
+    m = re.search('\nObject counts:(\n\s[^\n]*)*', contents, re.S)
     lines = m.group().split('\n')
     list = [l.split() for l in lines if re.match('\s+\d', l)]
     d = {}