The HTML number conversion regex operating against a byte string will only

convert individual bytes, so therefore incorrectly converts utf-8 encoded
characters. Decoding byte strings to unicode objects results in correct HTML
number encodings.

lib/spack/spack/cmd/test-install.py

@@ -27,6 +27,7 @@
 import itertools
 import re
 import os
+import codecs
 
 import llnl.util.tty as tty
 from llnl.util.filesystem import *
@@ -107,7 +108,7 @@ def __eq__(self, other):
 def fetch_log(path):
     if not os.path.exists(path):
         return list()
-    with open(path, 'rb') as F:
+    with codecs.open(path, 'rb', 'utf-8') as F:
         return list(line.strip() for line in F.readlines())