Make check_output of subprocess2 compatible with Python's subprocess.

subprocess2.py

Index: subprocess2.py
diff --git a/subprocess2.py b/subprocess2.py
index 9f547a63bbb84939d9ff29010abee0a99e23cb7d..16c2aac770921479efe4dd74a705ff6b6578ce78 100644
--- a/subprocess2.py
+++ b/subprocess2.py
@@ -33,8 +33,9 @@ SUBPROCESS_CLEANUP_HACKED = False
 
 class CalledProcessError(subprocess.CalledProcessError):
   """Augment the standard exception with more data."""
-  def __init__(self, returncode, cmd, cwd, stdout, stderr):
-    super(CalledProcessError, self).__init__(returncode, cmd)
+  def __init__(self, returncode, cmd, cwd, stdout, stderr, output=None):
+    # output kwarg is for compatability with python's subprocess.check_output
+    super(CalledProcessError, self).__init__(returncode, cmd, output=output)

[iannucci, 2014/09/18 20:59:03]

why not just output=stdout

[tandrii(chromium), 2014/09/18 22:23:05]

Done.

     self.stdout = stdout

[iannucci, 2014/09/18 20:59:03]

and then say `self.stdout = self.output` for compatibility

[tandrii(chromium), 2014/09/18 22:23:05]

Done.

     self.stderr = stderr
     self.cwd = cwd
@@ -475,7 +476,7 @@ def check_call_out(args, **kwargs):
   out, returncode = communicate(args, **kwargs)
   if returncode:
     raise CalledProcessError(
-        returncode, args, kwargs.get('cwd'), out[0], out[1])
+        returncode, args, kwargs.get('cwd'), out[0], out[1], output=out[0])
   return out