skpbench: use accumulative result as the default

tools/skpbench/parseskpbench.py

 #!/usr/bin/env python
 
 # Copyright 2016 Google Inc.
 #
 # Use of this source code is governed by a BSD-style license that can be
 # found in the LICENSE file.
 
 from __future__ import print_function
 from _benchresult import BenchResult
 from argparse import ArgumentParser
@@ skipping 16 matching lines @@
 (1) Install the "Office Editing for Docs, Sheets & Slides" Chrome extension:
     https://chrome.google.com/webstore/detail/office-editing-for-docs-s/gbkeegbaiigmenfmjfclcdgdpimamgkj
 
 (2) Designate Chrome os-wide as the default application for opening .csv files.
 
 (3) Run parseskpbench.py with the --open flag.
 
 """)
 
 __argparse.add_argument('-r', '--result',
-    choices=['median', 'accum', 'max', 'min'], default='median',
+    choices=['accum', 'median', 'max', 'min'], default='accum',
     help="result to use for cell values")
 __argparse.add_argument('-f', '--force',
     action='store_true', help='silently ignore warnings')
 __argparse.add_argument('-o', '--open',
     action='store_true',
     help="generate a temp file and open it (theoretically in a web browser)")
 __argparse.add_argument('-n', '--name',
     default='skpbench_%s' % datetime.now().strftime('%Y-%m-%d_%H.%M.%S.csv'),
     help="if using --open, a name for the temp file")
 __argparse.add_argument('sources',
@@ skipping 92 matching lines @@
     pathname = os.path.join(dirname, basename)
     with open(pathname, mode='w') as tmpfile:
       parser.print_csv(outfile=tmpfile)
     fileuri = urlparse.urljoin('file:', urllib.pathname2url(pathname))
     print('opening %s' % fileuri)
     webbrowser.open(fileuri)
 
 
 if __name__ == '__main__':
   main()