Add script to verify that image files exist for every actual_result checksum

tools/verify_images_for_gm_results.py

Index: tools/verify_images_for_gm_results.py
===================================================================
--- tools/verify_images_for_gm_results.py	(revision 0)
+++ tools/verify_images_for_gm_results.py	(revision 0)
@@ -0,0 +1,110 @@
+#!/usr/bin/python
+
+# Copyright (c) 2013 The Chromium Authors. All rights reserved.
+# Use of this source code is governed by a BSD-style license that can be
+# found in the LICENSE file.
+
+
+""" find_missing_images: Look through skia-autogen, searching for all checksums

[epoger, 2013/07/10 19:30:44]

Delete no-longer-accurate script name

[borenet, 2013/07/10 20:36:45]

Done.

+which should have corresponding files in Google Storage, and verify that those
+files exist. """
+
+
+import json
+import posixpath
+import subprocess
+import sys
+
+
+AUTOGEN_URL = 'http://skia-autogen.googlecode.com/svn/gm-actual'

[epoger, 2013/07/10 19:30:44]

optional: use argparse with defaults (like
https://code.google.com/p/skia/source/browse/trunk/tools/rebaseline.py ) to make
these overridable at the command line

[borenet, 2013/07/10 20:36:45]

Mind if I save this for the future?  I'd like to just move on.

[epoger, 2013/07/11 14:14:25]

No problem.

+GS_URL = 'gs://chromium-skia-gm/gm/'
+
+
+def FileNameToResultType(filename):

[epoger, 2013/07/10 19:30:44]

please add a brief description of each function

[borenet, 2013/07/10 20:36:45]

Done.

+  result = filename
+  for suffix in ['.png', '_8888', '_565', '_gpu', '_gpudebug', '_msaa4']:

[epoger, 2013/07/10 19:30:44]

I don't understand why we need this list of suffixes (which seems fragile).  I
think it makes more sense to build the gs_url like we do in
https://code.google.com/p/skia/source/browse/trunk/tools/rebaseline_imagefile...
, and just build up a list of complete gs_urls as we go (rather than
"checksums", which are actually partial GS paths in this case)

[borenet, 2013/07/10 20:36:45]

Done.

+    if result.endswith(suffix):
+      result = result[:len(result) - len(suffix)]
+  return result
+
+
+def FindChecksumsInJSON(json_file):
+  print 'svn cat %s' % json_file
+  output = subprocess.check_output(['svn', 'cat', json_file])
+  json_content = json.loads(output)
+  checksums = []
+  for dict_type in ['actual-results']:
+    for result_type in json_content[dict_type]:
+      if json_content[dict_type][result_type]:
+        for result in json_content[dict_type][result_type].keys():
+          hash_type, hash = json_content[dict_type][result_type][result]
+          checksum = posixpath.join(hash_type,
+                                    FileNameToResultType(result),
+                                    str(hash))
+          checksums.append(checksum)
+  return checksums
+
+
+def _FindJSONFiles(url, json_files):
+  proc = subprocess.Popen(['svn', 'ls', url], stdout=subprocess.PIPE,
+                          stderr=subprocess.STDOUT)
+  if proc.wait() != 0:
+    raise Exception('Failed to list svn directory.')
+  output = proc.communicate()[0].splitlines()
+  subdirs = []
+  for item in output:
+    if item.endswith(posixpath.sep):
+      subdirs.append(item)
+    elif item.endswith('.json'):
+      json_files.append(posixpath.join(url, item))
+    else:
+      print 'Warning: ignoring %s' % posixpath.join(url, item)
+  for subdir in subdirs:
+    _FindJSONFiles(posixpath.join(url, subdir), json_files)
+
+
+def FindJSONFiles(url):
+  print 'Searching for json files in %s' % url
+  json_files = []
+  _FindJSONFiles(url, json_files)
+  return json_files
+
+
+def FindChecksums(url):
+  checksums = set()
+  for json_file in FindJSONFiles(url):
+    print 'Looking for checksums in %s' % json_file
+    checksums.update(FindChecksumsInJSON(json_file))
+  return checksums
+
+
+def VerifyChecksum(checksum):
+  image_url = posixpath.join(GS_URL, checksum + '.png')
+  proc = subprocess.Popen(['gsutil', 'ls', image_url], stdout=subprocess.PIPE,
+                          stderr=subprocess.STDOUT)
+  if proc.wait() != 0:
+    return False
+  return True
+
+
+def VerifyChecksums(checksums):
+  print 'Verifying that images exist for checksums...'
+  missing = []
+  for checksum in checksums:
+    if not VerifyChecksum(checksum):
+      print 'Missing: %s' % checksum
+      missing.append(checksum)
+  return missing
+
+
+def Main():
+  checksums = FindChecksums(AUTOGEN_URL)
+  missing = VerifyChecksums(checksums)
+  if missing:
+    print 'Missing files:'

[epoger, 2013/07/10 19:30:44]

please print the number of missing files

[borenet, 2013/07/10 20:36:45]

Done.

+    print '\n'.join(missing)
+    return 1
+
+
+if __name__ == '__main__':
+  sys.exit(Main())