Add more informtion to the CSV generated

third_party/WebKit/Tools/Scripts/print-stale-test-expectations-entries

 #!/usr/bin/python
 #
 # Copyright (C) 2013 Google Inc. All rights reserved.
 #
 # Redistribution and use in source and binary forms, with or without
 # modification, are permitted provided that the following conditions are
 # met:
 #
 #     * Redistributions of source code must retain the above copyright
 # notice, this list of conditions and the following disclaimer.
@@ skipping 24 matching lines @@
 import optparse
 import sys
 import urllib2
 
 from webkitpy.common.host import Host
 from webkitpy.layout_tests.models.test_expectations import TestExpectationParser
 
 # FIXME: Make this a direct request to Monorail.
 GOOGLE_CODE_URL = 'https://www.googleapis.com/projecthosting/v2/projects/chromium/issues/%s?key=AIzaSyDgCqT1Dt5AZWLHo4QJjyMHaCjhnFacGF0'
 CRBUG_PREFIX = 'crbug.com/'
-
+CSV_ROW_HEADERS = 'crbug link, test file, days since last updated, status\n'

[qyearsley, 2016/08/19 00:37:18]

This would also make sense as a list of strings:

CSV_ROW_HEADERS = ['crbug link', 'test file', 'days since last updated',
'status']

[nainar, 2016/08/19 03:36:48]

Done.

 
 class StaleTestPrinter(object):
 
     def __init__(self, options):
         self.days = options.days
         self.csv_filename = options.create_csv
         self.host = Host()
         self.is_stale_results = {}
+        self.last_updated = {}
+        self.state = {}
 
     def print_stale_tests(self):
         port = self.host.port_factory.get()
         expectations = port.expectations_dict()
         parser = TestExpectationParser(port, all_tests=(), is_lint_mode=False)
         expectations_file, expectations_contents = expectations.items()[0]
         expectation_lines = parser.parse(expectations_file, expectations_contents)
         csv_rows = []
         for line in expectation_lines:
             row = self.check_expectations_line(line)
             if row:
                 csv_rows.append(row)
         if self.csv_filename:
             self.write_csv(csv_rows)
 
     def write_csv(self, rows):
+        # TODO(nainar): Add following to CSV       
+        # - Add days since last comment
+        # - Add whether or not the bug is closed
         row_strings = [', '.join(r) for r in rows]
-        contents = '\n'.join(row_strings) + '\n'
+        contents = CSV_ROW_HEADERS + '\n'.join(row_strings) + '\n'
         self.host.filesystem.write_text_file(self.csv_filename, contents)

[qyearsley, 2016/08/19 00:37:18]

Note: this function could also be implemented with the csv module
(https://pymotw.com/2/csv/).

It would look something like this:

def write_csv(header, rows):
  out = StringIO.StringIO()
  writer = csv.writer(out)
  writer.writerow(headers)
  for row in rows:
    writer.writerow(row)
  self.host.filesystem.write_text_file(self.csv_filename, out.getvalue())

The advantage of this is that the csv module would handle any comma or quote
characters in fields (e.g. if you included multiple comma-separated bug links in
one field, then that shouldn't be a problem), and it makes some parts of the
code simpler.

Anyway, if this change is made, it could be made in a separate CL.

[nainar, 2016/08/19 03:36:48]

Done.

 
     def check_expectations_line(self, line):
         """Checks the bugs in one test expectations line to see if they're stale.
 
         Args:
             line: A TestExpectationsLine instance.
 
         Returns:
             A CSV row (a list of strings), or None if there are no stale bugs.
         """
         bug_links, test_name = line.bugs, line.name
         try:
             if bug_links and all(self.is_stale(bug_link) for bug_link in bug_links):
                 print line.original_string.strip()
-                return [bug_links[0], test_name]
+                return [bug_links[0], test_name, str(self.last_updated[bug_links[0]]), self.state[bug_links[0]]]
         except urllib2.HTTPError as error:
             if error.code == 404:
                 message = 'got 404, bug does not exist.'
             elif error.code == 403:
                 message = 'got 403, not accessible. Not able to tell if it\'s stale.'
             else:
                 message = str(error)
             print >> sys.stderr, 'Error when checking %s: %s' % (','.join(bug_links), message)
         return None
 
     def is_stale(self, bug_link):
         if bug_link in self.is_stale_results:
             return self.is_stale_results[bug_link]
         # In case there's an error in the request, don't make the same request again.
         self.is_stale_results[bug_link] = False
         bug_number = bug_link.strip(CRBUG_PREFIX)
         url = GOOGLE_CODE_URL % bug_number
         response = urllib2.urlopen(url)
         parsed = json.loads(response.read())
         last_updated = parsed['updated']
         parsed_time = datetime.datetime.strptime(last_updated.split(".")[0] + "UTC", "%Y-%m-%dT%H:%M:%S%Z")
         time_delta = datetime.datetime.now() - parsed_time
         self.is_stale_results[bug_link] = time_delta.days > self.days
+        self.last_updated[bug_link] = time_delta.days
+        self.state[bug_link] = parsed['state']

[qyearsley, 2016/08/19 00:37:18]

It seems like storing the state of the bug isn't closely related to the main
purpose of this function (which is called is_stale) -- this is a signal that
refactoring would make the code clearer.

Currently, a lot of processing and storing things about bugs is done in
is_stale... maybe what we actually want to do is to process all of the bugs
first and store a list of dicts or objects with information about each bug,
*then* go through that list. Would that make sense?

[nainar, 2016/08/19 03:36:48]

Have done the refactors you have asked for.

         return self.is_stale_results[bug_link]
 
 
 def main(argv):
     option_parser = optparse.OptionParser()
     option_parser.add_option(
         '--days', type='int', default=90,
         help='Number of days to consider a bug stale.')
     option_parser.add_option(
         '--create-csv', type='string', default='',
         help='Filename of CSV file to write stale entries to. No file will be written if no name specified.')
     options, _ = option_parser.parse_args(argv)
     printer = StaleTestPrinter(options)
     printer.print_stale_tests()
     return 0
 
 
 if __name__ == '__main__':
     sys.exit(main(sys.argv[1:]))