Equalizing samples when comparing middle revision against good and bad.

scripts/slave/recipe_modules/auto_bisect/revision_state.py

 # Copyright 2015 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.
 
 """An interface for holding state and result of revisions in a bisect job.
 
 When implementing support for tests other than perf, one should extend this
 class so that the bisect module and recipe can use it.
 
 See perf_revision_state for an example.
@@ skipping 526 matching lines @@
       True if the results of testing this revision are significantly different
       from those of testing the earliest known bad revision.
       False if they are instead significantly different form those of testing
       the latest knwon good revision.
     """
 
     if self.bisector.is_return_code_mode():
       return self.overall_return_code == self.bisector.lkgr.overall_return_code
 
     while True:
+      truncate_length = min(len(x) for x in (self.bisector.lkgr.values,
+                                             self.bisector.fkbr.values))

[dtu, 2016/04/20 23:01:17]

A little bit weird for taking the min of just two values. Why not just
min(len(lkgr), len(fkbr))?

(In fact, you don't even need to take the min. Your previous strategy of doing
lkgr[:len(fkbr)] and fkbr[:len(lkgr)] works as well.)

(I also notice that this class is reaching into self.bisector a lot, and that's
the code smell "feature envy.")

[RobertoCN, 2016/04/20 23:13:33]

done. 

Agreed. revision_state will be gutted soon.

       diff_from_good = self.bisector.significantly_different(
-          self.bisector.lkgr.values, self.values)
+          self.bisector.lkgr.values[:truncate_length], self.values)
       diff_from_bad = self.bisector.significantly_different(
-          self.bisector.fkbr.values, self.values)
+          self.bisector.fkbr.values[:truncate_length], self.values)
 
       if diff_from_good and diff_from_bad:
         # Multiple regressions.
         # For now, proceed bisecting the biggest difference of the means.
         dist_from_good = abs(self.mean_value - self.bisector.lkgr.mean_value)
         dist_from_bad = abs(self.mean_value - self.bisector.fkbr.mean_value)
         if dist_from_good > dist_from_bad:
           # TODO(robertocn): Add way to handle the secondary regression
           #self.bisector.handle_secondary_regression(self, self.bisector.fkbr)
           return False
@@ skipping 38 matching lines @@
     else:
       next_revision_to_test.retest()
 
   def __repr__(self):
     if self.overall_return_code is not None:
       return ('RevisionState(rev=%s, values=%r, overall_return_code=%r, '
               'std_dev=%r)') % (self.revision_string(), self.values,
                                 self.overall_return_code, self.std_dev)
     return ('RevisionState(rev=%s, values=%r, mean_value=%r, std_dev=%r)' % (
         self.revision_string(), self.values, self.mean_value, self.std_dev))