Added support for versioning of layout test results of run-webkit-tests runs

Tools/Scripts/webkitpy/layout_tests/models/test_run_results.py

Index: Tools/Scripts/webkitpy/layout_tests/models/test_run_results.py
diff --git a/Tools/Scripts/webkitpy/layout_tests/models/test_run_results.py b/Tools/Scripts/webkitpy/layout_tests/models/test_run_results.py
index f203fa534bea2bbdd9c6015dc6172e06caf8116f..a0cac6a19c36cf05e67655472c72259ce7a574d7 100644
--- a/Tools/Scripts/webkitpy/layout_tests/models/test_run_results.py
+++ b/Tools/Scripts/webkitpy/layout_tests/models/test_run_results.py
@@ -163,7 +163,7 @@ def _interpret_test_failures(failures):
     return test_dict
 
 
-def summarize_results(port_obj, expectations, initial_results, retry_results, enabled_pixel_tests_in_retry, only_include_failing=False):
+def summarize_results(port_obj, expectations, initial_results, retry_results, enabled_pixel_tests_in_retry, only_include_failing=False, archived_results_path=None, revision_info=''):

[Dirk Pranke, 2014/06/17 17:58:58]

it feels a bit odd to me that the results object would know about the previous
results. What are the advantages of this approach vs keeping things completely
separate?

[patro, 2014/07/15 10:36:57]

Yes, things can be seperate there need not be any link .The script does the
ordering of the results based on the directory names (which have timestamps).

     """Returns a dictionary containing a summary of the test runs, with the following fields:
         'version': a version indicator
         'fixable': The number of fixable tests (NOW - PASS)
@@ -306,6 +306,12 @@ def summarize_results(port_obj, expectations, initial_results, retry_results, en
     results['build_number'] = port_obj.get_option('build_number')
     results['builder_name'] = port_obj.get_option('builder_name')
 
+    if port_obj.get_option("enable_versioned_results") and archived_results_path:
+        results['previous_results_location'] = archived_results_path
+        results['chromium_revision'] = ''
+        results['blink_revision'] = ''
+
+
     try:
         # Don't do this by default since it takes >100ms.
         # It's only used for uploading data to the flakiness dashboard.