Bugfix: Failed verification assumed data-driven testing.

QCOMPARE and friends could be used to ensure that performance
was measured for the intended code path: A failed verification
step would indicate an invalid benchmark result. This information
would then be stored in the database and taken into consideration
in a later analysis.

The feature falsely relied on explicit data tags being used and
was simply ignored for test functions without such tags
(i.e. without a corresponding _data() function).

1 files changed, 9 insertions, 1 deletions

diff --git a/scripts/uploadresults.py b/scripts/uploadresults.py
index 60c16a6..a50285c 100755
--- a/scripts/uploadresults.py
+++ b/scripts/uploadresults.py
@@ -79,12 +79,20 @@ def canonicalMetric(metric):
 def matchesFailedIncident(dataTag, incidents):
     for incident in incidents:
         if (incident.getAttribute("type") == "fail"):
+
+            dataTagElems = incident.getElementsByTagName("DataTag")
+
+            if dataTag == "": # special case
+                assert len(dataTagElems) == 0
+                return True
+
             try:
-                dataTagElem = incident.getElementsByTagName("DataTag")[0]
+                dataTagElem = dataTagElems[0]
             except:
                 continue
             if (dataTagElem.childNodes[0].data == dataTag):
                 return True
+
     return False