Bug 1184966 - perfherder should let harness do summarization

tests/etl/test_perf_data_adapters.py

@@ -3,6 +3,7 @@
 # file, you can obtain one at http://mozilla.org/MPL/2.0/.
 
 import json
+import zlib
 
 from tests.sampledata import SampleData
 from treeherder.etl.perf_data_adapters import TalosDataAdapter
@@ -50,4 +51,27 @@ def test_adapt_and_load():
         datum['blob'] = json.dumps({'talos_data': [datum['blob']]})
         tda.adapt_and_load(reference_data, job_data, datum)
 
+        # we upload a summary with a suite and subtest values, +1 for suite
+        if 'summary' in datum['blob']:
+            results = json.loads(zlib.decompress(tda.performance_artifact_placeholders[-1][4]))
+            data = json.loads(datum['blob'])['talos_data'][0]
+            assert results["blob"]["performance_series"]["geomean"] == data['summary']['suite']
+
+            # deal with the subtests now
+            subtests = len(data['summary']['subtests'])
+            for iter in range(0, subtests):
+                subresults = json.loads(zlib.decompress(tda.performance_artifact_placeholders[-1 - iter][4]))
+                if 'subtest_signatures' in subresults["blob"]['signature_properties']:
+                    # ignore summary signatures
+                    continue
+
+                subdata = data['summary']['subtests'][subresults["blob"]['signature_properties']['test']]
+                for datatype in ['min', 'max', 'mean', 'median', 'std']:
+                    print datatype
+                    assert subdata[datatype] == subresults["blob"]["performance_series"][datatype]
+        else:
+            # FIXME: the talos data blob we're currently using contains datums with summaries and those without
+            # we should probably test non-summarized data as well
+            pass
+
     assert result_count == len(tda.performance_artifact_placeholders)

tests/sample_data/artifacts/performance/talos_perf.json

@@ -26,6 +26,13 @@
                 "rss": false
             }
         },
+        "summary": {
+            "suite": 3141.00,
+            "subtests": {
+                "dhtml.html": {"min": 1, "max": 100, "std": 0.75, "mean": 50, "median": 50},
+                "tablemutation.html": {"min": 1, "max": 100, "std": 0.75, "mean": 50, "median": 50}
+            }
+        },
         "results": {
             "dhtml.html": [
                 1273,

treeherder/etl/perf_data_adapters.py

@@ -89,6 +89,11 @@ def _round(num):
         # that inflate the size of the stored data structure
         return round(num, 2)
 
+    @staticmethod
+    def _extract_summary_data(suite_data, summary):
+        suite_data["geomean"] = summary["suite"]
+        return suite_data
+
     @staticmethod
     def _calculate_summary_data(job_id, result_set_id, push_timestamp, results):
         values = []
@@ -108,6 +113,19 @@ def _calculate_summary_data(job_id, result_set_id, push_timestamp, results):
             "geomean": PerformanceDataAdapter._round(geomean)
         }
 
+    @staticmethod
+    def _extract_test_data(series_data, summary):
+        if not isinstance(summary, dict):
+            return series_data
+
+        series_data["min"] = PerformanceDataAdapter._round(summary["min"])
+        series_data["max"] = PerformanceDataAdapter._round(summary["max"])
+        series_data["std"] = PerformanceDataAdapter._round(summary["std"])
+        series_data["median"] = PerformanceDataAdapter._round(summary["median"])
+        series_data["mean"] = PerformanceDataAdapter._round(summary["mean"])
+
+        return series_data
+
     @staticmethod
     def _calculate_test_data(job_id, result_set_id, push_timestamp,
                              replicates):
@@ -328,8 +346,12 @@ def adapt_and_load(self, reference_data, job_data, datum):
 
                 series_data = self._calculate_test_data(
                     job_id, result_set_id, push_timestamp,
-                    talos_datum["results"][_test]
-                )
+                    talos_datum["results"][_test])
+
+                if "summary" in talos_datum and talos_datum["summary"]["subtests"][_test]:
+                    summary_data = talos_datum["summary"]["subtests"][_test]
+                    series_data = self._extract_test_data(series_data,
+                                                          summary_data)
 
                 obj = self._get_base_perf_obj(_job_guid, _name, _type,
                                               talos_datum,
@@ -357,6 +379,10 @@ def adapt_and_load(self, reference_data, job_data, datum):
                 summary_data = self._calculate_summary_data(
                     job_id, result_set_id, push_timestamp, talos_datum["results"])
 
+                if "summary" in talos_datum and "suite" in talos_datum["summary"]:
+                    summary_data = self._extract_summary_data(summary_data,
+                                                              talos_datum["summary"])
+
                 obj = self._get_base_perf_obj(_job_guid, _name, _type,
                                               talos_datum,
                                               summary_signature,