[pinboard] Compute averages over a fixed set of stories

Sometimes stories get disabled or removed, causing averages to shift due
to the different workloads of the stories that remain.

This CL specifies a fixed set of stories to use when computing averages
over the entire timeseries.

Note: To recompute averages for values in the past one needs to
remove the file _cached_data/pinboard/aggregated.pkl and run pinboard
again.

Bug: 958736
Change-Id: I3b5ec9ca57c74f61f16466c387b3b99b6edb687d
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1680186
Commit-Queue: Juan Antonio Navarro Pérez <perezju@chromium.org>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Cr-Commit-Position: refs/heads/master@{#673280}

tools/perf/cli_tools/pinboard/pinboard.py

@@ -43,6 +43,28 @@ MEASUREMENTS = set([
     'V8-Only:duration',
 ])
 
+# Compute averages over a fixed set of active stories. These may need to be
+# periodically updated.
+ACTIVE_STORIES = set([
+    'browse:chrome:newtab',
+    'browse:chrome:omnibox',
+    'browse:media:facebook_photos',
+    'browse:media:googleplaystore:2019',
+    'browse:media:imgur',
+    'browse:media:youtube',
+    'browse:news:cricbuzz',
+    'browse:news:toi',
+    'browse:shopping:amazon',
+    'browse:shopping:lazada',
+    'browse:social:facebook',
+    'browse:social:instagram',
+    'browse:tools:maps',
+    'load:media:facebook_photos',
+    'load:media:youtube:2018',
+    'load:news:irctc',
+    'load:news:wikipedia:2018'
+])
+
 
 def StartPinpointJobs(state, date):
   """Start new pinpoint jobs for the last commit on the given date."""
@@ -179,8 +201,9 @@ def GetRevisionResults(item):
   assert df['change'].str.contains(item['revision']).all(), (
       'Not all results match the expected git revision')
 
-  # Filter out and keep only the measurements we want.
+  # Filter out and keep only the measurements and stories that we want.
   df = df[df['name'].isin(MEASUREMENTS)]
+  df = df[df['story'].isin(ACTIVE_STORIES)]
 
   if not df.empty:
     # Aggregate over the results of individual stories.

tools/perf/cli_tools/pinboard/pinboard_unittest.py

@@ -168,7 +168,10 @@ class PinboardToolTests(unittest.TestCase):
     with open(filename, 'w') as f:
       f.writelines(csv)
 
-    df = pinboard.GetRevisionResults(item)
+    with mock.patch('cli_tools.pinboard.pinboard.ACTIVE_STORIES',
+                    new=['story1', 'story2']):
+      df = pinboard.GetRevisionResults(item)
+
     self.assertEqual(len(df.index), 2)  # Only two rows of output.
     self.assertTrue((df['revision'] == '2a66ba').all())
     self.assertTrue((df['benchmark'] == 'loading').all())