Modify list_system_health_stories to generate_system_health_csv

tools/perf/page_sets/system_health/system_health_story.py

Index: tools/perf/page_sets/system_health/system_health_story.py
diff --git a/tools/perf/page_sets/system_health/system_health_story.py b/tools/perf/page_sets/system_health/system_health_story.py
index aea6f800ea12f269597feaadbf78b913b41ea67f..e80331465afbe5d39a901f96ad9833bda3b1d0e3 100644
--- a/tools/perf/page_sets/system_health/system_health_story.py
+++ b/tools/perf/page_sets/system_health/system_health_story.py
@@ -87,6 +87,12 @@ class SystemHealthStory(page.Page):
     self._take_memory_measurement = take_memory_measurement
 
   @classmethod
+  def GetStoryDescription(cls):
+    if not cls.ABSTRACT_STORY and cls.__doc__:
+      return cls.__doc__

[perezju, 2017/05/18 09:52:35]

maybe just return the first line (title) of the doc string?

[nednguyen, 2017/05/18 14:02:58]

For cases like
https://cs.chromium.org/chromium/src/tools/perf/page_sets/system_health/searc...,
I think it's valuable for people to know exactly what the story does.

[perezju, 2017/05/19 13:48:29]

I think it's valuable to have a long full explanation next to the story
definition when doing a code search. But for the csv, I still think a short one
liner should be better.

This also forces us to come up with good short descriptions
for the stories, which could be surfaced in other places (e.g notes for leads?)

+    return 'N/A'
+
+  @classmethod
   def CanRun(cls, possible_browser):
     if (decorators.ShouldSkip(cls, possible_browser)[0] or
         cls.ShouldDisable(possible_browser)):