[Findit] Enable project classifier and component classifier

appengine/findit/model/crash/crash_config.py

Index: appengine/findit/model/crash/crash_config.py
diff --git a/appengine/findit/model/crash/crash_config.py b/appengine/findit/model/crash/crash_config.py
index 20e6c7a3558caa0a51d6cc4b5ecb29c9276b9ca7..4e66285a8b6c063a1a64022ec656bfe6c1122b74 100644
--- a/appengine/findit/model/crash/crash_config.py
+++ b/appengine/findit/model/crash/crash_config.py
@@ -4,6 +4,8 @@
 
 """Findit for crash (ClusterFuzz & Fracas/Chromecrash) configuration."""
 
+import re
+
 from google.appengine.ext import ndb
 
 from model.versioned_config import VersionedConfig
@@ -19,3 +21,83 @@ class CrashConfig(VersionedConfig):
   #   },
   # }
   fracas = ndb.JsonProperty(indexed=False, default={})
+
+  ################## Settings shared by Fracas/Clusterfuzz. ##################
+  # Project classifier settings:
+  # {
+  #   "host_directories": [
+  #     "src/chrome/browser/resources/",
+  #     "src/chrome/test/data/layout_tests/",
+  #     "src/media/",
+  #     "src/sdch/",
+  #     "src/testing/",
+  #     "src/third_party/WebKit/",
+  #     "src/third_party/",
+  #     "src/tools/",
+  #     "src/"
+  #   ],
+  #   # Where there is no dep_path found, use function and file_path makers to
+  #   # map a Result or StackFrame to a project name.
+  #     "function_marker_to_project_name": {
+  #       "org.chromium": "chromium",
+  #       "com.google.android.apps.chrome": "clank",
+  #       "android.": "android_os",
+  #       "com.android.": "android_os",
+  #   },
+  #   "file_path_marker_to_project_name": {
+  #     ("https___googleplex-android.googlesource."
+  #      "com_a_platform_manifest.git/"): "android_os",
+  #     "googleplex-android/": "android_os",
+  #   },
+  #
+  #   # Number of frames on top to consider when deciding the crashed project.
+  #   "top_n": 4,
+  #
+  #   # The chromium project should always have the highest rank priority (0).
+  #   # This dict assigns rank priorities to non chromium projects.
+  #   "non_chromium_project_rank_priority" = {
+  #     "clank": -1,
+  #     "android_os": -2,
+  #     "android_os_java": -2,
+  #     "src_internal": -3,
+  #     "others": -4,
+  #   }
+  # }
+  project_classifier = ndb.JsonProperty(indexed=False, default={})
+
+  # Component classifier settings:
+  # {
+  #   # Number of frames on top to consider when deciding the crashed
+  #   #component.
+  #   "top_n": 4,
+  #   "path_function_component": [
+  #     [r"src/third_party/WebKit/Source/core/layout", , "Blink>Layout"],
+  #     ...
+  #   ]
+  # }
+  component_classifier = ndb.JsonProperty(indexed=False, default={},
+                                          compressed=True)
+
+  # cached component classifer with all re patterns compiled.
+  cached_component_classifier = ndb.PickleProperty(indexed=False, default=None,

[stgao, 2016/05/24 00:10:58]

Do we plan to save this in NDB? If not, let's not make it a NDB property.

It is set in "def compiled_component_classifier" below, but never saved to NDB
through a put.

[Sharu Jiang, 2016/05/24 22:22:44]

Done.

+                                                   compressed=True)
+
+  def ClearCache(self):
+    self.cached_component_classifier = None
+
+  @property
+  def compiled_component_classifier(self):
+    """Returns the component classifier with all re patterns compiled."""
+    if self.cached_component_classifier is None:
+      compiled_path_function_component = []
+      for path, function, component in self.component_classifier[
+          'path_function_component']:
+        compiled_path_function_component.append(
+            [re.compile(path), re.compile(function), component])
+
+      self.cached_component_classifier = {
+          'top_n': self.component_classifier['top_n'],
+          'path_function_component': compiled_path_function_component
+      }
+
+    return self.cached_component_classifier