897104 - "TypeError: gControllerSources.getCache is not a function: testStateBeforeReload@chrome://mochitests/content/browser/browser/devtools/debugger/test/browser_dbg_sources-cache.js:117" on green debug test runs

Status: RESOLVED FIXED

(DevTools :: Debugger, defect)

Description

[Ed Morley [:emorley]]

A soon to be deployed TBPL parser improvement has found a whole bunch of log spam in green debug mochitest-browser-chrome runs.

a) Should these be making the test run fail?
b) We should handle/fix, so we don't spam the annotated failure summary once this TBPL patchset is rolled out.

eg:

https://tbpl-dev.allizom.org/php/getParsedLog.php?id=25604620&tree=Mozilla-Central#error3

Rev4 MacOSX Lion 10.7 mozilla-central debug test mochitest-browser-chrome on 2013-07-23 02:32:38 PDT for push fb4bf993a58a

slave: talos-r4-lion-044

{
03:30:28     INFO -  var foo = {
03:30:28     INFO -    a_test: function() {
03:30:28     INFO -    },
03:30:28     INFO -    n_test: function x() {
03:30:28     INFO -    },
03:30:28     INFO -    sub: {
03:30:28     INFO -      a_test: function() {
03:30:28     INFO -      },
03:30:28     INFO -      n_test: function y() {
03:30:28     INFO -      },
03:30:28     INFO -      sub: {
03:30:28     INFO -        a_test: function() {
03:30:28     INFO -        },
03:30:28     INFO -        n_test: function z() {
03:30:28     INFO -        },
03:30:28     INFO -        sub: {
03:30:28     INFO -          test_SAME_NAME: function test_SAME_NAME() {
03:30:28     INFO -          }
03:30:28     INFO -        }
03:30:28     INFO -      }
03:30:28     INFO -    }
03:30:28     INFO -  };
03:30:28     INFO -  DBG-FRONTEND: Handling tab navigation in the DebuggerView
03:30:28     INFO -  WARNING: NS_ENSURE_TRUE(mDisabledJSAndPlugins) failed: file ../../../../editor/composer/src/nsEditingSession.cpp, line 208
03:30:28     INFO -  WARNING: NS_ENSURE_TRUE(rootContent) failed: file ../../../../editor/composer/src/nsEditorSpellCheck.cpp, line 698
03:30:28     INFO -  TEST-PASS | chrome://mochitests/content/browser/browser/devtools/debugger/test/browser_dbg_sources-cache.js | There should be no sources present in the sources list during reload.
03:30:28     INFO -  TypeError: gControllerSources.getCache is not a function: testStateBeforeReload@chrome://mochitests/content/browser/browser/devtools/debugger/test/browser_dbg_sources-cache.js:117
}

http://hg.mozilla.org/mozilla-central/file/b717a7945dfb/browser/devtools/debugger/test/browser_dbg_sources-cache.js#l117

   114 function testStateBeforeReload() {
   115   is(gSources.itemCount, 0,
   116     "There should be no sources present in the sources list during reload.");
   117   is(gControllerSources.getCache().length, 0,
   118     "The sources cache should be empty during reload.");
   119   is(gDebugger.SourceUtils._labelsCache, gPrevLabelsCache,
   120     "The labels cache has been refreshed during reload and no new objects were created.");
   121   is(gDebugger.SourceUtils._groupsCache, gPrevGroupsCache,
   122     "The groups cache has been refreshed during reload and no new objects were created.");
   123   is(gDebugger.SourceUtils._labelsCache.size, 0,
   124     "There should be no labels cached during reload");
   125   is(gDebugger.SourceUtils._groupsCache.size, 0,
   126     "There should be no groups cached during reload");
   127 }

Comment 1

[Panos Astithas (he/him) [:past] (please ni?)]

Bug 896534 should cut down on the log spam, but we should certainly fix any errors. Do you have the complete list somewhere, or is it just this one?

Comment 2

[Victor Porof [:vporof][:vp]]

This should definitely have been failing :/ The function was removed in bug 886170. The |is| check can be safely removed since the function doesn't exist anymore.

Comment 3

[Ryan VanderMeulen [:RyanVM]]

Attachment: patch

Per comment 2.

Comment 4

[:Gavin Sharp [email: [e-mail]]]

I would love to understand how this exception wasn't being propagated to the mochitest onerror handler!

Comment 5

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/cdda582f86ce

Comment 6

[Panos Astithas (he/him) [:past] (please ni?)]

(In reply to :Gavin Sharp (use [e-mail] for email) from comment #4)
> I would love to understand how this exception wasn't being propagated to the
> mochitest onerror handler!

I think exception in event listeners do that?
I guess we should have been using our makeInfallible helper here:

http://mxr.mozilla.org/mozilla-central/source/toolkit/devtools/DevToolsUtils.js#41

Comment 7

[u279076]

Assuming no verification needed here. Please add the verifyme keyword and remove the [qa-] whiteboard tag to request verification.