Bug 200742 – [lldb-webkit] OptionSet summary shows size 0 sometimes for non-empty set

• Status: RESOLVED
• Resolution: FIXED
• Component: Tools / Tests

[Daniel Bates]

Sometimes the LLDB-pretty printed OptionSet summary for a non-empty set is 0.

[Daniel Bates]

Created attachment 376331 [details]
Patch

[Daniel Bates]

Comment on attachment 376331 [details]
Patch

Clearing flags on attachment: 376331

Committed r248970: <https://trac.webkit.org/changeset/248970>

[Daniel Bates]

All reviewed patches have been landed.  Closing bug.

[Radar WebKit Bug Importer]

<rdar://problem/54575345>

[Ryan Haddad]

There is a lldb_webkit_unittest test failure after this change:

[1903/1912] lldb_webkit_unittest.TestSummaryProviders.serial_test_WTFOptionSetProvider_empty failed:
  Traceback (most recent call last):
    File "/Volumes/Data/slave/mojave-release-tests-wk2/build/Tools/lldb/lldb_webkit_unittest.py", line 188, in serial_test_WTFOptionSetProvider_empty
      self.assertEqual(provider.get_child_at_index(0), None)
  AssertionError: <lldb.SBValue; proxy of <Swig Object of type 'lldb::SBValue *' at 0x1176af2a0> > != None
  
https://build.webkit.org/builders/Apple%20Mojave%20Release%20WK2%20(Tests)/builds/6078/steps/webkitpy-test/logs/stdio

[Daniel Bates]

Not near a computer. Roll it out and I will fix offljne

[Ryan Haddad]

Reverted r248970 for reason:

Caused lldb_webkit_unittest.TestSummaryProviders.serial_test_WTFOptionSetProvider_empty to fail.

Committed r248980: <https://trac.webkit.org/changeset/248980>

[Daniel Bates]

Something happened to EWS. It use to run these tests.

[Daniel Bates]

Created attachment 377058 [details]
To land

[Daniel Bates]

Comment on attachment 377058 [details]
To land

Clearing flags on attachment: 377058

Committed r249030: <https://trac.webkit.org/changeset/249030>

[Daniel Bates]

All reviewed patches have been landed.  Closing bug.

[Daniel Bates]

Comment on attachment 377058 [details]
To land

View in context: https://bugs.webkit.org/attachment.cgi?id=377058&action=review

> Tools/lldb/lldb_webkit_unittest.py:188
> -        self.assertEqual(provider.get_child_at_index(0), None)
> +        self.assertEqual(provider.get_child_at_index(0).GetName(), 'm_storage')

In retrospect, probably would be better for this to assert provider.size == 0 and should add a new test to check that the last child of the provider is always 'm_storage'.

[Daniel Bates]

Comment on attachment 377058 [details]
To land

View in context: https://bugs.webkit.org/attachment.cgi?id=377058&action=review

>> Tools/lldb/lldb_webkit_unittest.py:188
>> +        self.assertEqual(provider.get_child_at_index(0).GetName(), 'm_storage')
> 
> In retrospect, probably would be better for this to assert provider.size == 0 and should add a new test to check that the last child of the provider is always 'm_storage'.

Filed bug #201056 for this.