Remove test result whitespace stripping from blink_gc_plugin test script

Remove test result whitespace stripping from blink_gc_plugin test script

To make things easier for future developers, the test script now parses
the options using argparse and provides a --reset-results flag to easily
update the baselines.

BUG=none

Committed: https://crrev.com/5171d85366fe1bd7c1de43e866adc3c3e50bc419
Cr-Commit-Position: refs/heads/master@{#329904}

R=thakis@chromium.org

Committed: https://chromium.googlesource.com/chromium/src/+/fc70d1951701938d38b5fd2cd7eab39b3e06e9c0

[dcheng, 2015-05-14 13:42:52]

dcheng@chromium.org changed reviewers:
+ thakis@chromium.org

[dcheng, 2015-05-14 13:42:53]

Just some miscellaneous cleanup.

[Nico, 2015-05-14 19:29:37]

lgtm

Some incoherent rambling below, feel free to ignore :-)

https://codereview.chromium.org/1132293003/diff/1/tools/clang/blink_gc_plugin...
File tools/clang/blink_gc_plugin/tests/cycle_sub.txt (right):

https://codereview.chromium.org/1132293003/diff/1/tools/clang/blink_gc_plugin...
tools/clang/blink_gc_plugin/tests/cycle_sub.txt:6: 
i wonder if we should change the tool to not print a superfluous newline here

https://codereview.chromium.org/1132293003/diff/1/tools/clang/blink_gc_plugin...
File tools/clang/blink_gc_plugin/tests/test.py (right):

https://codereview.chromium.org/1132293003/diff/1/tools/clang/blink_gc_plugin...
tools/clang/blink_gc_plugin/tests/test.py:121: help='If specified, overwrites
the expected results in place.')
(fwiw, I always thought the webkit baseline test runner had way too many flags
and I could never figure out which ones to use. if this script ends up getting
tons of flags, I'm going to delete most of them. Having one seems fine though)

[Nico, 2015-05-14 19:29:53]

The CQ bit was checked by thakis@chromium.org

[commit-bot: I haz the power, 2015-05-14 19:30:49]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1132293003/1

[commit-bot: I haz the power, 2015-05-14 19:44:17]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2015-05-14 19:45:12]

Patchset 1 (id:??) landed as
https://crrev.com/5171d85366fe1bd7c1de43e866adc3c3e50bc419
Cr-Commit-Position: refs/heads/master@{#329904}

[hans, 2015-05-15 20:09:40]

hans@chromium.org changed reviewers:
+ hans@chromium.org

[hans, 2015-05-15 20:09:41]

The tests now fail on Mac for me:

[...]
2: Testing cycle_super_neg.cpp... failed: no expected file found
2: Testing destructor_access_finalized_field.cpp... passed!
2: Testing destructor_in_nonfinalized_class.cpp... passed!
2: Testing fields_require_tracing.cpp... passed!
2: Testing finalize_after_dispatch.cpp... passed!
2: Testing garbage_collected_mixin.cpp... passed!
2: Testing ignore_class.cpp... passed!
2: Testing ignore_fields.cpp... passed!
2: Testing inner_class.cpp... passed!
2: Testing left_most_gc_base.cpp... passed!
2: Testing member_in_offheap_class.cpp... passed!
2: Testing non_virtual_trace.cpp... passed!
2: Testing own_ptr_to_gc_managed_class.cpp... passed!
2: Testing part_object_to_gc_derived_class.cpp... passed!
2: Testing persistent_field_in_gc_managed_class.cpp... passed!
2: Testing polymorphic_class_with_non_virtual_trace.cpp... passed!
2: Testing pure_virtual_trace.cpp... passed!
2: Testing raw_ptr_to_gc_managed_class.cpp... passed!
2: Testing ref_ptr_to_gc_managed_class.cpp... passed!
2: Testing register_weak_members_template.cpp... failed: no expected file found
2: Testing stack_allocated.cpp... passed!
2: Testing templated_class_with_local_class_requires_trace.cpp... passed!
2: Testing trace_after_dispatch.cpp... passed!
2: Testing trace_after_dispatch_impl.cpp... failed: no expected file found
2: Testing trace_after_dispatch_impl_error.cpp... passed!
2: Testing trace_collections.cpp... passed!
2: Testing trace_if_needed.cpp... passed!
2: Testing trace_templated_super.cpp... passed!
2: Testing traceimpl.cpp... failed: no expected file found
2: Testing traceimpl_dependent_scope.cpp... passed!
2: Testing traceimpl_derived_from_templated_base.cpp... failed: no expected file
found
2: Testing traceimpl_error.cpp... passed!
2: Testing traceimpl_omitted_trace.cpp... failed: no expected file found
2: Testing traceimpl_overloaded.cpp... failed: no expected file found
2: Testing traceimpl_overloaded_error.cpp... passed!
2: Testing virtual_and_trace_after_dispatch.cpp... passed!
2: Testing weak_fields_require_tracing.cpp... passed!
2: Ran 51 tests: 44 succeeded, 7 failed
2:     cycle_super_neg
2:     register_weak_members_template
2:     trace_after_dispatch_impl
2:     traceimpl
2:     traceimpl_derived_from_templated_base
2:     traceimpl_omitted_trace
2:     traceimpl_overloaded
2/2 Test #2: blink_gc_plugin_test .............***Failed    1.28 sec

[hans, 2015-05-15 20:13:39]

A revert of this CL (patchset #1 id:1) has been created in
https://codereview.chromium.org/1139113004/ by hans@chromium.org.

The reason for reverting is: This broke the blink_gc_plugin tests on Mac (I
didn't try Linux yet). Reverting it solved the problem locally..

[Nico, 2015-05-15 20:15:27]

Feel free to revert
On May 15, 2015 1:09 PM, <hans@chromium.org> wrote:

> The tests now fail on Mac for me:
>
> [...]
> 2: Testing cycle_super_neg.cpp... failed: no expected file found
> 2: Testing destructor_access_finalized_field.cpp... passed!
> 2: Testing destructor_in_nonfinalized_class.cpp... passed!
> 2: Testing fields_require_tracing.cpp... passed!
> 2: Testing finalize_after_dispatch.cpp... passed!
> 2: Testing garbage_collected_mixin.cpp... passed!
> 2: Testing ignore_class.cpp... passed!
> 2: Testing ignore_fields.cpp... passed!
> 2: Testing inner_class.cpp... passed!
> 2: Testing left_most_gc_base.cpp... passed!
> 2: Testing member_in_offheap_class.cpp... passed!
> 2: Testing non_virtual_trace.cpp... passed!
> 2: Testing own_ptr_to_gc_managed_class.cpp... passed!
> 2: Testing part_object_to_gc_derived_class.cpp... passed!
> 2: Testing persistent_field_in_gc_managed_class.cpp... passed!
> 2: Testing polymorphic_class_with_non_virtual_trace.cpp... passed!
> 2: Testing pure_virtual_trace.cpp... passed!
> 2: Testing raw_ptr_to_gc_managed_class.cpp... passed!
> 2: Testing ref_ptr_to_gc_managed_class.cpp... passed!
> 2: Testing register_weak_members_template.cpp... failed: no expected file
> found
> 2: Testing stack_allocated.cpp... passed!
> 2: Testing templated_class_with_local_class_requires_trace.cpp... passed!
> 2: Testing trace_after_dispatch.cpp... passed!
> 2: Testing trace_after_dispatch_impl.cpp... failed: no expected file found
> 2: Testing trace_after_dispatch_impl_error.cpp... passed!
> 2: Testing trace_collections.cpp... passed!
> 2: Testing trace_if_needed.cpp... passed!
> 2: Testing trace_templated_super.cpp... passed!
> 2: Testing traceimpl.cpp... failed: no expected file found
> 2: Testing traceimpl_dependent_scope.cpp... passed!
> 2: Testing traceimpl_derived_from_templated_base.cpp... failed: no
> expected file
> found
> 2: Testing traceimpl_error.cpp... passed!
> 2: Testing traceimpl_omitted_trace.cpp... failed: no expected file found
> 2: Testing traceimpl_overloaded.cpp... failed: no expected file found
> 2: Testing traceimpl_overloaded_error.cpp... passed!
> 2: Testing virtual_and_trace_after_dispatch.cpp... passed!
> 2: Testing weak_fields_require_tracing.cpp... passed!
> 2: Ran 51 tests: 44 succeeded, 7 failed
> 2:     cycle_super_neg
> 2:     register_weak_members_template
> 2:     trace_after_dispatch_impl
> 2:     traceimpl
> 2:     traceimpl_derived_from_templated_base
> 2:     traceimpl_omitted_trace
> 2:     traceimpl_overloaded
> 2/2 Test #2: blink_gc_plugin_test .............***Failed    1.28 sec
>
> https://codereview.chromium.org/1132293003/
>

To unsubscribe from this group and stop receiving emails from it, send an email
to chromium-reviews+unsubscribe@chromium.org.

[hans, 2015-05-15 20:18:00]

On Fri, May 15, 2015 at 1:15 PM, Nico Weber <thakis@chromium.org> wrote:
> Feel free to revert

Already in the cq :-)

I'll also see if I can hook these tests up to run on the ToT builders.

To unsubscribe from this group and stop receiving emails from it, send an email
to chromium-reviews+unsubscribe@chromium.org.

[dcheng, 2015-05-27 21:55:26]

Hans, if you patch this in locally, are you still seeing failures?

I tried running tests with this patch on Linux/Mac/Windows, and I see no
failures.

[Nico, 2015-05-27 21:59:27]

Did anything change? The tests were red on all fyi bots (which run plugin tests)

[dcheng, 2015-05-27 22:02:13]

On 2015/05/27 at 21:59:27, thakis wrote:
> Did anything change? The tests were red on all fyi bots (which run plugin
tests)

I haven't changed anything. It's just difficult for me to figure out how to fix
it if I can't reproduce any failures =(

[hans, 2015-05-28 15:48:28]

On 2015/05/27 22:02:13, dcheng wrote:
> On 2015/05/27 at 21:59:27, thakis wrote:
> > Did anything change? The tests were red on all fyi bots (which run plugin
> tests)
> 
> I haven't changed anything. It's just difficult for me to figure out how to
fix
> it if I can't reproduce any failures =(

I applied this patch locally on Linux and all tests pass.

I think the problem is with how it was committed. Looking at
https://crrev.com/5171d85366fe1bd7c1de43e866adc3c3e50bc419, it seems
tools/clang/blink_gc_plugin/tests/cycle_super_neg.txt was *deleted*, but your
patch doesn't delete it, just removes all the contents.

Maybe landing it manually would work?

[dcheng, 2015-05-28 16:00:29]

On 2015/05/28 at 15:48:28, hans wrote:
> On 2015/05/27 22:02:13, dcheng wrote:
> > On 2015/05/27 at 21:59:27, thakis wrote:
> > > Did anything change? The tests were red on all fyi bots (which run plugin
> > tests)
> > 
> > I haven't changed anything. It's just difficult for me to figure out how to
fix
> > it if I can't reproduce any failures =(
> 
> I applied this patch locally on Linux and all tests pass.
> 
> I think the problem is with how it was committed. Looking at
https://crrev.com/5171d85366fe1bd7c1de43e866adc3c3e50bc419, it seems
tools/clang/blink_gc_plugin/tests/cycle_super_neg.txt was *deleted*, but your
patch doesn't delete it, just removes all the contents.
> 
> Maybe landing it manually would work?

Hmm... I'll try that and see what happens.

[commit-bot: I haz the power, 2015-05-28 16:02:31]

Patchset 1 (id:??) landed as
https://crrev.com/fc70d1951701938d38b5fd2cd7eab39b3e06e9c0
Cr-Commit-Position: refs/heads/master@{#331790}

[dcheng, 2015-05-28 16:02:46]

Committed patchset #1 (id:1) manually as
fc70d1951701938d38b5fd2cd7eab39b3e06e9c0 (presubmit successful).