Issue 932493004: Move more tests to SkipByDesign

Issue 932493004: Move more tests to SkipByDesign (Closed)

Created:
5 years, 10 months ago by kevmoo

Modified:
5 years, 10 months ago

Reviewers:
ricow1, Bill Hesse

CC:
reviews_dartlang.org

Base URL:
https://dart.googlecode.com/svn/branches/bleeding_edge/dart

Target Ref:
refs/heads/master

Visibility:
Public.

More Reviews

Description

Move more tests to SkipByDesign R=whesse@google.com Committed: https://code.google.com/p/dart/source/detail?r=43977

Patch Set 3 : nits #

Created: 5 years, 10 months ago

Download [raw] [tar.bz2]

	Unified diffs	Side-by-side diffs	Delta from patch set	Stats (+17 lines, -22 lines)			Patch
M	pkg/pkg.status	View	1 2	3 chunks	+10 lines, -15 lines	0 comments	Download
M	tests/lib/lib.status	View	1 2	2 chunks	+2 lines, -2 lines	0 comments	Download
M	tests/standalone/standalone.status	View	1 2	1 chunk	+4 lines, -4 lines	0 comments	Download
M	tests/try/try.status	View		1 chunk	+1 line, -1 line	0 comments	Download

Messages

Total messages: 5 (1 generated)

Expand Messages | Collapse Messages | Show Generated Messages | Hide Generated Messages

Bill Hesse

lgtm https://codereview.chromium.org/932493004/diff/20001/pkg/pkg.status File pkg/pkg.status (right): https://codereview.chromium.org/932493004/diff/20001/pkg/pkg.status#newcode151 pkg/pkg.status:151: */test/analyzer_test: SkipByDesign # No need to run analysis ...

5 years, 10 months ago (2015-02-23 08:53:26 UTC) #3

kevmoo

Rolled back a lot of changes – you bring up good points A lot of ...

5 years, 10 months ago (2015-02-23 21:30:18 UTC) #4

Rolled back a lot of changes – you bring up good points

A lot of these can be covered in a more thorough triage

https://codereview.chromium.org/932493004/diff/20001/pkg/pkg.status
File pkg/pkg.status (right):

https://codereview.chromium.org/932493004/diff/20001/pkg/pkg.status#newcode151
pkg/pkg.status:151: */test/analyzer_test: SkipByDesign # No need to run analysis
tests on browser bots
On 2015/02/23 08:53:25, Bill Hesse wrote:
> Why is this?  Could there be a better explanation?  Is it not possible?

I'm not sure, honestly. I'm just running with what's already here.

This is good to go through when we do the linter

https://codereview.chromium.org/932493004/diff/20001/pkg/pkg.status#newcode156
pkg/pkg.status:156: docgen/test/*: SkipByDesign  # Uses dart:io
On 2015/02/23 08:53:25, Bill Hesse wrote:
> I know these are slow, but it seems like there should be a way to make these
run
> on the browser, using HTML5 features.  It would be a good stress test.  Maybe
we
> should file an issue.

docgen is actually going away soonish – well, replaced by a project on github

https://codereview.chromium.org/932493004/diff/20001/pkg/pkg.status#newcode156
pkg/pkg.status:156: docgen/test/*: SkipByDesign  # Uses dart:io
On 2015/02/23 08:53:25, Bill Hesse wrote:
> Add . to the end to be consistent.

Done.

https://codereview.chromium.org/932493004/diff/20001/tests/lib/lib.status
File tests/lib/lib.status (right):

https://codereview.chromium.org/932493004/diff/20001/tests/lib/lib.status#new...
tests/lib/lib.status:322: profiler/metrics_num_test: SkipByDesign # Because of a
int / double type test.
On 2015/02/23 08:53:25, Bill Hesse wrote:
> Isn't there an issue number for this (the dart2js int/double semantics)?  Is
> this really "byDesign", or an issue that is "will not fix" or "as intended". 
It
> is easy to imagine a better world in which this wouldn't be skipped.

Agreed. I'm going to leave this Skip – and mark it as it should be triaged.

Ideally we shouldn't have to skip an entire suite because test case uses
num/double semantics

https://codereview.chromium.org/932493004/diff/20001/tests/standalone/standal...
File tests/standalone/standalone.status (right):

https://codereview.chromium.org/932493004/diff/20001/tests/standalone/standal...
tests/standalone/standalone.status:39: [ $compiler == none && ($runtime == drt
|| $runtime == dartium || $runtime == ContentShellOnAndroid) ]
On 2015/02/23 08:53:25, Bill Hesse wrote:
> Should this line instead say "compiler = none && browser"?

Maybe? We'll have to wait until the linter work to dig in on these

https://codereview.chromium.org/932493004/diff/20001/tests/standalone/standal...
tests/standalone/standalone.status:43: io/*: SkipByDesign # Don't run tests
using dart:io in the browser
On 2015/02/23 08:53:25, Bill Hesse wrote:
> Make all "dart:io" comments the same.

Done.

https://codereview.chromium.org/932493004/diff/20001/tests/standalone/standal...
tests/standalone/standalone.status:44: package/*: SkipByDesign # Do not run
those in Dartium.
On 2015/02/23 08:53:25, Bill Hesse wrote:
> Why don't we run package tests in Dartium?  Isn't this a time issue, rather
than
> a design issue? 

Rolling back the contentious ones – should triage later

https://codereview.chromium.org/932493004/diff/20001/tests/standalone/standal...
tests/standalone/standalone.status:68: typed_data_test: SkipByDesign #
dart:typed_data support needed.
On 2015/02/23 08:53:25, Bill Hesse wrote:
> I thought we had typed_data on dart2js.  Is this just incomplete support, so
> this is actually an issue, or is it really not supported.  There are typed
> arrays on the web.

Rolling back the contentious ones – should triage later

https://codereview.chromium.org/932493004/diff/20001/tests/standalone/standal...
tests/standalone/standalone.status:82: left_shift_bit_and_op_test: SkipByDesign
# Integers exceed dart2js precision.
On 2015/02/23 08:53:25, Bill Hesse wrote:
> Again, if we have an issue for this, I think it shouldn't be SkipByDesign, but
> Skip with an issue.  Or is this not the plan.  Is Skip due to an unfixable
issue
> the same as Skip due to the inherent properties of the platform?

Acknowledged.

https://codereview.chromium.org/932493004/diff/20001/tests/standalone/standal...
tests/standalone/standalone.status:83: pow_test: SkipByDesign # Precision > 53
bits.
On 2015/02/23 08:53:25, Bill Hesse wrote:
> Just say the int/double issue here, if there isn't a separate precision issue.

> Make the comments the same for all 3 tests.

Acknowledged.

https://codereview.chromium.org/932493004/diff/20001/tests/standalone/standal...
tests/standalone/standalone.status:86: javascript_int_overflow_test: Skip
On 2015/02/23 08:53:25, Bill Hesse wrote:
> Probably the same issue.

Acknowledged.

https://codereview.chromium.org/932493004/diff/20001/tests/standalone/standal...
tests/standalone/standalone.status:89: vmservice/*: SkipByDesign # Do not run
standalone vm service tests with dart2js.
On 2015/02/23 08:53:25, Bill Hesse wrote:
> Should we have a !$standalone variable and section, for things that are
skipped
> on both Dartium and dart2js?

Acknowledged.

https://codereview.chromium.org/932493004/diff/20001/tests/standalone/standal...
tests/standalone/standalone.status:93: unboxed_int_converter_test: Skip
On 2015/02/23 08:53:25, Bill Hesse wrote:
> Probably int/double issue.

Acknowledged.

kevmoo

5 years, 10 months ago (2015-02-23 21:31:20 UTC) #5

Message was sent while issue was closed.

Committed patchset #3 (id:40001) manually as 43977 (presubmit successful).

Expand Messages | Collapse Messages | Show Generated Messages | Hide Generated Messages