Omission from https://codereview.chromium.org/23595002/.

Omission from https://codereview.chromium.org/23595002/.

Committed: https://code.google.com/p/dart/source/detail?r=26731

[ahe, 2013-08-27 17:43:59]

TBR to make build green.

[ahe, 2013-08-27 17:44:18]

Committed patchset #1 manually as r26731 (presubmit successful).

[kustermann, 2013-08-28 00:28:39]

LGTM

https://codereview.chromium.org/23566003/diff/1/dart/tests/html/html.status
File dart/tests/html/html.status (right):

https://codereview.chromium.org/23566003/diff/1/dart/tests/html/html.status#n...
dart/tests/html/html.status:247: datalistelement_test: Fail #
http://dartbug.com/12797
We try to unify our comments: -> "# Issue 12797"

[ahe, 2013-08-28 05:06:38]

Thank you, Martin!

https://codereview.chromium.org/23566003/diff/1/dart/tests/html/html.status
File dart/tests/html/html.status (right):

https://codereview.chromium.org/23566003/diff/1/dart/tests/html/html.status#n...
dart/tests/html/html.status:247: datalistelement_test: Fail #
http://dartbug.com/12797
On 2013/08/28 00:28:39, kustermann wrote:
> We try to unify our comments: -> "# Issue 12797"

Then we should unify on something unambiguous. Where is this documented?

[kustermann, 2013-08-28 07:59:04]

On 2013/08/28 05:06:38, ahe wrote:
> Thank you, Martin!
> 
> https://codereview.chromium.org/23566003/diff/1/dart/tests/html/html.status
> File dart/tests/html/html.status (right):
> 
>
https://codereview.chromium.org/23566003/diff/1/dart/tests/html/html.status#n...
> dart/tests/html/html.status:247: datalistelement_test: Fail #
> http://dartbug.com/12797
> On 2013/08/28 00:28:39, kustermann wrote:
> > We try to unify our comments: -> "# Issue 12797"
> 
> Then we should unify on something unambiguous. Where is this documented?

Currently nowhere. But sgjesse is making *a lot* of status file updates lately,
and he's using this scheme.

I proposed actually a different scheme, something like BUG(dart:12797),
BUG(co19:123,dart:456).
This would make it unambiguous and clear.

"tools/testing/dart/status_file_paser.dart" contains a simple regexp
("IssueNumberPattern") to parse these numbers. Eventually, we want to make
presubmit scripts to keep our status files in a consistent and clean way.

[ahe, 2013-08-28 08:11:21]

On 2013/08/28 07:59:04, kustermann wrote:
> On 2013/08/28 05:06:38, ahe wrote:
> > Thank you, Martin!
> > 
> > https://codereview.chromium.org/23566003/diff/1/dart/tests/html/html.status
> > File dart/tests/html/html.status (right):
> > 
> >
>
https://codereview.chromium.org/23566003/diff/1/dart/tests/html/html.status#n...
> > dart/tests/html/html.status:247: datalistelement_test: Fail #
> > http://dartbug.com/12797
> > On 2013/08/28 00:28:39, kustermann wrote:
> > > We try to unify our comments: -> "# Issue 12797"
> > 
> > Then we should unify on something unambiguous. Where is this documented?
> 
> Currently nowhere. But sgjesse is making *a lot* of status file updates
lately,
> and he's using this scheme.
> 
> I proposed actually a different scheme, something like BUG(dart:12797),
> BUG(co19:123,dart:456).
> This would make it unambiguous and clear.
> 
> "tools/testing/dart/status_file_paser.dart" contains a simple regexp
> ("IssueNumberPattern") to parse these numbers. Eventually, we want to make
> presubmit scripts to keep our status files in a consistent and clean way.

Thank you, I'll follow any unambiguous and documented convention. I feel there
is a strong reason for using a URL as that both unambiguous and self-documenting
(external people don't need to read any documentation to follow a reference to a
bug).

[kustermann, 2013-08-28 08:35:25]

On 2013/08/28 08:11:21, ahe wrote:
> On 2013/08/28 07:59:04, kustermann wrote:
> > On 2013/08/28 05:06:38, ahe wrote:
> > > Thank you, Martin!
> > > 
> > >
https://codereview.chromium.org/23566003/diff/1/dart/tests/html/html.status
> > > File dart/tests/html/html.status (right):
> > > 
> > >
> >
>
https://codereview.chromium.org/23566003/diff/1/dart/tests/html/html.status#n...
> > > dart/tests/html/html.status:247: datalistelement_test: Fail #
> > > http://dartbug.com/12797
> > > On 2013/08/28 00:28:39, kustermann wrote:
> > > > We try to unify our comments: -> "# Issue 12797"
> > > 
> > > Then we should unify on something unambiguous. Where is this documented?
> > 
> > Currently nowhere. But sgjesse is making *a lot* of status file updates
> lately,
> > and he's using this scheme.
> > 
> > I proposed actually a different scheme, something like BUG(dart:12797),
> > BUG(co19:123,dart:456).
> > This would make it unambiguous and clear.
> > 
> > "tools/testing/dart/status_file_paser.dart" contains a simple regexp
> > ("IssueNumberPattern") to parse these numbers. Eventually, we want to make
> > presubmit scripts to keep our status files in a consistent and clean way.
> 
> Thank you, I'll follow any unambiguous and documented convention. I feel there
> is a strong reason for using a URL as that both unambiguous and
self-documenting
> (external people don't need to read any documentation to follow a reference to
a
> bug).

I don't think that external people are reading a lot of status files.
The real advantage of having a URL is IMHO that you can just click on it.
So we're balancing ease of use and compactness (allowing urls also means that
people can use different URLs to refer to the same bug).

[ahe, 2013-08-28 08:42:22]

On 2013/08/28 08:35:25, kustermann wrote:
> I don't think that external people are reading a lot of status files.

That might be, but it would be nice to have a conventions that works for
comments in code as well.

> The real advantage of having a URL is IMHO that you can just click on it.

Certainly.

> So we're balancing ease of use and compactness (allowing urls also means that
> people can use different URLs to refer to the same bug).

If you're willing to enforce a syntax, you can also validate that URLs match
patterns.