Add expectAsyncX and expectAsyncUntilX methods, and deprecate the old methods.

Add expectAsyncX and expectAsyncUntilX methods, and deprecate the old methods.

Fixes #436.

R=nweiz@google.com

Committed: https://github.com/dart-lang/test/commit/7babd03969368b21e51de65a3e84d43925f6e009

[floitsch, 2016-11-21 16:18:50]

floitsch@google.com changed reviewers:
+ leafp@google.com, nweiz@google.com

[floitsch, 2016-11-21 16:18:50]

Started work on changing the package.

Some comments:
1. Tests weren't all successful, but I get the same result as on master.

2. There is some cleanup possible (but I expect lots of code to go away anyway
when the deprecated methods are removed.

3. @leafp: the code is currently not strong mode clean:
"The return type '([dynamic]) → T' isn't a 'F1(P0) → T', as defined by the
method 'expectAsync1'."
I'm not sure why this is. As soon as it takes the same number of arguments and
returns the same type it should be correct.

4. All uses could be automatically changed. (There were only 2 for the `until`
so I did those by hand.

- git grep -l 'expectAsync(()' | xargs sed -i
's/expectAsync(()/expectAsync0(()/g'
- git grep -l 'expectAsync(([^,]*)' | xargs sed -i
's/expectAsync((\([^,]\+\))/expectAsync1((\1)/g'
- git grep -l 'expectAsync(([^,]*,[^,]*)' | xargs sed -i
's/expectAsync((\([^,]\+\),\([^,]\+\))/expectAsync2((\1,\2)/g'
- git grep -l 'expectAsync(([^,]*,[^,]*,[^,]*)' | xargs sed -i
's/expectAsync((\([^,]\+\),\([^,]\+\),\([^,]\+\))/expectAsync3((\1,\2,\3)/g'
- git grep -l 'expectAsync(([^,]*,[^,]*,[^,]*,[^,]*)' | xargs sed -i
's/expectAsync((\([^,]\+\),\([^,]\+\),\([^,]\+\),\([^,]\+\))/expectAsync4((\1,\2,\3,\4)/g'
- git grep -l 'expectAsync(([^,]*,[^,]*,[^,]*,[^,]*,[^,]*)' | xargs sed -i
's/expectAsync((\([^,]\+\),\([^,]\+\),\([^,]\+\),\([^,]\+\),\([^,]\+\))/expectAsync5((\1,\2,\3,\4,\5)/g'
- git grep -l 'expectAsync(([^,]*,[^,]*,[^,]*,[^,]*,[^,]*,[^,]*)' | xargs sed -i
's/expectAsync((\([^,]\+\),\([^,]\+\),\([^,]\+\),\([^,]\+\),\([^,]\+\),\([^,]\+\))/expectAsync6((\1,\2,\3,\4,\5,\6)/g'

[floitsch, 2016-11-21 16:49:52]

On 2016/11/21 16:18:50, floitsch wrote:
> Started work on changing the package.
> 
> Some comments:
> 1. Tests weren't all successful, but I get the same result as on master.
> 
> 2. There is some cleanup possible (but I expect lots of code to go away anyway
> when the deprecated methods are removed.
> 
> 3. @leafp: the code is currently not strong mode clean:
> "The return type '([dynamic]) → T' isn't a 'F1(P0) → T', as defined by the
> method 'expectAsync1'."
> I'm not sure why this is. As soon as it takes the same number of arguments and
> returns the same type it should be correct.

This seems to have been a spurious error from the IDE.
(Probably triggered because I edited the files without having run `pub get`
before.).
I get no warnings/errors/hints anymore.

> 
> 4. All uses could be automatically changed. (There were only 2 for the `until`
> so I did those by hand.
> 
> - git grep -l 'expectAsync(()' | xargs sed -i
> 's/expectAsync(()/expectAsync0(()/g'
> - git grep -l 'expectAsync(([^,]*)' | xargs sed -i
> 's/expectAsync((\([^,]\+\))/expectAsync1((\1)/g'
> - git grep -l 'expectAsync(([^,]*,[^,]*)' | xargs sed -i
> 's/expectAsync((\([^,]\+\),\([^,]\+\))/expectAsync2((\1,\2)/g'
> - git grep -l 'expectAsync(([^,]*,[^,]*,[^,]*)' | xargs sed -i
> 's/expectAsync((\([^,]\+\),\([^,]\+\),\([^,]\+\))/expectAsync3((\1,\2,\3)/g'
> - git grep -l 'expectAsync(([^,]*,[^,]*,[^,]*,[^,]*)' | xargs sed -i
>
's/expectAsync((\([^,]\+\),\([^,]\+\),\([^,]\+\),\([^,]\+\))/expectAsync4((\1,\2,\3,\4)/g'
> - git grep -l 'expectAsync(([^,]*,[^,]*,[^,]*,[^,]*,[^,]*)' | xargs sed -i
>
's/expectAsync((\([^,]\+\),\([^,]\+\),\([^,]\+\),\([^,]\+\),\([^,]\+\))/expectAsync5((\1,\2,\3,\4,\5)/g'
> - git grep -l 'expectAsync(([^,]*,[^,]*,[^,]*,[^,]*,[^,]*,[^,]*)' | xargs sed
-i
>
's/expectAsync((\([^,]\+\),\([^,]\+\),\([^,]\+\),\([^,]\+\),\([^,]\+\),\([^,]\+\))/expectAsync6((\1,\2,\3,\4,\5,\6)/g'

[kevmoo, 2016-11-21 16:54:00]

kevmoo@google.com changed reviewers:
+ kevmoo@google.com

[kevmoo, 2016-11-21 16:54:01]

Also, there are some analysis errors w/ strong mode at 1.21.0-dev.7.0 – and some
duplicate typedefs that can be removed

See

https://github.com/kevmoo/test/commit/03adae99d1c484e7c6e568e5197835fd6857b9ee

Also: remove the analysis exclusion in the .analysis_options for the test
directory

https://codereview.chromium.org/2515303002/diff/1/pubspec.yaml
File pubspec.yaml (right):

https://codereview.chromium.org/2515303002/diff/1/pubspec.yaml#newcode2
pubspec.yaml:2: version: 0.13.0
Since this is just a deprecation, we could call it 0.12.17, right?

[floitsch, 2016-11-21 17:10:29]

On 2016/11/21 16:54:01, kevmoo wrote:
> Also, there are some analysis errors w/ strong mode at 1.21.0-dev.7.0 – 
Yes. I'm getting them again now.
I believe that they are mistake. I must not add the `[` optional brackets `]`
since that would break the existing code for determining the arity of the
function.

> and some
> duplicate typedefs that can be removed

Thanks. Good point.
I used the original typedefs and just added types to them.
Also made them public.
> 
> See
> 
> https://github.com/kevmoo/test/commit/03adae99d1c484e7c6e568e5197835fd6857b9ee
> 
> Also: remove the analysis exclusion in the .analysis_options for the test
> directory
done.
> 
> https://codereview.chromium.org/2515303002/diff/1/pubspec.yaml
> File pubspec.yaml (right):
> 
> https://codereview.chromium.org/2515303002/diff/1/pubspec.yaml#newcode2
> pubspec.yaml:2: version: 0.13.0
> Since this is just a deprecation, we could call it 0.12.17, right?

[floitsch, 2016-11-21 17:10:38]

https://codereview.chromium.org/2515303002/diff/1/pubspec.yaml
File pubspec.yaml (right):

https://codereview.chromium.org/2515303002/diff/1/pubspec.yaml#newcode2
pubspec.yaml:2: version: 0.13.0
On 2016/11/21 16:54:00, kevmoo wrote:
> Since this is just a deprecation, we could call it 0.12.17, right?

Reverted my change.
I'm not sure how '-dev' works. Do I even need to update the version number for a
`-dev`?

[kevmoo, 2016-11-21 17:28:43]

On 2016/11/21 17:10:38, floitsch wrote:
> https://codereview.chromium.org/2515303002/diff/1/pubspec.yaml
> File pubspec.yaml (right):
> 
> https://codereview.chromium.org/2515303002/diff/1/pubspec.yaml#newcode2
> pubspec.yaml:2: version: 0.13.0
> On 2016/11/21 16:54:00, kevmoo wrote:
> > Since this is just a deprecation, we could call it 0.12.17, right?
> 
> Reverted my change.
> I'm not sure how '-dev' works. Do I even need to update the version number for
a
> `-dev`?

-dev means "in progress". Leave it like that until the commit we publish.

[nweiz, 2016-11-21 22:08:03]

https://codereview.chromium.org/2515303002/diff/40001/CHANGELOG.md
File CHANGELOG.md (right):

https://codereview.chromium.org/2515303002/diff/40001/CHANGELOG.md#newcode6
CHANGELOG.md:6: * `expectAsyncUntil0`, `expectAsyncUntil1`, ...
`expectAsyncUntil6`.
It would be nice to have more details here. I'd write:

"Deprecate `expectAsync()` and `expectAsyncUntil()`, since they currently can't
be made to work cleanly in strong mode. They're replaced with separate methods
for each number of callback arguments:"

https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
File lib/src/frontend/expect_async.dart (right):

https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
lib/src/frontend/expect_async.dart:87: (maxExpected == 0 && minExpected > 0) ?
minExpected : maxExpected,
Nit: Please don't include unrelated whitespace diffs.

https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
lib/src/frontend/expect_async.dart:145: T _max0() => _max6();
Since these are being accessed outside the class now, make them public.

https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
lib/src/frontend/expect_async.dart:216: /// (by default 1).
Remove the old documentation.

https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
lib/src/frontend/expect_async.dart:218: /// This function is deprecated. Use
[expectAsync0], [expectAsync1],
"is deprecated because it doesn't work well with strong mode."

https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
lib/src/frontend/expect_async.dart:235: @deprecated
Use @Deprecated and mention that it will be removed in 0.13.0.

https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
lib/src/frontend/expect_async.dart:259: /// [reason] should be the reason the
callback is expected to be called.
Add a paragraph describing the difference between the various expectAsyncN
methods. Maybe:

"This takes a callback with zero arguments. See also [expectAsync1],
[expectAsync2], [expectAsync3], [expectAsync4], [expectAsync5], or
[expectAsync6] for callbacks with different numbers of arguments."

https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
lib/src/frontend/expect_async.dart:260: Func0<dynamic/*=T*/ >
expectAsync0/*<T>*/(dynamic/*=T*/ callback(),
Nit: "<dynamic/*=T*/>"

https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
lib/src/frontend/expect_async.dart:263: throw new StateError("expectAsync() may
only be called within a test.");
"expectAsync0()"

https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
lib/src/frontend/expect_async.dart:271: /// See [expectAsync0].
I'd prefer to copy the docs from [expectAsync0] rather than just refer to them.
I'm willing to make maintenance a little more annoying in exchange for saving
users a click.

[floitsch, 2016-11-22 14:14:27]

PTAL.

Also, the analyzer is still complaining.
@Leaf: do you have any idea why?

https://codereview.chromium.org/2515303002/diff/40001/CHANGELOG.md
File CHANGELOG.md (right):

https://codereview.chromium.org/2515303002/diff/40001/CHANGELOG.md#newcode6
CHANGELOG.md:6: * `expectAsyncUntil0`, `expectAsyncUntil1`, ...
`expectAsyncUntil6`.
On 2016/11/21 22:08:02, nweiz wrote:
> It would be nice to have more details here. I'd write:
> 
> "Deprecate `expectAsync()` and `expectAsyncUntil()`, since they currently
can't
> be made to work cleanly in strong mode. They're replaced with separate methods
> for each number of callback arguments:"

Done.

https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
File lib/src/frontend/expect_async.dart (right):

https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
lib/src/frontend/expect_async.dart:87: (maxExpected == 0 && minExpected > 0) ?
minExpected : maxExpected,
On 2016/11/21 22:08:02, nweiz wrote:
> Nit: Please don't include unrelated whitespace diffs.

I tried, but writing multiline function signatures without invoking the dart
formatter turned out to be too difficult.

I undid this change (and will do the same for any other I can see).

https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
lib/src/frontend/expect_async.dart:145: T _max0() => _max6();
On 2016/11/21 22:08:02, nweiz wrote:
> Since these are being accessed outside the class now, make them public.

Done.

https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
lib/src/frontend/expect_async.dart:216: /// (by default 1).
On 2016/11/21 22:08:02, nweiz wrote:
> Remove the old documentation.

Done.

https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
lib/src/frontend/expect_async.dart:218: /// This function is deprecated. Use
[expectAsync0], [expectAsync1],
On 2016/11/21 22:08:02, nweiz wrote:
> "is deprecated because it doesn't work well with strong mode."

Done.

https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
lib/src/frontend/expect_async.dart:235: @deprecated
On 2016/11/21 22:08:02, nweiz wrote:
> Use @Deprecated and mention that it will be removed in 0.13.0.

Done.

https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
lib/src/frontend/expect_async.dart:259: /// [reason] should be the reason the
callback is expected to be called.
On 2016/11/21 22:08:03, nweiz wrote:
> Add a paragraph describing the difference between the various expectAsyncN
> methods. Maybe:
> 
> "This takes a callback with zero arguments. See also [expectAsync1],
> [expectAsync2], [expectAsync3], [expectAsync4], [expectAsync5], or
> [expectAsync6] for callbacks with different numbers of arguments."

Done.

https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
lib/src/frontend/expect_async.dart:260: Func0<dynamic/*=T*/ >
expectAsync0/*<T>*/(dynamic/*=T*/ callback(),
On 2016/11/21 22:08:02, nweiz wrote:
> Nit: "<dynamic/*=T*/>"

I don't understand.
If it's the spacing: I used the dartformatter.

https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
lib/src/frontend/expect_async.dart:263: throw new StateError("expectAsync() may
only be called within a test.");
On 2016/11/21 22:08:02, nweiz wrote:
> "expectAsync0()"

Done.

https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
lib/src/frontend/expect_async.dart:271: /// See [expectAsync0].
On 2016/11/21 22:08:02, nweiz wrote:
> I'd prefer to copy the docs from [expectAsync0] rather than just refer to
them.
> I'm willing to make maintenance a little more annoying in exchange for saving
> users a click.

Done.

[vsm, 2016-11-22 14:33:38]

On 2016/11/22 14:14:27, floitsch wrote:
> PTAL.
> 
> Also, the analyzer is still complaining.
> @Leaf: do you have any idea why?

DBC - Leaf is out till next Monday.

Does it work if you change:

T max1([a0 = _PLACEHOLDER])

to:

T max1([Object a0 = _PLACEHOLDER])

etc.?

> 
> https://codereview.chromium.org/2515303002/diff/40001/CHANGELOG.md
> File CHANGELOG.md (right):
> 
> https://codereview.chromium.org/2515303002/diff/40001/CHANGELOG.md#newcode6
> CHANGELOG.md:6: * `expectAsyncUntil0`, `expectAsyncUntil1`, ...
> `expectAsyncUntil6`.
> On 2016/11/21 22:08:02, nweiz wrote:
> > It would be nice to have more details here. I'd write:
> > 
> > "Deprecate `expectAsync()` and `expectAsyncUntil()`, since they currently
> can't
> > be made to work cleanly in strong mode. They're replaced with separate
methods
> > for each number of callback arguments:"
> 
> Done.
> 
>
https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
> File lib/src/frontend/expect_async.dart (right):
> 
>
https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
> lib/src/frontend/expect_async.dart:87: (maxExpected == 0 && minExpected > 0) ?
> minExpected : maxExpected,
> On 2016/11/21 22:08:02, nweiz wrote:
> > Nit: Please don't include unrelated whitespace diffs.
> 
> I tried, but writing multiline function signatures without invoking the dart
> formatter turned out to be too difficult.
> 
> I undid this change (and will do the same for any other I can see).
> 
>
https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
> lib/src/frontend/expect_async.dart:145: T _max0() => _max6();
> On 2016/11/21 22:08:02, nweiz wrote:
> > Since these are being accessed outside the class now, make them public.
> 
> Done.
> 
>
https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
> lib/src/frontend/expect_async.dart:216: /// (by default 1).
> On 2016/11/21 22:08:02, nweiz wrote:
> > Remove the old documentation.
> 
> Done.
> 
>
https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
> lib/src/frontend/expect_async.dart:218: /// This function is deprecated. Use
> [expectAsync0], [expectAsync1],
> On 2016/11/21 22:08:02, nweiz wrote:
> > "is deprecated because it doesn't work well with strong mode."
> 
> Done.
> 
>
https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
> lib/src/frontend/expect_async.dart:235: @deprecated
> On 2016/11/21 22:08:02, nweiz wrote:
> > Use @Deprecated and mention that it will be removed in 0.13.0.
> 
> Done.
> 
>
https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
> lib/src/frontend/expect_async.dart:259: /// [reason] should be the reason the
> callback is expected to be called.
> On 2016/11/21 22:08:03, nweiz wrote:
> > Add a paragraph describing the difference between the various expectAsyncN
> > methods. Maybe:
> > 
> > "This takes a callback with zero arguments. See also [expectAsync1],
> > [expectAsync2], [expectAsync3], [expectAsync4], [expectAsync5], or
> > [expectAsync6] for callbacks with different numbers of arguments."
> 
> Done.
> 
>
https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
> lib/src/frontend/expect_async.dart:260: Func0<dynamic/*=T*/ >
> expectAsync0/*<T>*/(dynamic/*=T*/ callback(),
> On 2016/11/21 22:08:02, nweiz wrote:
> > Nit: "<dynamic/*=T*/>"
> 
> I don't understand.
> If it's the spacing: I used the dartformatter.
> 
>
https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
> lib/src/frontend/expect_async.dart:263: throw new StateError("expectAsync()
may
> only be called within a test.");
> On 2016/11/21 22:08:02, nweiz wrote:
> > "expectAsync0()"
> 
> Done.
> 
>
https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
> lib/src/frontend/expect_async.dart:271: /// See [expectAsync0].
> On 2016/11/21 22:08:02, nweiz wrote:
> > I'd prefer to copy the docs from [expectAsync0] rather than just refer to
> them.
> > I'm willing to make maintenance a little more annoying in exchange for
saving
> > users a click.
> 
> Done.

[floitsch, 2016-11-22 14:49:46]

On 2016/11/22 14:33:38, vsm wrote:
> On 2016/11/22 14:14:27, floitsch wrote:
> > PTAL.
> > 
> > Also, the analyzer is still complaining.
> > @Leaf: do you have any idea why?
> 
> DBC - Leaf is out till next Monday.
> 
> Does it work if you change:
> 
> T max1([a0 = _PLACEHOLDER])
> 
> to:
> 
> T max1([Object a0 = _PLACEHOLDER])
> 
> etc.?
> 
> > 
> > https://codereview.chromium.org/2515303002/diff/40001/CHANGELOG.md
> > File CHANGELOG.md (right):
> > 
> > https://codereview.chromium.org/2515303002/diff/40001/CHANGELOG.md#newcode6
> > CHANGELOG.md:6: * `expectAsyncUntil0`, `expectAsyncUntil1`, ...
> > `expectAsyncUntil6`.
> > On 2016/11/21 22:08:02, nweiz wrote:
> > > It would be nice to have more details here. I'd write:
> > > 
> > > "Deprecate `expectAsync()` and `expectAsyncUntil()`, since they currently
> > can't
> > > be made to work cleanly in strong mode. They're replaced with separate
> methods
> > > for each number of callback arguments:"
> > 
> > Done.
> > 
> >
>
https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
> > File lib/src/frontend/expect_async.dart (right):
> > 
> >
>
https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
> > lib/src/frontend/expect_async.dart:87: (maxExpected == 0 && minExpected > 0)
?
> > minExpected : maxExpected,
> > On 2016/11/21 22:08:02, nweiz wrote:
> > > Nit: Please don't include unrelated whitespace diffs.
> > 
> > I tried, but writing multiline function signatures without invoking the dart
> > formatter turned out to be too difficult.
> > 
> > I undid this change (and will do the same for any other I can see).
> > 
> >
>
https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
> > lib/src/frontend/expect_async.dart:145: T _max0() => _max6();
> > On 2016/11/21 22:08:02, nweiz wrote:
> > > Since these are being accessed outside the class now, make them public.
> > 
> > Done.
> > 
> >
>
https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
> > lib/src/frontend/expect_async.dart:216: /// (by default 1).
> > On 2016/11/21 22:08:02, nweiz wrote:
> > > Remove the old documentation.
> > 
> > Done.
> > 
> >
>
https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
> > lib/src/frontend/expect_async.dart:218: /// This function is deprecated. Use
> > [expectAsync0], [expectAsync1],
> > On 2016/11/21 22:08:02, nweiz wrote:
> > > "is deprecated because it doesn't work well with strong mode."
> > 
> > Done.
> > 
> >
>
https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
> > lib/src/frontend/expect_async.dart:235: @deprecated
> > On 2016/11/21 22:08:02, nweiz wrote:
> > > Use @Deprecated and mention that it will be removed in 0.13.0.
> > 
> > Done.
> > 
> >
>
https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
> > lib/src/frontend/expect_async.dart:259: /// [reason] should be the reason
the
> > callback is expected to be called.
> > On 2016/11/21 22:08:03, nweiz wrote:
> > > Add a paragraph describing the difference between the various expectAsyncN
> > > methods. Maybe:
> > > 
> > > "This takes a callback with zero arguments. See also [expectAsync1],
> > > [expectAsync2], [expectAsync3], [expectAsync4], [expectAsync5], or
> > > [expectAsync6] for callbacks with different numbers of arguments."
> > 
> > Done.
> > 
> >
>
https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
> > lib/src/frontend/expect_async.dart:260: Func0<dynamic/*=T*/ >
> > expectAsync0/*<T>*/(dynamic/*=T*/ callback(),
> > On 2016/11/21 22:08:02, nweiz wrote:
> > > Nit: "<dynamic/*=T*/>"
> > 
> > I don't understand.
> > If it's the spacing: I used the dartformatter.
> > 
> >
>
https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
> > lib/src/frontend/expect_async.dart:263: throw new StateError("expectAsync()
> may
> > only be called within a test.");
> > On 2016/11/21 22:08:02, nweiz wrote:
> > > "expectAsync0()"
> > 
> > Done.
> > 
> >
>
https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
> > lib/src/frontend/expect_async.dart:271: /// See [expectAsync0].
> > On 2016/11/21 22:08:02, nweiz wrote:
> > > I'd prefer to copy the docs from [expectAsync0] rather than just refer to
> > them.
> > > I'm willing to make maintenance a little more annoying in exchange for
> saving
> > > users a click.
> > 
> > Done.

That did it. Thanks.
I don't think they should be necessary, but they aren't hurting either, so I
would just keep them.

[vsm, 2016-11-22 15:25:40]

On 2016/11/22 14:49:46, floitsch wrote:
> On 2016/11/22 14:33:38, vsm wrote:
> > On 2016/11/22 14:14:27, floitsch wrote:
> > > PTAL.
> > > 
> > > Also, the analyzer is still complaining.
> > > @Leaf: do you have any idea why?
> > 
> > DBC - Leaf is out till next Monday.
> > 
> > Does it work if you change:
> > 
> > T max1([a0 = _PLACEHOLDER])
> > 
> > to:
> > 
> > T max1([Object a0 = _PLACEHOLDER])
> > 
> > etc.?
> > 
> > > 
> > > https://codereview.chromium.org/2515303002/diff/40001/CHANGELOG.md
> > > File CHANGELOG.md (right):
> > > 
> > >
https://codereview.chromium.org/2515303002/diff/40001/CHANGELOG.md#newcode6
> > > CHANGELOG.md:6: * `expectAsyncUntil0`, `expectAsyncUntil1`, ...
> > > `expectAsyncUntil6`.
> > > On 2016/11/21 22:08:02, nweiz wrote:
> > > > It would be nice to have more details here. I'd write:
> > > > 
> > > > "Deprecate `expectAsync()` and `expectAsyncUntil()`, since they
currently
> > > can't
> > > > be made to work cleanly in strong mode. They're replaced with separate
> > methods
> > > > for each number of callback arguments:"
> > > 
> > > Done.
> > > 
> > >
> >
>
https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
> > > File lib/src/frontend/expect_async.dart (right):
> > > 
> > >
> >
>
https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
> > > lib/src/frontend/expect_async.dart:87: (maxExpected == 0 && minExpected >
0)
> ?
> > > minExpected : maxExpected,
> > > On 2016/11/21 22:08:02, nweiz wrote:
> > > > Nit: Please don't include unrelated whitespace diffs.
> > > 
> > > I tried, but writing multiline function signatures without invoking the
dart
> > > formatter turned out to be too difficult.
> > > 
> > > I undid this change (and will do the same for any other I can see).
> > > 
> > >
> >
>
https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
> > > lib/src/frontend/expect_async.dart:145: T _max0() => _max6();
> > > On 2016/11/21 22:08:02, nweiz wrote:
> > > > Since these are being accessed outside the class now, make them public.
> > > 
> > > Done.
> > > 
> > >
> >
>
https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
> > > lib/src/frontend/expect_async.dart:216: /// (by default 1).
> > > On 2016/11/21 22:08:02, nweiz wrote:
> > > > Remove the old documentation.
> > > 
> > > Done.
> > > 
> > >
> >
>
https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
> > > lib/src/frontend/expect_async.dart:218: /// This function is deprecated.
Use
> > > [expectAsync0], [expectAsync1],
> > > On 2016/11/21 22:08:02, nweiz wrote:
> > > > "is deprecated because it doesn't work well with strong mode."
> > > 
> > > Done.
> > > 
> > >
> >
>
https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
> > > lib/src/frontend/expect_async.dart:235: @deprecated
> > > On 2016/11/21 22:08:02, nweiz wrote:
> > > > Use @Deprecated and mention that it will be removed in 0.13.0.
> > > 
> > > Done.
> > > 
> > >
> >
>
https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
> > > lib/src/frontend/expect_async.dart:259: /// [reason] should be the reason
> the
> > > callback is expected to be called.
> > > On 2016/11/21 22:08:03, nweiz wrote:
> > > > Add a paragraph describing the difference between the various
expectAsyncN
> > > > methods. Maybe:
> > > > 
> > > > "This takes a callback with zero arguments. See also [expectAsync1],
> > > > [expectAsync2], [expectAsync3], [expectAsync4], [expectAsync5], or
> > > > [expectAsync6] for callbacks with different numbers of arguments."
> > > 
> > > Done.
> > > 
> > >
> >
>
https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
> > > lib/src/frontend/expect_async.dart:260: Func0<dynamic/*=T*/ >
> > > expectAsync0/*<T>*/(dynamic/*=T*/ callback(),
> > > On 2016/11/21 22:08:02, nweiz wrote:
> > > > Nit: "<dynamic/*=T*/>"
> > > 
> > > I don't understand.
> > > If it's the spacing: I used the dartformatter.
> > > 
> > >
> >
>
https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
> > > lib/src/frontend/expect_async.dart:263: throw new
StateError("expectAsync()
> > may
> > > only be called within a test.");
> > > On 2016/11/21 22:08:02, nweiz wrote:
> > > > "expectAsync0()"
> > > 
> > > Done.
> > > 
> > >
> >
>
https://codereview.chromium.org/2515303002/diff/40001/lib/src/frontend/expect...
> > > lib/src/frontend/expect_async.dart:271: /// See [expectAsync0].
> > > On 2016/11/21 22:08:02, nweiz wrote:
> > > > I'd prefer to copy the docs from [expectAsync0] rather than just refer
to
> > > them.
> > > > I'm willing to make maintenance a little more annoying in exchange for
> > saving
> > > > users a click.
> > > 
> > > Done.
> 
> That did it. Thanks.
> I don't think they should be necessary, but they aren't hurting either, so I
> would just keep them.

Do you want to file a tracking bug?  At a minimum, we could give a more
understandable / actionable error.

[nweiz, 2016-11-23 21:37:19]

LGTM!

[floitsch, 2016-11-24 10:47:47]

Description was changed from

==========
Add expectAsyncX and expectAsyncUntilX methods, and deprecate the old methods.

Fixes #436.
==========

to

==========
Add expectAsyncX and expectAsyncUntilX methods, and deprecate the old methods.

Fixes #436.

R=nweiz@google.com

Committed:
https://github.com/dart-lang/test/commit/7babd03969368b21e51de65a3e84d43925f6...
==========

[floitsch, 2016-11-24 10:47:48]

Committed patchset #5 (id:80001) manually as
7babd03969368b21e51de65a3e84d43925f6e009 (presubmit successful).