Send notificatinos after no-op changes.

Send notificatinos after no-op changes.

This should solve the problem when client remove some part of the code,
so removes also error markers for it, and then quickly restores it,
so for the server it looks as an empty/no-op change. But information is
lost, and the server should resend it to the client.

R=brianwilkerson@google.com
BUG=

Committed: https://code.google.com/p/dart/source/detail?r=43521

[Brian Wilkerson, 2015-02-05 20:07:59]

LGTM

https://codereview.chromium.org/899753004/diff/1/pkg/analyzer/lib/src/generat...
File pkg/analyzer/lib/src/generated/engine.dart (right):

https://codereview.chromium.org/899753004/diff/1/pkg/analyzer/lib/src/generat...
pkg/analyzer/lib/src/generated/engine.dart:1883: _workManager.add(source,
SourcePriority.UNKNOWN);
If the state was INVALID (or anything other than VALID or FLUSHED, which are
already handled), will this schedule the work a second time?

[scheglov, 2015-02-05 20:41:41]

New patchsets have been uploaded after l-g-t-m from brianwilkerson@google.com

[scheglov, 2015-02-05 20:42:03]

https://codereview.chromium.org/899753004/diff/1/pkg/analyzer/lib/src/generat...
File pkg/analyzer/lib/src/generated/engine.dart (right):

https://codereview.chromium.org/899753004/diff/1/pkg/analyzer/lib/src/generat...
pkg/analyzer/lib/src/generated/engine.dart:1883: _workManager.add(source,
SourcePriority.UNKNOWN);
On 2015/02/05 20:07:59, Brian Wilkerson wrote:
> If the state was INVALID (or anything other than VALID or FLUSHED, which are
> already handled), will this schedule the work a second time?

Done.

[scheglov, 2015-02-05 20:42:26]

Committed patchset #2 (id:20001) manually as 43521 (presubmit successful).