some fixes for --strong warnings

some fixes for --strong warnings
I didn't go all the way, in case we want to change some of these. Thoughts?

R=leafp@google.com

Committed: https://github.com/dart-lang/dev_compiler/commit/47adda72e4c60079b144dd573c39c20260fd0c31

[Jennifer Messerly, 2015-07-20 22:27:50]

jmesserly@google.com changed reviewers:
+ vsm@google.com

[Jennifer Messerly, 2015-07-22 21:51:19]

jmesserly@google.com changed reviewers:
+ leafp@google.com
- vsm@google.com

[Jennifer Messerly, 2015-07-22 21:51:19]

Leaf, would you mind taking a look at these? Do they match your expectations?

[Leaf, 2015-07-22 22:29:15]

lgtm

https://codereview.chromium.org/1245013002/diff/1/lib/src/codegen/code_genera...
File lib/src/codegen/code_generator.dart (right):

https://codereview.chromium.org/1245013002/diff/1/lib/src/codegen/code_genera...
lib/src/codegen/code_generator.dart:54: }() as List<String>;
This shouldn't be necessary, but currently is.  I need to fix this.

https://codereview.chromium.org/1245013002/diff/1/lib/src/codegen/js_codegen....
File lib/src/codegen/js_codegen.dart (right):

https://codereview.chromium.org/1245013002/diff/1/lib/src/codegen/js_codegen....
lib/src/codegen/js_codegen.dart:412: return _statement(<JS.Statement>[result,
copyMembers]);
Is this necessary?  I would have expected your changes to the argument type of
_statement to allow inference to fill this in.

https://codereview.chromium.org/1245013002/diff/1/lib/src/server/dependency_g...
File lib/src/server/dependency_graph.dart (right):

https://codereview.chromium.org/1245013002/diff/1/lib/src/server/dependency_g...
lib/src/server/dependency_graph.dart:228:
graph.nodeFromUri(uri.resolve(tag.attributes['href']));
Is there a rationale behind using the T v = e pattern in some places and the var
v = e as T pattern in others?

[Jennifer Messerly, 2015-07-22 22:35:52]

thanks!

https://codereview.chromium.org/1245013002/diff/1/lib/src/codegen/js_codegen....
File lib/src/codegen/js_codegen.dart (right):

https://codereview.chromium.org/1245013002/diff/1/lib/src/codegen/js_codegen....
lib/src/codegen/js_codegen.dart:412: return _statement(<JS.Statement>[result,
copyMembers]);
On 2015/07/22 22:29:15, Leaf wrote:
> Is this necessary?  I would have expected your changes to the argument type of
> _statement to allow inference to fill this in.  

Done. Reverted this.

https://codereview.chromium.org/1245013002/diff/1/lib/src/server/dependency_g...
File lib/src/server/dependency_graph.dart (right):

https://codereview.chromium.org/1245013002/diff/1/lib/src/server/dependency_g...
lib/src/server/dependency_graph.dart:228:
graph.nodeFromUri(uri.resolve(tag.attributes['href']));
On 2015/07/22 22:29:15, Leaf wrote:
> Is there a rationale behind using the T v = e pattern in some places and the
var
> v = e as T pattern in others?

I have no idea ... the messages seemed inconsistent :|
here, I think it complained about htmlResourceNodes requiring
ResourceSourceNode.
Somewhere else, it forced me to switch a `T v = e` to `var e as T` to get rid of
a warning ...

[Jennifer Messerly, 2015-07-22 22:37:29]

https://codereview.chromium.org/1245013002/diff/20001/lib/src/utils.dart
File lib/src/utils.dart (right):

https://codereview.chromium.org/1245013002/diff/20001/lib/src/utils.dart#newc...
lib/src/utils.dart:385: var map = _objectMap[typeProvider] as Map<String,
DartType>;
for some reason, this was a warning the old way. But not this way. No idea why
...

[Jennifer Messerly, 2015-07-22 22:37:58]

Committed patchset #2 (id:20001) manually as
47adda72e4c60079b144dd573c39c20260fd0c31 (presubmit successful).