Make import failures log errors

Make import failures log errors

BUG=
R=eseidel@chromium.org

Committed: https://chromium.googlesource.com/external/mojo/+/ed3e4e60bed67985d1c49257a5c61354fc3a27f3

[eseidel, 2014-11-06 23:08:51]

eseidel@chromium.org changed reviewers:
+ eseidel@chromium.org

[eseidel, 2014-11-06 23:08:52]

https://codereview.chromium.org/711513003/diff/40001/sky/engine/core/frame/Fr...
File sky/engine/core/frame/FrameConsole.cpp (left):

https://codereview.chromium.org/711513003/diff/40001/sky/engine/core/frame/Fr...
sky/engine/core/frame/FrameConsole.cpp:99: if (consoleMessage->source() ==
CSSMessageSource || consoleMessage->source() == NetworkMessageSource)
Why did you need to remove this?

[rafaelw, 2014-11-06 23:10:45]

https://codereview.chromium.org/711513003/diff/40001/sky/engine/core/frame/Fr...
File sky/engine/core/frame/FrameConsole.cpp (left):

https://codereview.chromium.org/711513003/diff/40001/sky/engine/core/frame/Fr...
sky/engine/core/frame/FrameConsole.cpp:99: if (consoleMessage->source() ==
CSSMessageSource || consoleMessage->source() == NetworkMessageSource)
Loggin to stddout happens the call below to .chromeClient().addMessageToConsole.
I suspect that DevTools uses the messageStorage() above.

On 2014/11/06 23:08:52, eseidel wrote:
> Why did you need to remove this?

[rafaelw, 2014-11-06 23:14:11]

https://codereview.chromium.org/711513003/diff/40001/sky/engine/core/frame/Fr...
File sky/engine/core/frame/FrameConsole.cpp (left):

https://codereview.chromium.org/711513003/diff/40001/sky/engine/core/frame/Fr...
sky/engine/core/frame/FrameConsole.cpp:99: if (consoleMessage->source() ==
CSSMessageSource || consoleMessage->source() == NetworkMessageSource)
On 2014/11/06 23:10:45, rafaelw wrote:
> Loggin to stddout happens the call below to
.chromeClient().addMessageToConsole.
> I suspect that DevTools uses the messageStorage() above.

Note: I didn't fully follow the path that Blink/Devtools takes. In blink, import
failures do long to the console (as well as the network tab) despite this return
for CSS & Network types - which made me assure they were arriving their somehow
via the messageStorage.

> 
> On 2014/11/06 23:08:52, eseidel wrote:
> > Why did you need to remove this?
>

[eseidel, 2014-11-06 23:19:52]

lgtm

[abarth-chromium, 2014-11-06 23:24:05]

abarth@chromium.org changed reviewers:
+ abarth@chromium.org

[abarth-chromium, 2014-11-06 23:24:05]

Yeah, I changed the way we issue the network requests for imports so that we can
feed the data directly to the parser thread.  Basically, we bypass several
layers of abstraction and just talk directly to Mojo's network_service.  I
suspect we lost the console logging in cutting through those layers.

[rafaelw, 2014-11-06 23:47:11]

Committed patchset #4 (id:60001) manually as
ed3e4e60bed67985d1c49257a5c61354fc3a27f3 (presubmit successful).