Bug 51551 – [chromium] warnings when building on Windows

• Status: RESOLVED
• Resolution: WONTFIX
• Component: WebKit Misc.

[Evan Martin]

We still get a few warnings when we build on Windows.  We should fix these, and then make warnings into errors so that they don't regress.

Here is a full list from a recent run:

e:\b\build\slave\win\build\src\third_party\WebKit\Tools\DumpRenderTree\TestNetscapePlugIn\Tests\NPRuntimeRemoveProperty.cpp(66) : warning C4244: 'initializing' : conversion from 'const double' to 'int32_t', possible loss of data

e:\b\build\slave\win\build\src\third_party\webkit\webcore\dom\ViewportArguments.h(78) : warning C4305: 'initializing' : truncation from '' to 'bool'
(this is bug 50982 and the code is truly wrong, possibly causing real problems)

e:\b\build\slave\win\build\src\third_party\WebKit\Tools\DumpRenderTree\chromium\LayoutTestController.cpp(1433) : warning C4244: 'argument' : conversion from 'float' to 'int', possible loss of data

[Evan Martin]

Oops, left one out:
..\html\TextMetrics.h(39) : warning C4244: 'return' : conversion from 'const float' to 'unsigned int', possible loss of data

[Evan Martin]

James, I'm trying to fix all the Windows build warnings.
One of the last one is near a comment you added:

+    // FIXME: Is coercing to an integer value intentional? If so, there should be a static_cast<unsigned>() here.
     unsigned width() const { return m_width; }

Did you ever track down that issue?

[Evan Martin]

(see bug 38073 for more context)

[Evan Martin]

CC original author of the mystery code.

[Evan Martin]

Oops, sorry hyatt, wrong bug.  :(