Fix ImageDecoder::frameIsCompleteAtIndex - fully received instead of decoded.

third_party/WebKit/Source/platform/image-decoders/bmp/BMPImageReader.h

Index: third_party/WebKit/Source/platform/image-decoders/bmp/BMPImageReader.h
diff --git a/third_party/WebKit/Source/platform/image-decoders/bmp/BMPImageReader.h b/third_party/WebKit/Source/platform/image-decoders/bmp/BMPImageReader.h
index d150a56bb5a874a98d140a44c218e9b5d9333258..efe4dbb0f853a58695cf392bc6cc597f18ffefdf 100644
--- a/third_party/WebKit/Source/platform/image-decoders/bmp/BMPImageReader.h
+++ b/third_party/WebKit/Source/platform/image-decoders/bmp/BMPImageReader.h
@@ -355,6 +355,11 @@ private:
     bool m_seenNonZeroAlphaPixel;
     bool m_seenZeroAlphaPixel;
 
+    // Variables m_seenNonZeroAlphaPixel and m_seenZeroAlphaPixel are tracking
+    // values used for alpha channel.  m_hasPixelsWithAlpha tracks if decoding
+    // saw pixels that are transparent.

[scroggo_chromium, 2016/05/16 20:32:08]

I think this comment is confusing. It describes the variables above (although
better explanations exist elsewhere) and then partially describes
m_hasPixelsWithAlpha.

I think it would be clearer to drop the first sentence and instead say something
like:

// Whether this image is considered to have transparent pixels, after
// taking m_seenNonZeroAlphaPixel and m_seenZeroAlphaPixel into
// consideration.

[aleksandar.stojiljkovic, 2016/05/22 15:41:53]

reverted the code.

+    bool m_hasPixelsWithAlpha;
+
     // BMPs-in-ICOs have a few differences from standalone BMPs, so we need to
     // know if we're in an ICO container.
     bool m_isInICO;