Get rid of leaks in SkCodec::NewFromStream.

src/codec/SkCodec_libbmp.cpp

Index: src/codec/SkCodec_libbmp.cpp
diff --git a/src/codec/SkCodec_libbmp.cpp b/src/codec/SkCodec_libbmp.cpp
index 75c5715bda1df787b6e503c3346a1a172590bde7..3d2391b1703ef44222bdf4b5dc8d4c48fcad7c48 100644
--- a/src/codec/SkCodec_libbmp.cpp
+++ b/src/codec/SkCodec_libbmp.cpp
@@ -496,8 +496,13 @@ bool SkBmpCodec::ReadHeader(SkStream* stream, bool isIco, SkCodec** codecOut) {
  *
  */
 SkCodec* SkBmpCodec::NewFromStream(SkStream* stream, bool isIco) {
+    SkAutoTDelete<SkStream> streamDeleter(stream);
     SkCodec* codec = NULL;
     if (ReadHeader(stream, isIco, &codec)) {

[mtklein, 2015/04/03 13:13:05]

Might be less error prone to have ReadHeader take ownership too:

if (ReadHeader(stream.detach(), isIco, &codec)) {
  return codec;
}

[scroggo, 2015/04/03 13:59:53]

Yeah, I was not sure the best way to handle this. The problem with that approach
is that we call ReadHeader in two places:

This time, it make sense for ReadHeader to take ownership and pass to the codec
if it creates one.

The other time, after a rewind, we already have a codec, and it already owns the
stream. In this case we pass NULL for codec.

We could make ReadHeader behave differently with regards to taking ownership
depending on whether a codec pointer is passed in, but that seems ugly to me.
(Maybe no worse than this decision though?)

+        // codec has taken ownership of stream, so we do not need to
+        // delete it.
+        SkASSERT(codec);
+        streamDeleter.detach();
         return codec;
     }
     return NULL;