Remove old C++03 move emulation code.

base/files/file.cc

Index: base/files/file.cc
diff --git a/base/files/file.cc b/base/files/file.cc
index 47b9f88f1888d1d3cf4b70070b460bd8e8880289..036d43b33c50d6efd816d64005c6a18ce4ab34e9 100644
--- a/base/files/file.cc
+++ b/base/files/file.cc
@@ -50,13 +50,12 @@ File::File(Error error_details)
       async_(false) {
 }
 
-File::File(RValue other)
-    : file_(other.object->TakePlatformFile()),
-      tracing_path_(other.object->tracing_path_),
-      error_details_(other.object->error_details()),
-      created_(other.object->created()),
-      async_(other.object->async_) {
-}
+File::File(File&& other)
+    : file_(other.TakePlatformFile()),
+      tracing_path_(other.tracing_path_),
+      error_details_(other.error_details()),
+      created_(other.created()),
+      async_(other.async_) {}
 
 File::~File() {
   // Go through the AssertIOAllowed logic.
@@ -72,15 +71,14 @@ File File::CreateForAsyncHandle(PlatformFile platform_file) {
   return file.Pass();
 }
 
-File& File::operator=(RValue other) {
-  if (this != other.object) {

[dcheng, 2015/10/13 21:03:09]

As far as I can tell, these checks were all defensive. I've changed them all to
DCHECK for now.

I've seen arguments both for and against a self-move check in the move
assignment operator. I'm open to either way, but it might be nice to have some
sort of convention about this in Chrome.

-    Close();
-    SetPlatformFile(other.object->TakePlatformFile());
-    tracing_path_ = other.object->tracing_path_;
-    error_details_ = other.object->error_details();
-    created_ = other.object->created();
-    async_ = other.object->async_;
-  }
+File& File::operator=(File&& other) {
+  DCHECK_NE(this, &other);
+  Close();

[danakj, 2015/10/15 23:35:05]

This close would be destructive if this == &other. So it's good to that the
DCHECK. Assuming it doesn't actually happen, then we should early out. :)
Without destructive behaviour like this (which I expect to be rare), DCHECK or
early out both seem overly scared and we can just ignore that and move things
from other to this, IMO.

+  SetPlatformFile(other.TakePlatformFile());
+  tracing_path_ = other.tracing_path_;

[danakj, 2015/10/15 23:35:05]

we should probably be move()ing all the arguments in a move assignment. TODO?

[dcheng, 2015/10/16 00:40:01]

It seems weird to std::move() an integral type.

FilePath isn't moveable today and it might be in the future, but this seems like
it'd be a lot of TODOs to add. Maybe this should just be something that a
diagnostic could detect? "Field with moveable type copied instead of moved in
move constructor".

[danakj, 2015/10/16 00:44:24]

It's okay if it's not moveable, the code would work fine if it did. move() casts
to an rvalue but isn't guaranteed to not copy. For ints ya prob doesnt make
sense.

+  error_details_ = other.error_details();
+  created_ = other.created();
+  async_ = other.async_;
   return *this;
 }