Remove unused includes from core/inspector, core/loader and core/page

Remove unused includes from core/inspector, core/loader and core/page

Continue removing unused includes as suggested by include-what-you-use.
I've manually looked through the changes here so all the removals should
be actually unused. As usual, if a .cpp only needs a fwd decl of a type
used in a method signature, I assume that the header would already have
included or declared the type and remove the include from the .cpp.

BUG=242216
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=153301

[pdr., 2013-06-24 14:33:05]

On 2013/06/24 11:54:13, Timothy Loh wrote:

rs=me

LGTM!

[yurys, 2013-06-24 15:00:44]

https://codereview.chromium.org/17569013/diff/1/Source/core/inspector/HeapGra...
File Source/core/inspector/HeapGraphSerializer.cpp (left):

https://codereview.chromium.org/17569013/diff/1/Source/core/inspector/HeapGra...
Source/core/inspector/HeapGraphSerializer.cpp:38: #include
"wtf/MemoryInstrumentationHashMap.h"
Please revert these two includes. Removing them would break memory reporting for
m_address2NodeIdMap and m_roots fields as they would be considered as
not-instrumented.

https://codereview.chromium.org/17569013/diff/1/Source/core/inspector/Injecte...
File Source/core/inspector/InjectedScript.h (right):

https://codereview.chromium.org/17569013/diff/1/Source/core/inspector/Injecte...
Source/core/inspector/InjectedScript.h:40: #include "wtf/PassRefPtr.h"
Why not remove PassRefPtr as well? Forward declaration should be enough to
declare return types as PassRefPtr.

https://codereview.chromium.org/17569013/diff/1/Source/core/inspector/Inspect...
File Source/core/inspector/InspectorController.cpp (left):

https://codereview.chromium.org/17569013/diff/1/Source/core/inspector/Inspect...
Source/core/inspector/InspectorController.cpp:73: #include
<wtf/MemoryInstrumentationVector.h>
This one should be preserved to provide access to the Vector<> memory
instrumentation (used in reportMemoryUsage) method.

https://codereview.chromium.org/17569013/diff/1/Source/core/inspector/Inspect...
File Source/core/inspector/InspectorDOMStorageAgent.cpp (left):

https://codereview.chromium.org/17569013/diff/1/Source/core/inspector/Inspect...
Source/core/inspector/InspectorDOMStorageAgent.cpp:51: #include
<wtf/MemoryInstrumentationHashMap.h>
Please return MemoryInstrumentationHashMap.h back for the same reasons as above.
In general you shouldn't remove MemoryInstrumentation*.h includes as they are
used to calculate WTF containers size in reportMemoryUsage methods.

https://codereview.chromium.org/17569013/diff/1/Source/core/inspector/MemoryI...
File Source/core/inspector/MemoryInstrumentationImpl.cpp (left):

https://codereview.chromium.org/17569013/diff/1/Source/core/inspector/MemoryI...
Source/core/inspector/MemoryInstrumentationImpl.cpp:38: #include
<wtf/MemoryInstrumentationHashSet.h>
ditto. It used to estimate size of m_allocatedObjects

[darktears, 2013-06-24 15:25:01]

On 2013/06/24 15:00:44, Yury Semikhatsky wrote:
>
https://codereview.chromium.org/17569013/diff/1/Source/core/inspector/HeapGra...
> File Source/core/inspector/HeapGraphSerializer.cpp (left):
> 
>
https://codereview.chromium.org/17569013/diff/1/Source/core/inspector/HeapGra...
> Source/core/inspector/HeapGraphSerializer.cpp:38: #include
> "wtf/MemoryInstrumentationHashMap.h"
> Please revert these two includes. Removing them would break memory reporting
for
> m_address2NodeIdMap and m_roots fields as they would be considered as
> not-instrumented.
> 
>
https://codereview.chromium.org/17569013/diff/1/Source/core/inspector/Injecte...
> File Source/core/inspector/InjectedScript.h (right):
> 
>
https://codereview.chromium.org/17569013/diff/1/Source/core/inspector/Injecte...
> Source/core/inspector/InjectedScript.h:40: #include "wtf/PassRefPtr.h"
> Why not remove PassRefPtr as well? Forward declaration should be enough to
> declare return types as PassRefPtr.
> 
>
https://codereview.chromium.org/17569013/diff/1/Source/core/inspector/Inspect...
> File Source/core/inspector/InspectorController.cpp (left):
> 
>
https://codereview.chromium.org/17569013/diff/1/Source/core/inspector/Inspect...
> Source/core/inspector/InspectorController.cpp:73: #include
> <wtf/MemoryInstrumentationVector.h>
> This one should be preserved to provide access to the Vector<> memory
> instrumentation (used in reportMemoryUsage) method.
> 
>
https://codereview.chromium.org/17569013/diff/1/Source/core/inspector/Inspect...
> File Source/core/inspector/InspectorDOMStorageAgent.cpp (left):
> 
>
https://codereview.chromium.org/17569013/diff/1/Source/core/inspector/Inspect...
> Source/core/inspector/InspectorDOMStorageAgent.cpp:51: #include
> <wtf/MemoryInstrumentationHashMap.h>
> Please return MemoryInstrumentationHashMap.h back for the same reasons as
above.
> In general you shouldn't remove MemoryInstrumentation*.h includes as they are
> used to calculate WTF containers size in reportMemoryUsage methods.
> 
>
https://codereview.chromium.org/17569013/diff/1/Source/core/inspector/MemoryI...
> File Source/core/inspector/MemoryInstrumentationImpl.cpp (left):
> 
>
https://codereview.chromium.org/17569013/diff/1/Source/core/inspector/MemoryI...
> Source/core/inspector/MemoryInstrumentationImpl.cpp:38: #include
> <wtf/MemoryInstrumentationHashSet.h>
> ditto. It used to estimate size of m_allocatedObjects

My 2 cents. I'm not at all familiar with the memory instrumentation code but it
should not compile when you remove these includes (I'm assuming that the CL at
least built). Silently breaking the
code because you removed an include is most likely calling for future mistakes
that we won't be able to catch unless someone like Yury stops by. We really
should enforce a compilation failure if by mistake these includes are removed.

[yurys, 2013-06-24 15:50:49]

On 2013/06/24 15:25:01, darktears wrote:
> 
> My 2 cents. I'm not at all familiar with the memory instrumentation code but
it
> should not compile when you remove these includes (I'm assuming that the CL at
> least built). Silently breaking the
> code because you removed an include is most likely calling for future mistakes
> that we won't be able to catch unless someone like Yury stops by. We really
> should enforce a compilation failure if by mistake these includes are removed.

As instrumentation couldn't be added to all existing classes the instrumentation
works in a way that allows to precisely estimate memory usage for the
instrumented objects and use simple sizeof() for the rest. So if no
instrumentation is available for a class the compilation won't break but only
part of the object size may be counted. There is a few layout tests for the
memory instrumentation which should catch some of the problems, but they do not
cover all parts.

Putting the instrumentation directly into WTF collections would eliminate the
problem. Before Blink fork  we encountered a strong resistance from Maciej who
insisted on considering WTF as it were STL which we could not modify and not
adding invasive instrumentation into the collections. So we ended up
implementing it externally. Now that this code is in Blink we could move the
instrumentation into the containers and get rid of separate
MemoryInstrumentation*.h headers. At the moment we are considering alternative
solutions that would be easier to maintain and until we agree on how to proceed
I'd rather leave the instrumentation as is.

[yurys, 2013-06-24 16:38:20]

On 2013/06/24 15:25:01, darktears wrote:
> 
> My 2 cents. I'm not at all familiar with the memory instrumentation code but
it
> should not compile when you remove these includes (I'm assuming that the CL at
> least built). Silently breaking the
> code because you removed an include is most likely calling for future mistakes
> that we won't be able to catch unless someone like Yury stops by. We really
> should enforce a compilation failure if by mistake these includes are removed.

Actually there should be a link-time error if there is a WTF collection member
but corresponding MemoryInstrumentation<Collection>.h is not included. It seems
that it didn't happen on this patch because of some transitive dependencies.

[Timothy Loh, 2013-06-24 17:22:52]

On 2013/06/24 16:38:20, Yury Semikhatsky wrote:
> On 2013/06/24 15:25:01, darktears wrote:
> > 
> > My 2 cents. I'm not at all familiar with the memory instrumentation code but
> it
> > should not compile when you remove these includes (I'm assuming that the CL
at
> > least built). Silently breaking the
> > code because you removed an include is most likely calling for future
mistakes
> > that we won't be able to catch unless someone like Yury stops by. We really
> > should enforce a compilation failure if by mistake these includes are
removed.
> 
> Actually there should be a link-time error if there is a WTF collection member
> but corresponding MemoryInstrumentation<Collection>.h is not included. It
seems
> that it didn't happen on this patch because of some transitive dependencies.

In any case, I've added these back for now and will be more careful about
removing
these includes in the future. Thanks for pointing this out.

As for the PassRefPtr that wasn't removed, include-what-you-use doesn't quite
like all the using directives we use in wtf/, so I was going to leave the wtf/
includes
that it didn't pick up for another patch. I've removed this particular include
anyway.

[commit-bot: I haz the power, 2013-06-28 05:29:55]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/timloh@chromium.org/17569013/18001

[commit-bot: I haz the power, 2013-06-28 05:51:26]

Sorry for I got bad news for ya.
Compile failed with a clobber build on mac_layout.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=mac_layout...
Your code is likely broken or HEAD is junk. Please ensure your
code is not broken then alert the build sheriffs.
Look at the try server FAQ for more details.

[commit-bot: I haz the power, 2013-06-28 06:05:00]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/timloh@chromium.org/17569013/46002

[commit-bot: I haz the power, 2013-06-28 06:43:05]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/timloh@chromium.org/17569013/40004

[commit-bot: I haz the power, 2013-06-28 08:11:53]

Failed to apply patch for Source/core/loader/DocumentLoader.cpp:
While running patch -p1 --forward --force --no-backup-if-mismatch;
  patching file Source/core/loader/DocumentLoader.cpp
  Hunk #1 FAILED at 30.
  Hunk #2 FAILED at 63.
  2 out of 2 hunks FAILED -- saving rejects to file
Source/core/loader/DocumentLoader.cpp.rej

Patch:       Source/core/loader/DocumentLoader.cpp
Index: Source/core/loader/DocumentLoader.cpp
diff --git a/Source/core/loader/DocumentLoader.cpp
b/Source/core/loader/DocumentLoader.cpp
index
cf39671beba238041579c99e93cac45c02d99849..cf17c906e4b9929a046b7bf64d569aeb45a2adb0
100644
--- a/Source/core/loader/DocumentLoader.cpp
+++ b/Source/core/loader/DocumentLoader.cpp
@@ -30,23 +30,13 @@
 #include "config.h"
 #include "core/loader/DocumentLoader.h"
 
-#include <wtf/Assertions.h>
-#include <wtf/MemoryInstrumentationHashMap.h>
-#include <wtf/MemoryInstrumentationHashSet.h>
-#include <wtf/MemoryInstrumentationVector.h>
-#include <wtf/text/CString.h>
-#include <wtf/text/WTFString.h>
-#include <wtf/unicode/Unicode.h>
 #include "core/dom/Document.h"
 #include "core/dom/DocumentParser.h"
 #include "core/dom/Event.h"
 #include "core/dom/WebCoreMemoryInstrumentation.h"
-#include "core/history/HistoryItem.h"
-#include "core/html/HTMLFormElement.h"
 #include "core/html/HTMLFrameOwnerElement.h"
 #include "core/inspector/InspectorInstrumentation.h"
 #include "core/loader/DocumentWriter.h"
-#include "core/loader/FormState.h"
 #include "core/loader/FrameLoader.h"
 #include "core/loader/FrameLoaderClient.h"
 #include "core/loader/ResourceLoader.h"
@@ -63,9 +53,12 @@
 #include "core/page/FrameTree.h"
 #include "core/page/Page.h"
 #include "core/page/Settings.h"
-#include "core/platform/Logging.h"
 #include "weborigin/SchemeRegistry.h"
-#include "weborigin/SecurityPolicy.h"
+#include "wtf/Assertions.h"
+#include "wtf/MemoryInstrumentationHashMap.h"
+#include "wtf/MemoryInstrumentationHashSet.h"
+#include "wtf/MemoryInstrumentationVector.h"
+#include "wtf/text/WTFString.h"
 
 namespace WebCore {

[commit-bot: I haz the power, 2013-06-28 15:18:29]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/timloh@chromium.org/17569013/54002

[commit-bot: I haz the power, 2013-06-28 17:03:09]

Failed to apply patch for Source/core/loader/DocumentLoader.cpp:
While running patch -p1 --forward --force --no-backup-if-mismatch;
  patching file Source/core/loader/DocumentLoader.cpp
  Hunk #1 succeeded at 41 (offset 6 lines).
  Hunk #2 FAILED at 55.
  1 out of 2 hunks FAILED -- saving rejects to file
Source/core/loader/DocumentLoader.cpp.rej

Patch:       Source/core/loader/DocumentLoader.cpp
Index: Source/core/loader/DocumentLoader.cpp
diff --git a/Source/core/loader/DocumentLoader.cpp
b/Source/core/loader/DocumentLoader.cpp
index
a0fce6d46eb6eed46b7cf3b3b0f7a8891677cd91..af439ea49c378fc8b036fd491e1d6cff55b2f8e0
100644
--- a/Source/core/loader/DocumentLoader.cpp
+++ b/Source/core/loader/DocumentLoader.cpp
@@ -35,12 +35,9 @@
 #include "core/dom/DocumentParser.h"
 #include "core/dom/Event.h"
 #include "core/dom/WebCoreMemoryInstrumentation.h"
-#include "core/history/HistoryItem.h"
-#include "core/html/HTMLFormElement.h"
 #include "core/html/HTMLFrameOwnerElement.h"
 #include "core/inspector/InspectorInstrumentation.h"
 #include "core/loader/DocumentWriter.h"
-#include "core/loader/FormState.h"
 #include "core/loader/FrameLoader.h"
 #include "core/loader/FrameLoaderClient.h"
 #include "core/loader/ResourceLoader.h"
@@ -58,16 +55,12 @@
 #include "core/page/FrameTree.h"
 #include "core/page/Page.h"
 #include "core/page/Settings.h"
-#include "core/platform/Logging.h"
 #include "weborigin/SchemeRegistry.h"
-#include "weborigin/SecurityPolicy.h"
 #include "wtf/Assertions.h"
 #include "wtf/MemoryInstrumentationHashMap.h"
 #include "wtf/MemoryInstrumentationHashSet.h"
 #include "wtf/MemoryInstrumentationVector.h"
-#include "wtf/text/CString.h"
 #include "wtf/text/WTFString.h"
-#include "wtf/unicode/Unicode.h"
 
 namespace WebCore {

[commit-bot: I haz the power, 2013-06-29 13:18:31]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/timloh@chromium.org/17569013/65001

[commit-bot: I haz the power, 2013-06-29 14:39:12]

Change committed as 153301