Quota support.

Add quota support.

Add a 5mb quota to DOM Storage.  Most of the details are in an upstream patch (https://bugs.webkit.org/show_bug.cgi?id=29991 ).

BUG=16876
TEST=I added a layout test upstream and it works.


Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=27942

[michaeln, 2009-10-02 01:11:32]

thnx for the link to the webkit side... (Ja, I see It!)

[darin (slow to review), 2009-10-02 02:59:27]

http://codereview.chromium.org/255050/diff/4001/5004
File webkit/api/public/WebStorageNamespace.h (right):

http://codereview.chromium.org/255050/diff/4001/5004#newcode52
Line 52: static const unsigned kNoQuota = UINT_MAX;
nit: noQuota per webkit style.

http://codereview.chromium.org/255050/diff/4001/5006
File webkit/glue/webkitclient_impl.h (right):

http://codereview.chromium.org/255050/diff/4001/5006#newcode73
Line 73: virtual void setCursorForPlugin(const WebKit::WebCursorInfo&,
nitty-nit: this file is pretty uniform about wrapping the parameters by breaking
after the "("

[michaeln, 2009-10-02 07:43:08]

http://codereview.chromium.org/255050/diff/4001/5011
File chrome/browser/in_process_webkit/storage_namespace.h (right):

http://codereview.chromium.org/255050/diff/4001/5011#newcode58
Line 58: // The quota for each storage area.  Suggested by the spec.
nano nit: one space after the period

http://codereview.chromium.org/255050/diff/4001/5002
File webkit/api/src/WebStorageNamespaceImpl.cpp (right):

http://codereview.chromium.org/255050/diff/4001/5002#newcode38
Line 38: #include "WebKitClient.h"
why do we need the new include?

http://codereview.chromium.org/255050/diff/4001/5006
File webkit/glue/webkitclient_impl.h (right):

http://codereview.chromium.org/255050/diff/4001/5006#newcode89
Line 89: virtual WebKit::WebStorageNamespace* createSessionStorageNamespace();
I see there is noQuota for session storage? Should there be? Data will linger
for as long as a tab is open. If there are no limits for session storage, seems
like that could be a surface area for malice. The usual recourse of hitting the
back button on a misbehaving page won't remedy sessionStorage maliciousness.

[jorlow, 2009-10-02 22:30:41]

Oh...forgot to send this out... Michael: this is why I was talking as if I had
already responded to your question.  :-)

http://codereview.chromium.org/255050/diff/4001/5004
File webkit/api/public/WebStorageNamespace.h (right):

http://codereview.chromium.org/255050/diff/4001/5004#newcode52
Line 52: static const unsigned kNoQuota = UINT_MAX;
On 2009/10/02 02:59:27, darin wrote:
> nit: noQuota per webkit style.

Done.

http://codereview.chromium.org/255050/diff/4001/5002
File webkit/api/src/WebStorageNamespaceImpl.cpp (right):

http://codereview.chromium.org/255050/diff/4001/5002#newcode38
Line 38: #include "WebKitClient.h"
On 2009/10/02 07:43:08, michaeln wrote:
> why do we need the new include?

Historical.  Deleted.

http://codereview.chromium.org/255050/diff/4001/5006
File webkit/glue/webkitclient_impl.h (right):

http://codereview.chromium.org/255050/diff/4001/5006#newcode73
Line 73: virtual void setCursorForPlugin(const WebKit::WebCursorInfo&,
On 2009/10/02 02:59:27, darin wrote:
> nitty-nit: this file is pretty uniform about wrapping the parameters by
breaking
> after the "("

Done.

http://codereview.chromium.org/255050/diff/4001/5006#newcode89
Line 89: virtual WebKit::WebStorageNamespace* createSessionStorageNamespace();
On 2009/10/02 07:43:08, michaeln wrote:
> I see there is noQuota for session storage? Should there be? Data will linger
> for as long as a tab is open. If there are no limits for session storage,
seems
> like that could be a surface area for malice. The usual recourse of hitting
the
> back button on a misbehaving page won't remedy sessionStorage maliciousness.

Hmmm....interesting point.  I'm going to hold off for now, but I'll keep this in
mind when I'm looking at enabling SessionStorage.  I'll probably also run it
past the security guys and/or WhatWG.

[jorlow, 2009-10-02 23:05:08]

Posted a new version.

The upstream half is landed btw.

[michaeln, 2009-10-02 23:35:24]

LGTM