[webkit-reviews] review denied: [Bug 47389] IndexedDB does not have a quota mechanism. : [Attachment 70184] Patch

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Thu Oct 7 22:51:23 PDT 2010


Jeremy Orlow <jorlow at chromium.org> has denied Andrei Popescu
<andreip at google.com>'s request for review:
Bug 47389: IndexedDB does not have a quota mechanism.
https://bugs.webkit.org/show_bug.cgi?id=47389

Attachment 70184: Patch
https://bugs.webkit.org/attachment.cgi?id=70184&action=review

------- Additional Comments from Jeremy Orlow <jorlow at chromium.org>
View in context: https://bugs.webkit.org/attachment.cgi?id=70184&action=review

close

> LayoutTests/storage/indexeddb/database-basics.html:124
> +function checkQuotaEnforcing()

I think this stuff would be best in its own test.

> WebCore/storage/IDBDatabaseBackendImpl.cpp:155
> +	   transaction->abort();

return?

> WebCore/storage/IDBDatabaseBackendImpl.cpp:230
> +	  
callbacks->onError(IDBDatabaseError::create(IDBDatabaseException::UNKNOWN_ERR,
"Error writing data to stable storage."));

Add a fixme to spec this.

> WebCore/storage/IDBObjectStoreBackendImpl.cpp:136
> +    return (resultCode != SQLResultOk) && (resultCode != SQLResultDone);

don't need ()'s

When could a result be either one or the other?  Isn't it always defined which
it'll be in a non-error case?

> WebCore/storage/IDBObjectStoreBackendImpl.cpp:192
> +    RefPtr<IDBTransactionBackendInterface> transactionPtr = transaction;

um..you put the Ptr suffix in the wrong place  :-)

> WebCore/storage/IDBObjectStoreBackendImpl.cpp:246
> +	  
callbacks->onError(IDBDatabaseError::create(IDBDatabaseException::UNKNOWN_ERR,
"Error writing data to stable storage."));

fixmes...here and anywhere u use unkown_err

> WebKit/chromium/src/WebIDBFactoryImpl.cpp:61
> +void WebIDBFactoryImpl::open(const WebString& name, const WebString&
description, WebIDBCallbacks* callbacks, const WebSecurityOrigin& origin,
WebFrame*, const WebString& dataDir, unsigned long long maximumSize)

int64 should work, i think.  i'm pretty sure i used it somewhere else in this


More information about the webkit-reviews mailing list