[webkit-reviews] review granted: [Bug 106377] IndexedDB: Allow createIndex/createObjectStore to be asynchronous : [Attachment 182040] Patch

[bugzilla-daemon at webkit.org]

Tony Chang <tony at chromium.org> has granted Alec Flett
<alecflett at chromium.org>'s request for review:
Bug 106377: IndexedDB: Allow createIndex/createObjectStore to be asynchronous
https://bugs.webkit.org/show_bug.cgi?id=106377

Attachment 182040: Patch
https://bugs.webkit.org/attachment.cgi?id=182040&action=review

------- Additional Comments from Tony Chang <tony at chromium.org>
View in context: https://bugs.webkit.org/attachment.cgi?id=182040&action=review


> Source/WebCore/ChangeLog:35
> +	   * Modules/indexeddb/IDBDatabaseBackendImpl.cpp: Add all
create/delete operations.

Nit: I would say where these are copied from.

> Source/WebCore/Modules/indexeddb/IDBBackingStore.cpp:608
> +// FIXME: This should do some error handling rather than plowing ahead when
bad data is encountered.

Are there security implications of plowing ahead when bad data is encountered?

> Source/WebCore/Modules/indexeddb/IDBDatabaseBackendImpl.h:69
> +    virtual PassRefPtr<IDBObjectStoreBackendInterface>
createObjectStore(int64_t id, const String& name, const IDBKeyPath&, bool
autoIncrement, IDBTransactionBackendInterface*, ExceptionCode&)  {
ASSERT_NOT_REACHED(); return 0; }
> +    virtual void createObjectStore(int64_t transactionId, int64_t
objectStoreId, const String& name, const IDBKeyPath&, bool autoIncrement);
> +    virtual void deleteObjectStore(int64_t, IDBTransactionBackendInterface*,
ExceptionCode&) { ASSERT_NOT_REACHED(); }
> +    virtual void deleteObjectStore(int64_t transactionId, int64_t
objectStoreId);

Should these have OVERRIDE?