[Webkit-unassigned] [Bug 145002] New: Fix the incorrectness that AtomicString could have symbolic StringImpl

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=145002

            Bug ID: 145002
           Summary: Fix the incorrectness that AtomicString could have
                    symbolic StringImpl
    Classification: Unclassified
           Product: WebKit
           Version: 528+ (Nightly build)
          Hardware: Unspecified
                OS: Unspecified
            Status: NEW
          Severity: Normal
          Priority: P2
         Component: Web Template Framework
          Assignee: webkit-unassigned at lists.webkit.org
          Reporter: utatane.tea at gmail.com

In issue 144848, AtomicStringImpl's incorrectness is resolved.
But AtomicString's incorrectness still remains.

There are several plans.

1. Introduce UniquedString, AtomicString, SymbolString and replace the current AtomicString with UniquedString

The simplest plan. Introduce UniquedString and SymbolString as the same to UniquedStringImpl and SymbolImpl.
And replace the current AtomicString use with UniquedString.

Pros:
+ Simple, easy.

Cons:
+ Introduce more and more classes for strings :(

2. Drop AtomicString and use RefPtr<UniquedStringImpl> / Ref<UniquedStringImpl> directly

Drop the AtomicString use cases completely and use RefPtr<UniquedStringImpl> / Ref<UniquedStringImpl> instead.

Pros:
+ We can drop the class for strings! (AtomicString)

Cons:
+ Since there's code using `operator==` for AtomicString, rewriting is not mechanical. (We need to consider about PassRefPtr<UniquedStringImpl>...)
+ WTFString remains.


In both, IDL change is involved.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.webkit.org/pipermail/webkit-unassigned/attachments/20150514/51797738/attachment.html>