Refactoring: window.btoa() and window.atob() should be implemented on DOMWindow

• Component: DOM
• Status: RESOLVED FIXED

[Hajime Morrita]

Currently window.btoa() and window.atob() are separetely implemented in both JSC and V8 binding.
They should share code as much as possible and the code should be on DOMWindow.

[Hajime Morrita]

Created attachment 49997 [details]
patch v0

[WebKit Review Bot]

Attachment 49997 [details] did not pass style-queue:

Failed to run "WebKitTools/Scripts/check-webkit-style" exit_code: 1
WebCore/page/DOMWindow.cpp:832:  Missing spaces around =  [whitespace/operators] [4]
WebCore/page/DOMWindow.cpp:832:  Missing spaces around <  [whitespace/operators] [3]
Total errors found: 2 in 7 files


If any of these errors are false positives, please file a bug against check-webkit-style.

[Hajime Morrita]

Created attachment 49998 [details]
v1; fix style violations

[Darin Adler]

Comment on attachment 49998 [details]
v1; fix style violations

> -    return jsString(exec, String(out.data(), out.size()));
> +    return jsString(exec, String(result.characters(), result.length()));

This is wrong. It makes an extra copy of the entire string! It should just be jsString(exec, result).

[Hajime Morrita]

Created attachment 50075 [details]
v2; follow the feedback, cleanup a little

[Hajime Morrita]

Thank you for reviewing quickly!

(In reply to comment #4)
> This is wrong. It makes an extra copy of the entire string! It should just be
> jsString(exec, result).
Fixed.
And removed unnecessary code that last patch has.

[Darin Adler]

Comment on attachment 50075 [details]
v2; follow the feedback, cleanup a little

> +static bool hasMultibyteCharacters(const String& string)
> +{
> +    for (unsigned i = 0; i < string.length(); i++) {
> +        if (string[i] > 0xFF)
> +            return true;
> +    }
> +
> +    return false;
> +}

This is not a great name for this function. I don’t really think of U+0040 as a “single byte” character nor do I think of U+0100 as “multi-byte”.

Patch otherwise seems OK.

r=me

[Hajime Morrita]

Created attachment 50083 [details]
v3; rename hasMultibyteCharacters to isSafeToConvertCharList

[Hajime Morrita]

Thank you for reviewing!

> This is not a great name for this function. I don’t really think of U+0040 as a
> “single byte” character nor do I think of U+0100 as “multi-byte”.
I agree that "multi-byte character" is not right. 
But I'm not sure if there is some good concept to represent this case.
So I chose the name intend to be valid locally in this context.

[Darin Adler]

Comment on attachment 50083 [details]
v3; rename hasMultibyteCharacters to isSafeToConvertCharList

> +static bool isSafeToConvertCharList(const String& string)

This is probably better than the old name, but I would have used a name like containsOnlyEightBitCharacters. Sorry I didn't suggest a name last time. r=me either with your name or with my suggested name.

[WebKit Commit Bot]

Comment on attachment 50083 [details]
v3; rename hasMultibyteCharacters to isSafeToConvertCharList

Clearing flags on attachment: 50083

Committed r55619: <http://trac.webkit.org/changeset/55619>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.