Bug 115781 – [BlackBerry] Fix usage of BlackBerry::Platform::String

• Status: RESOLVED
• Resolution: FIXED
• Component: WebKit BlackBerry

[Alberto Garcia]

We're creating BlackBerry::Platform::String passing UTF-8 data to its char* constructor, which is interpreted as Latin1.

[Alberto Garcia]

Created attachment 201044 [details]
Patch

[Benjamin Poulain]

Comment on attachment 201044 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=201044&action=review

> Source/WebKit/blackberry/Api/WebSettings.cpp:209
> +    DEFINE_STATIC_LOCAL(AtomicString, WebKitMonospace, ("-webkit-monospace"));
> +    DEFINE_STATIC_LOCAL(AtomicString, WebKitSansSerif, ("-webkit-sans-serif"));
> +    DEFINE_STATIC_LOCAL(AtomicString, WebKitSerif, ("-webkit-serif"));
> +    DEFINE_STATIC_LOCAL(AtomicString, WebKitStandard, ("-webkit-standard"));

Given that "settings" is static local, you will never hit this path again. Why would you define those static?

It also looks like your "WebSettings" class takes a String, not an AtomicString.

[Arvid Nilsson]

(In reply to comment #2)
> (From update of attachment 201044 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=201044&action=review
> 
> > Source/WebKit/blackberry/Api/WebSettings.cpp:209
> > +    DEFINE_STATIC_LOCAL(AtomicString, WebKitMonospace, ("-webkit-monospace"));
> > +    DEFINE_STATIC_LOCAL(AtomicString, WebKitSansSerif, ("-webkit-sans-serif"));
> > +    DEFINE_STATIC_LOCAL(AtomicString, WebKitSerif, ("-webkit-serif"));
> > +    DEFINE_STATIC_LOCAL(AtomicString, WebKitStandard, ("-webkit-standard"));
> 
> Given that "settings" is static local, you will never hit this path again. Why would you define those static?
> 
> It also looks like your "WebSettings" class takes a String, not an AtomicString.

One possible reasoning behind this could be:
"-webkit-monospace" will be added into the AtomicString table sooner or later anyway, or could already be in the table at this point.
That depends on if/how -webkit-monospace is used down in WebCore.

[Alberto Garcia]

Anyway I think Benjamin is right in that it's not necessary to define
these static. I'll update the patch to fix that.

I'll also add a few more changes from a different internal commit that
can also go here.

[Alberto Garcia]

Created attachment 201057 [details]
Patch

Second try.

[Rob Buis]

Comment on attachment 201057 [details]
Patch

LGTM.

[WebKit Commit Bot]

Comment on attachment 201057 [details]
Patch

Clearing flags on attachment: 201057

Committed r149769: <http://trac.webkit.org/changeset/149769>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.