Re: [webkit-dev] Optimization of buffer allocation of Vector and StringBuilder

2011-10-11 Thread Darin Adler
On Oct 11, 2011, at 3:32 AM, Xianzhu Wang (王显著) wrote: > add a StringImpl::reallocate() method Sounds good. Before StringBuilder, StringImpl objects were never resized, so this is a recently-introduced issue. -- Darin ___ webkit-dev mailing lis

Re: [webkit-dev] Pixel testing and WebKit2

2011-10-11 Thread Martin Robinson
On Tue, Oct 11, 2011 at 3:24 PM, Simon Fraser wrote: > After various fixes, pixel tests now work reasonably well in > WebKitTestRunning (i.e. when running with WebKit2). Do any of these fixes apply to other platforms as well? I'd like to have GTK+ pixel tests matching between WebKit1 and WebKit2

Re: [webkit-dev] Pixel testing and WebKit2

2011-10-11 Thread Dirk Schulze
Am 12.10.2011 um 00:24 schrieb Simon Fraser: > For this reason, I'd like to propose that on Mac, all pixel testing is done > using WebKit2, which you can get by passing "-2" to run-webkit-tests or > new-run-webkit-tests. I can fix the scripts to print a warning if you pass > --pixel without -2

[webkit-dev] Pixel testing and WebKit2

2011-10-11 Thread Simon Fraser
After various fixes, pixel tests now work reasonably well in WebKitTestRunning (i.e. when running with WebKit2). One final step is enabling mock scrollbars in DRT/WTR . The goal of using mock scrollbars is to reduce the pixel differences between pl

Re: [webkit-dev] How do I turn off timeout for unresponsive plug-ins, to allow long debug sessions

2011-10-11 Thread Rudi Sherry
Hi Anders, > in if you're running Safari with a WebKit nightly, you can disable the > timeout using ...does that mean the shipping Safari won't do that? Is there any way to disable this with the shipping Safari? It doesn't appear to be disabling it with Safari 5.1 (the real deal, not a night

[webkit-dev] Reading from Object

2011-10-11 Thread Soheil Servati Beiragh
In RenderBlockLineLayout.cpp, I'm trying to read face of the font from pointer to a structure named f below: RenderText* t = toRenderText(o); const Font& f = t->style(firstLine)->font(); the data I need exist in: f.m_fontlist.m_ptr->m_cachePrimarySimpleFontData->m_platformData.m_face I have

Re: [webkit-dev] Making PopupMenuClient testable

2011-10-11 Thread Fady Samuel
*ping* I'd love to get some feedback on https://bugs.webkit.org/show_bug.cgi?id=69631. :-) Thanks, Fady On Mon, Oct 10, 2011 at 5:27 PM, Fady Samuel wrote: > Please review then? :-) > > Fady > > On Mon, Oct 10, 2011 at 9:11 PM, Darin Adler wrote: > > On Oct 10, 2011, at 7:57 AM, Fady Samuel

Re: [webkit-dev] Optimization of buffer allocation of Vector and StringBuilder

2011-10-11 Thread 王显著
I found replacing malloc/memcpy with realloc() can reduce many actual allocations and copies. I tested with DumpRenderTree (using chromium's tcmalloc) accessing 20 popular websites. In StringBuilder using realloc() instead of malloc/memcpy when applicable, 45.3% of realloc()s return the same pointe

Re: [webkit-dev] Fwd: Typography in webkit

2011-10-11 Thread Meir TOLEDANO
Hi to everyone, I am working on web project which deals with lots of fonts (thousands and thousands...) so I need to use the @font-face css features and I would like to have precision about that: 1. How, the browser validate a font? For example if the font have glyph codes in CC (control charact

[webkit-dev] Reg: Full-width space character rendering

2011-10-11 Thread just2 contribute
Hi, On seeing some of the bugs on Text component, I came across one bug; https://bugs.webkit.org/show_bug.cgi?id=38226 So for this bug, can we midify the WTF code for the function; static inline bool isSpaceOrNewline(UChar c) - like; this is checking whether it is an ASCII space or a Unicode