- webkit-efl - lists.webkit.org

HEADS-UP: ewk_view_setting_scripts_window_open_{get, set} have been renamed
by Raphael Kubo da Costa 01 Nov '11

01 Nov '11

Hey there. After SVN commit 98908, the ewk_view_setting_scripts_window_open_* family of functions has been renamed. - ewk_view_setting_scripts_window_open_get is now ewk_view_setting_scripts_can_open_windows_get. - ewk_view_setting_scripts_window_open_set is now ewk_view_setting_scripts_can_open_windows_set. Since r93833 we had the API in a weird state where the calls to (dis)allow scripts to close windows did not match the naming scheme of the calls to (dis)allow scripts to open windows, which was pretty weird. The naming scheme of the former made much more sense, so I've chosen to adapt the latter, older functions. -- Raphael Kubo da Costa ProFUSION embedded systems http://profusion.mobi

1 0

Heads-up: API change in ewk_view_run_open_panel
by Raphael Kubo da Costa 14 Oct '11

14 Oct '11

Due to r97336 and r97338, which changed some parameter types in WebCore, our implementation of ewk_run_open_panel (and the expected virtual method to be implemented in child classes) has had its signature changed. The accept_types parameter, which as previously a const char*, is now a Vector<String> in ewk_view_run_open_panel, which then turns it into an Eina_List for child classes. Please note that, as the documentation says, child implementations do NOT need to free the lists or its elements themselves. -- Raphael Kubo da Costa ProFUSION embedded systems http://profusion.mobi

1 0

About removing ewk_xxx_forward | backward_possible()
by Gyuyoung Kim 13 Oct '11

13 Oct '11

Hello efl folks, It seems to me that there are unnecessary APIs in ewk port. The APIs are ewk_frame_back_possible(), ewk_frame_forward_possible(). I think the APIs can be enough to replace by ewk_frame_navigate_possible(). We just add a comment regarding how to check if backword and forward is possible via steps parameter of ewk_frame_navigate_possible(). 597 Eina_Bool ewk_frame_back_possible(Evas_Object* o) 598 { 599 return ewk_frame_navigate_possible(o, -1); 600 } 601 602 Eina_Bool ewk_frame_forward_possible(Evas_Object* o) 603 { 604 return ewk_frame_navigate_possible(o, 1); 605 } 606 607 Eina_Bool ewk_frame_navigate_possible(Evas_Object* o, int steps) 608 { 609 EWK_FRAME_SD_GET_OR_RETURN(o, sd, EINA_FALSE); 610 EINA_SAFETY_ON_NULL_RETURN_VAL(sd->frame, EINA_FALSE); 611 WebCore::Page* page = sd->frame->page(); 612 return page->canGoBackOrForward(steps); 613 } Ewk_view.cpp also has same APIs which using ewk_frame_back | forward_possible(). 1496 Eina_Bool ewk_view_back_possible(Evas_Object* o) 1497 { 1498 EWK_VIEW_SD_GET_OR_RETURN(o, sd, EINA_FALSE); 1499 return ewk_frame_back_possible(sd->main_frame); 1500 } 1501 1502 Eina_Bool ewk_view_forward_possible(Evas_Object* o) 1503 { 1504 EWK_VIEW_SD_GET_OR_RETURN(o, sd, EINA_FALSE); 1505 return ewk_frame_forward_possible(sd->main_frame); 1506 } 1507 1508 Eina_Bool ewk_view_navigate_possible(Evas_Object* o, int steps) 1509 { 1510 EWK_VIEW_SD_GET_OR_RETURN(o, sd, EINA_FALSE); 1511 return ewk_frame_navigate_possible(sd->main_frame, steps); 1512 } 1513 I'd like to know how do webkit-efl developers think this removal. Thanks, Gyuyoung.

2 3

POR FAVOR RESPUESTA PARA MÁS DETALLES.
by Dr Halidou Zida 07 Oct '11

07 Oct '11

Dr. Halidou Zida Estimado amigo, Tengo una propuesta para usted, sin embargo, esto no es obligatorio ni yo de ninguna manera que obligan a cumplir el contenido en contra de su voluntad. Soy el Dr. Halidou Zida, yo trabajo con un banco (uno de los principales bancos en frica). Aqu , en este banco existe una cuenta inactiva durante los ltimos 10 a os, que pertenece a un ciudadano estadounidense que es ya tarde, el Sr. John Cuthbertson, quien muri en Alaska Airlines en Alaska avi n vuelo 261 el 31 de enero de 2000. Cuando descubr que no hab a dep sitos o retiros de esta cuenta para este per odo de tiempo, me decid a llevar a cabo una investigaci n de sistemas y descubri que ninguno de los miembros de la familia ni las relaciones de la persona es consciente de finales de esta cuenta. Ahora quiero una cuenta en el extranjero, donde el banco de transferencia de este fondo en. Despu s de eso, voy a destruir todos los documentos relacionados con esta cuenta. Se trata de una red de cuidado y para los ltimos once meses he trabajado a cabo todo lo necesario para una operaci n de enganche libre. La cantidad m s todos los intereses acumulados es EE.UU. $ 15,500,000.00 (quince millones, quinientos mil d lares de EE.UU. solamente). Ahora mis preguntas son: - 1. ¿Se puede manejar este proyecto? 2. ¿Te puedo dar esta confianza? 3. ¿Cu l ser su comisi n si usted puede facilitar el movimiento de este fondo en su cuenta? Tenga en cuenta estos y volver a m tan pronto como sea posible con su nombre completo, tel fono particular y n mero de fax, direcci n postal correspondiente, etc Por ltimo, es mi humilde petici n de que la informaci n contenida en este documento se otorga con todas las de confidencialidad y el secreto es necesario que se merece espero su respuesta urgente a esta petici n, Atentamente, Atentamente, Dr. Zida Halidou

1 0

Re: [webkit-efl] I'd like to refactor zoom related APIs.
by Ryuan Choi 05 Oct '11

05 Oct '11

> I think scale_set can be replaced zoom_set. But, I'm not sure if scale_set > is more clear than zoom_set for users. scale_set doesn't replace zoom_set, but ewk_view_zoom_set lost functionality of center zoom. This patch provide three different zoom feature, scale_set and zoom_set and text_zoom_set. But, as we discussed in private channel, scale_set and zoom_set is still confused. So, How do you think about ewk_{frame|view}_page_zoom_set instead of ewk_{frame|view}_zoom_set ? > In addition, it looks gtk and qt port still use zoom instead scale term. IIRC, gtk doesn't support scalePage feature yet. you can find it from WebKit2. Best Regards, Ryuan Choi

2 1

I'd like to refactor zoom related APIs.
by Ryuan Choi 05 Oct '11

05 Oct '11

You can see history in https://bugs.webkit.org/show_bug.cgi?id=62842 In a short, I'd like to introduce scalePage feature of webkit. But, current ewk_{frame|view}_zoom_{get|set} is not good to expand. So, I'd like to change zoom related APIs like gustavo mentioned. 1) remove last two argument(center position) of ewk_view_zoom_set. 2) remove ewk_{view|frame}_text_only_{get|set} 3) introduce ewk_{view|frame}_text_zoom_{get|set}, ewk_view_scale_{get|set} ewk_tiled_backingstore will use ewk_view_scale_set instead of ewk_view_zoom_set. please visit at https://bugs.webkit.org/show_bug.cgi?id=62842 if you want to see changes. Best Regards, Ryuan Choi +------------------------------------------------------------------------ Listen and listen and listen -----------------------+

2 1

Re: [webkit-efl] I want to change *.c to *.cpp
by Ryuan Choi 24 Sep '11

24 Sep '11

> > Do you really need the extern "C" declarations in the headers? This code > will only be used by C++ files. > Yes, I need it. > If not, it makes linking errors. I realize that this is because ewk_tiled_backing_store declare extern "C". I removed. > > > > > And I dislike using typeof here, as it will really break the build for > > anything that is not gcc or clang. I suggest getting rid of the macros > > (you could even turn them into proper template functions) or passing > > another parameter to them specifying the data type. > > I agree, I'll remove those macro. > I thought that I'll prepare next bug for this bug to be focused renaming. Update like you mentioned. please see attached file. > > > Or, taking another step back, is this all really needed? We are have to > > make so many adjustments to make code which is clearly C code into an > > uglified C++ version, and IIRC the reason given for that was some kind > > of #include problem. Couldn't we fix that instead of spending so much > > effort into this? > > Previous bug is for #include problem. so that I didn't touch other c files. > But now, we are in phase to improve WebKit/efl code quality. > > Because we need to change our tiled backing store until we decide whether we use WebCore version, > We need reviewer's help and these cleanup will help them review easily.

1 0

Re: [webkit-efl] I want to change *.c to *.cpp
by 최병운 24 Sep '11

24 Sep '11

> Do you really need the extern "C" declarations in the headers? This code > will only be used by C++ files. Yes, I need it. If not, it makes linking errors. > > And I dislike using typeof here, as it will really break the build for > anything that is not gcc or clang. I suggest getting rid of the macros > (you could even turn them into proper template functions) or passing > another parameter to them specifying the data type. I agree, I'll remove those macro. I thought that I'll prepare next bug for this bug to be focused renaming. > Or, taking another step back, is this all really needed? We are have to > make so many adjustments to make code which is clearly C code into an > uglified C++ version, and IIRC the reason given for that was some kind > of #include problem. Couldn't we fix that instead of spending so much > effort into this? Previous bug is for #include problem. so that I didn't touch other c files. But now, we are in phase to improve WebKit/efl code quality. Because we need to change our tiled backing store until we decide whether we use WebCore version, We need review's help and these cleanup will help them review easily. Best Regards, Ryuan Choi

1 0

Re: [webkit-efl] I want to change *.c to *.cpp
by Ryuan Choi 24 Sep '11

24 Sep '11

> > Ok, committed in r63568 of EFL. Thank you. Attached file is simple diff after renaming for easy review. If there are no objection, I'll create a patch Best Regards, Ryuan Choi

2 2

Re: [webkit-efl] I want to change *.c to *.cpp
by Lucas De Marchi 23 Sep '11

23 Sep '11

Hi Ryuan On Fri, Sep 23, 2011 at 6:51 AM, Ryuan Choi <ryuan.choi(a)samsung.com> wrote: > >> I thought you would change all *.c and not just ewk_view_single.c and >> ewk_view_tiled.c. > > I'd like to share patch to fix above issue. > Now I'm waiting internal review for it. > > There are build break issues, so that I tried to not only rename but also fix breaks, pointer operator and cast. > > BTW, I realize that EINA_INLIST_FOREACH make build break with cpp. > For now, I redefine EINA_INLIST_FOREACH. > Should we need to fix it? Damn C++ that doesn't allow the conversion from void* to type*. We have some options: 1) Use "-fpermissive" flag and live with the warning (there doesn't seem to be an option to turn it off in gcc). 2) Patch eina to provide a c++ version of the macro. This would mean using typeof() though, which is a gcc feature. Looking at how it's implemented, it would be very simple, just casting to typeof(ptr) instead of to void*. 3) Use a different version of the macro in WebKit, like you did. 4) Don't do this conversion from c to c++. Ryuan, was this the only issue of converting from C to C++? I like option 2. I made a simple patch to eina that fixes it if we are using C++, but I didn't committed it yet. I'd like to hear opinions from others. regards, Lucas De Marchi

1 1