Hi Pawel, Are you referring to favicons here? Then I'd suggest, generally, looking at the core functions I wrote for my favicon branch, that really should be merged to trunk quite quickly to avoid duplication of work; treating that as a kind of 'API' to retrieve favicons; then beyond that there is the question of implementing a cache-to-disk, that clearly belongs in a new branch; As for the text input, my only comment is that at current projections, I'd hope to complete the customisation branch perhaps in time for Friday [although my car may need major work so I may have limited time for now], while it may well be that theme support would naturally be a simple addition to the customisation branch; at that stage then the only branch for me to work at would be hotlist, so I'd really encourage you to fully complete your treeview/hotlist branch first, before touching text input — in a new branch please :-) Best Mark http://www.halloit.com Key ID 046B65CF Paweł Blokus wrote:

Paweł Blokus wrote: Hi Pawel, Me I'm trying to resist the temptation to work in the middle of the night :) generally to see for instance what my favicon branch does, I'd suggest svn diff svn://svn.netsurf-browser.org/trunk/netsurf@7236 svn://svn.netsurf-browser.org/branches/MarkieB/favs | <your favourite editor stdin command here> perhaps I should add commentary to some of the less-commented code, though I hope it's relatively clear what it does; hopefully we've already identified the main overlap areas I've been writing my reports direct to my mentor, so you'll see virtually no sign of commentary in the lists either; it looks as though that should change too :) this is the general schema of how my project is progressing; I notice incidentally that you have listed keyboard nav as one of your later milestones, are you very keen to achieve that? As I was eyeing keyboard nav perhaps ahead of trying my hand at a windows port; keyboard nav being a more approachable project; the question being whether you are that confident of the speed of your progress that you definitely want to reserve keyboard nav all to yourself? Maybe I could tackle keyboard nav, then as long as you've completed your primary milestones in time, we'll work at the windows port together? Feature Status Now /http://www.netsurf-browser.org/developers/gsoc/2009ideas/ Find Migration from riscos/amiga to core, Implementation in GTK; branch awaiting review/merger Bookmarks Branch development awaiting glade branch review/merger; currently waiting for Paul B's core work to complete Improved right-click menu Improved implementation in GTK; more improvement in toolbar customisation branch, right-click aspect now complete Adjust toolbar buttons to match GTK prefs Now Implemented in trunk File Uploads Branch ready for review/merger View Source Implemented in GTK; could add features; could migrate to core Documentation No progress yet Favicons core/GTK implementation branch written; awaiting review/merger /http://www.netsurf-browser.org/documentation/progress/ Documentation GTK See above [yet to implement] Find Text GTK See above [branch ready] Full Save GTK GTK implementation branch written; migration of riscos implementation to core; amiga implementation needs merger in view of added core functionality; branch awaiting review/merger History-global GTK Branch ready for review / merger Hotlist/Bookmarks GTK See above [waiting] Save as text GTK GTK implementation branch written; awaiting review/merger Toolbar customisation GTK prefs complete; full customisation implementation being worked at now Themes GTK No additional progress yet; foreseen in toolbar customisation though /Additional/ Web search/ / core / GTK implementation branch written; awaiting review / merger Improved tabbed browsing More implementation dispersed in patches/branches : patch partially undrawn new tab; view source in new tab; focus new tab, blank new tab; More progress possible glade file simplification more rational organisation of Main NetSurf glade file; made into several files, one for every GTK window As the main project is virtually complete, I daresay it adds little to send my proposal too :) Best Mark http://www.halloit.com Key ID 046B65CF

Chris Young wrote: Hi Chris, the main difficulty is that I have no way that I know of, of compiling for Amiga to check my work; one more branch that I have yet to properly check for Amiga compatibility is search branch, involving a similar migration of functions from riscos/amiga to core; so the best would be for you to check that both branches compile for Amiga, possibly debug the Amiga compile for search branch, possibly patch the full save branch? As, in the full save branch, I simply avoided touching the Amiga branch at all, recognising the difficulty I was in in search branch; when I say 'needs merger in view of added core functionality' that's really a question of code optimisation, it should work untouched; longer term though it's a case of 'rewiring' the Amiga front to the changed core functions Best Mark http://www.halloit.com Key ID 046B65CF

Hi Mark I fixed save_complete by replacing amiga/save_complete.c with gtk/gtk_save.c and amiga/save_complete.h with the attached (a better approach is to search through my files for #include "amiga/save_complete.h" and replace it with #incude "desktop/save_complete.h" and then delete amiga/save_complete.h) There is a minor problem as I don't get an icon for index, I'm just adding that back in now. I've also improved the search code by adding Show All and Case Sensitive options. The two strings ShowAll and CaseSensitive will need to be added to the Messages files. Regards Chris

Hello i have add this code, to get sdl better working, SDL_VideoModeOK return the depth of the screen so netsurf use always best color depth in window mode. please take a look, if thats possible to add.i test and it work ok. its in file fb_frontend_sdl.c framebuffer_t *fb_os_init(int argc, char** argv) { framebuffer_t *newfb; int fb_width; int fb_height; int fb_depth; if ((option_window_width != 0) && (option_window_height != 0)) { fb_width = option_window_width; fb_height = option_window_height; } else { fb_width = 800; fb_height = 600; } fb_depth = option_fb_depth; if ((fb_depth != 32) && (fb_depth != 16) && (fb_depth != 8)) fb_depth = 16; /* sanity checked depth in bpp */ newfb = calloc(1, sizeof(framebuffer_t)); if (newfb == NULL) return NULL; if ( SDL_Init(SDL_INIT_VIDEO) < 0 ) { fprintf(stderr, "Unable to init SDL: %s\n", SDL_GetError()); return NULL; } atexit(SDL_Quit); newfb->width = fb_width; newfb->height = fb_height; + fb_depth = SDL_VideoModeOK(newfb->width,newfb->height,fb_depth, SDL_SWSURFACE); + newfb->bpp = fb_depth; + if (!fb_depth)return 0; sdl_screen = SDL_SetVideoMode(newfb->width, newfb->height, newfb->bpp, SDL_SWSURFACE ); if ( sdl_screen == NULL ) { fprintf(stderr, "Unable to set video: %s\n", SDL_GetError()); free(newfb); return NULL; } + SDL_EnableKeyRepeat(300, 50); Regards

On Mon, 8 Jun 2009 02:18:14 +0200, Pawe³ Blokus wrote: I looked at your select widget corewidgets code some time back and forgot to comment, but this could affect some of your code structure and your other corewidgets. In my (Amiga) port I set everything up to write to an off-screen bitmap, and then transfer that to the visible display. With NetSurf as-is, you call content_redraw (or history_redraw) from a platform-specific function, which allows any setup required before drawing, and any other activity after drawing, to be performed. In your code, the plotter functions are being called without the platform code getting any warning, or knowing when it is finished. In my port, that means the select menu never displays, which obviously isn't what we want and isn't trivial to fix without modifying the core. I would suggest to add a couple of platform-specific functions, something like: gui_redraw_start(struct gui_window *g) gui_redraw_end(struct gui_window *g) which "top and tail" any internal redraws and allow the platform code to do whatever is required, or a NOP. I don't know how vital this is to other ports, but at a guess the plotter code would at least need to know which window it is supposed to be rendering to. In my case I need something like this executed after the plotter functions have finished: { struct IBox *bbox; GetAttr(SPACE_AreaBox,g->gadgets[GID_BROWSER],(ULONG *)&bbox); BltBitMapRastPort(glob.bm,0,0,g->win->RPort,bbox->Left,bbox->Top, bbox->Width,bbox->Height,0x0C0); } (although really it should be limited to the area that has actually been drawn over) and nothing beforehand. You'll need to check the normal redraw code for other platforms to see if they have any particular requirements. Regards Chris