December 2012 - Dillo-dev - lists.mailman3.com

Patch for xembed (text corrected)
by podarcis＠lavabit.com Jan. 14, 2013

Jan. 14, 2013

Hi all, Here is a patch to make embedding dillo into an X-window work. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ diff -r 9bad41ea1dde src/xembed.cc --- a/src/xembed.cc Wed Dec 05 13:03:04 2012 +0100 +++ b/src/xembed.cc Sun Dec 09 10:12:36 2012 +0100 // TODO: Implement more XEMBED support; -void Xembed::show() { +void Xembed::create() { createInternal(xid); setXembedInfo(1); Fl::event_dispatch(event_handler); diff -r 9bad41ea1dde src/xembed.hh --- a/src/xembed.hh Wed Dec 05 13:03:04 2012 +0100 +++ b/src/xembed.hh Sun Dec 09 10:12:36 2012 +0100 @@ -14,10 +14,11 @@ public: Xembed(uint32_t xid, int _w, int _h) : Fl_Window(_w, _h) { - this->xid = xid; + this->xid = xid; create(); }; - void show(); + void create(); int handle(int event); + void resize(int x, int y, int w, int h); }; #endif ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ tested with this FLTK program: ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #include <FL/Fl.H> #include <FL/Fl_Window.H> #include <FL/x.H> #include <cstdio> #include <iostream> #include <sstream> #include <string> using namespace std; int main( int argc_, char *argv_[] ) { Fl_Window dw( 800, 600, "dillo embedded" ); dw.resizable( &dw ); dw.show(); cout << "dw.xid = " << (unsigned long)fl_xid(&dw) << endl; ostringstream cmd; cmd << "dillo -f -g " << dw.w() << "x" << dw.h() << " --xid " << (unsigned long)fl_xid(&dw); if ( argc_ > 1 ) cmd << " " << argv_[1]; cmd << " &"; cout << "system '" << cmd.str() << "'" << endl; system( cmd.str().c_str() ); return Fl::run(); } ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ What does NOT work is automatic resizing of the embedded dillo window, when the size is changed. This would be nice to have. I do not know enough of the embedding stuff, but I wonder if the static int event_handler() ever gets called with type ClientMessage - at least I could not make it get there.. podarcis

4 11

ui colors again
by corvid＠lavabit.com Jan. 13, 2013

Jan. 13, 2013

Today I dug up the old ui colors patch from last September, ripped out the part that used system colors as defaults that was a problem*, and played around with it a little. I really like configurable colors, so, since I haven't heard anyone say anything like feature freeze yet, I'll make another push for this thing. One difference from the previous code is that I'm using inactive() on the images for inactive toolbar buttons because our gray pixmaps tend not to look so good with different UI colors. It might be good to pull it somewhat closer to the background color, but I'm not sure yet. * Under X, FLTK tries to "Read colors that KDE writes to the xrdb database", which may or may not work fine for those people running KDE, but...

3 20

wizard, switching between UIs
by corvid＠lavabit.com Jan. 2, 2013

Jan. 2, 2013

Yesterday I was playing with ctrl-tab, of course, and I noticed how the state of the focused tab would sometimes change. I went through a few rounds of fixing this in different ways before finding something that didn't seem icky. Detailed comment in patch.

2 3

Render bug
by Hole.destructor＠gmx.de Jan. 2, 2013

Jan. 2, 2013

Dear Dillo developers, sorry for not creating a proper bug report, but the "How to reproduce" message field is too small for the example file. :) The issue ========= After closing the "Save As" dialog, the background color of the HTML page is not redrawn correctly (leaving white rectangles instead of the background color). How to reproduce ================ 1. open the appended html file (note the grey background) 2. click on "file" link 3. move the "Save As" window over the grey background 4. close the "Save As" window --> parts of the background are drawn white, instead of grey Best, Alex

3 3

CAPI ignores download requests when not in cache or handled by the downloads DPI.
by onepoint＠starurchin.org Jan. 2, 2013

Jan. 2, 2013

I wondered why "Save link as" silently failed on some of my local web pages if I had not already visited the link. It turns out they were "file:///" links and "Save link as" sends a download request which the CAPI silently ignores if the URL is not cached and is not handled by the downloads DPI. I added a MSG_WARN to log this but haven't changed any behaviour. I still found it surprising and was wondering if it could be improved? Regards, Jeremy Henty

4 5

hyphenator valgrind error
by corvid＠lavabit.com Jan. 2, 2013

Jan. 2, 2013

==30983== Invalid read of size 1 ==30983== at 0x81151B6: fl_utf8back (fl_utf.c:259) ==30983== by 0x80A2320: dw::fltk::FltkPlatform::prevGlyph(char const*, int) (fltkplatform.cc:599) ==30983== by 0x809EDA3: dw::Hyphenator::hyphenateSingleWord(dw::core::Platform*, char*, int, lout::misc::SimpleVector<int>*) (hyphenator.cc:327) ==30983== by 0x809E9F2: dw::Hyphenator::hyphenateWord(dw::core::Platform*, char const*, int*) (hyphenator.cc:256) ==30983== by 0x809C087: dw::Textblock::hyphenateWord(int) (textblock_linebreaking.cc:762) ==30983== by 0x809BB6E: dw::Textblock::wordWrap(int, bool) (textblock_linebreaking.cc:609) ==30983== by 0x809B6A9: dw::Textblock::processWord(int) (textblock_linebreaking.cc:441) ==30983== by 0x8098753: dw::Textblock::addText0(char const*, unsigned int, short, dw::core::style::Style*, dw::core::Requisition*) (textblock.cc:1694) ==30983== by 0x8097B02: dw::Textblock::addText(char const*, unsigned int, dw::core::style::Style*) (textblock.cc:1478) ==30983== by 0x8073F5C: Html_process_word(DilloHtml*, char const*, int) (html.cc:1216) ==30983== by 0x807A348: Html_write_raw(DilloHtml*, char*, int, int) (html.cc:3923) ==30983== by 0x80727A9: DilloHtml::write(char*, int, int) (html.cc:531) ==30983== Address 0x661e6bf is 1 bytes before a block of size 16 alloc'd ==30983== at 0x402AC41: malloc (vg_replace_malloc.c:270) ==30983== by 0x80A223F: dw::fltk::FltkPlatform::textToLower(char const*, int) (fltkplatform.cc:584) ==30983== by 0x809E86A: dw::Hyphenator::hyphenateWord(dw::core::Platform*, char const*, int*) (hyphenator.cc:223) ==30983== by 0x809C087: dw::Textblock::hyphenateWord(int) (textblock_linebreaking.cc:762) ==30983== by 0x809BB6E: dw::Textblock::wordWrap(int, bool) (textblock_linebreaking.cc:609) ==30983== by 0x809B6A9: dw::Textblock::processWord(int) (textblock_linebreaking.cc:441) ==30983== by 0x8098753: dw::Textblock::addText0(char const*, unsigned int, short, dw::core::style::Style*, dw::core::Requisition*) (textblock.cc:1694) ==30983== by 0x8097B02: dw::Textblock::addText(char const*, unsigned int, dw::core::style::Style*) (textblock.cc:1478) ==30983== by 0x8073F5C: Html_process_word(DilloHtml*, char const*, int) (html.cc:1216) ==30983== by 0x807A348: Html_write_raw(DilloHtml*, char*, int, int) (html.cc:3923) ==30983== by 0x80727A9: DilloHtml::write(char*, int, int) (html.cc:531) ==30983== by 0x8079D85: Html_callback(int, _CacheClient*) (html.cc:3799)

2 2

Dpi_connect_socket retry
by p37sitdu＠lavabit.com Jan. 1, 2013

Jan. 1, 2013

Dpi_connect_socket function from src/IO/dpi.c (also duplicated in test/cookies.c) has retry argument. It is set to TRUE on first call, and Dpi_connect_socket calls itself recursively with retry == FALSE in case of error. This retry function is not working. If connection fails, at line 645 sock_fd is set, but ret is still set to -1. Even if the second attempt to connect is successful, -1 is returned and conected sock_fd is lost. It is possible to fix it by adding "ret = sock_fd", but I think it is better to just remove retries completely. It is unlikely that connecting using loopback interface will fail on first try and succeed on retry. Also, comment says "the server closes sock_fd on error". However, sock_fd should still be closed by client to free allocated file descriptor. OS don't know if Dillo is going to read from it (in which case error such as ECONNRESET should happen) so file descriptor can't be reused. File descriptor should be closed in case of error in a_Dpip_build_cmd too. Patch to fix this function and its calls in src/IO/dpi.c and test/cookies.c is attached.

2 6

show_url pref
by corvid＠lavabit.com Dec. 30, 2012

Dec. 30, 2012

Since it's been broken throughout the dillo3 era, the other day I made it so that you can't disable it -- and now I'm wondering what the value is in even having the pref. Removing Location doesn't give us room for anything more important. It's kind of like having a show_viewport preference.

2 1

PATCH: uicmd.cc prefs.save_dir cleanup
by onepoint＠starurchin.org Dec. 30, 2012

Dec. 30, 2012

uicmd.cc exports a get()/set() interface to the default save directory that is never used elsewhere, and used only trivially internally. It also rereads prefs.save_dir at every save, which is redundant. It would be simpler to have an init() interface that is called once. There would be no change to the user. Patch attached. Thoughts? Regards, Jeremy Henty

2 2

findbar behaviour
by corvid＠lavabit.com Dec. 28, 2012

Dec. 28, 2012

When I want to change the case-sensitivity of a search, I don't like having to switch to the mouse and aim it toward the little checkbox. So I looked up why we disable tabbing to the other widgets (and found nothing). Then a little experimentation showed that triggering shortcuts causes widgets to get focus, which we don't want. With that answered, I tried letting the input do its own callbacks upon Enter or Shift-Enter, and letting everything accept focus. This isn't heavily tested, but I wanted to see how the idea goes over first...

2 8