- Dillo-dev - lists.mailman3.com

prefs.show_url
by corvid＠lavabit.com Sept. 3, 2011

Sept. 3, 2011

I just found that turning off prefs.show_url messes up the appearance of the UI.

1 0

valgrind still unhappy despite expect changes
by jcid＠dillo.org Sept. 2, 2011

Sept. 2, 2011

On Sun, Aug 28, 2011 at 06:53:54PM +0000, corvid wrote: > > I can't reproduce this with valgrind-3.6.1 on wheezy. > > > > No valgrind error report before or after these debug messages: > > > > a_Nav_cancel_expect_if_eq:: calling a_Nav_cancel_expect > > Dillo: normal exit! > > > > I tested by pushing a remote URL on CLI and closing the window > > before the page data starts flowing (i.e. to cancel the > > "expect"), and also with the cancelled download code path. > > Neither one produced an error. > > I saw that there was another log entry the other day > http://starurchin.org/dillo/valgrind/7ea9fc809376ddf7dde2908e2ecf999aea2741… > > Do you think something sneaky could be going on, or does it seem > like it must come down to configuration differences somehow? AFAIS, either one is possible... Me not being able to reproduce it says it's most probably be OK for the codepaths I tested (FWIW, the ones I can see). If there's an untested codepath, well... OTOH, valgrind produces some false positives (its docs say it) and gcc optimization can also play a role... So, I'd try to find a simple test case that produces the log. Then, test this problematic codepath here, and see what comes out of the exercise. Can you reproduce the bug? Do you have a simple test case? -- Cheers Jorge.-

4 16

valgrind doctree leak
by corvid＠lavabit.com Aug. 28, 2011

Aug. 28, 2011

(from Jeremy's logs) ==2186== 928 (32 direct, 896 indirect) bytes in 1 blocks are definitely lost in loss record 815 of 958 ==2186== at 0x40275CC: operator new(unsigned int) (vg_replace_malloc.c:255) ==2186== by 0x806E258: Doctree::push() (doctree.hh:67) ==2186== by 0x806C1FC: StyleEngine::startElement(int) (styleengine.cc:93) ==2186== by 0x8076DB8: Html_process_tag(DilloHtml*, char*, int) (html.cc:3507) ==2186== by 0x8077A33: Html_write_raw(DilloHtml*, char*, int, int) (html.cc:3822) ==2186== by 0x8070543: DilloHtml::write(char*, int, int) (html.cc:552) ==2186== by 0x80775BD: Html_callback(int, _CacheClient*) (html.cc:3725) ==2186== by 0x80627AD: Cache_process_queue (cache.c:1195) ==2186== by 0x80629A2: Cache_delayed_process_queue_callback (cache.c:1260) ==2186== by 0x40C0876: Fl::wait(double) (in /usr/lib/libfltk.so.1.3) ==2186== by 0x40C0A30: Fl::run() (in /usr/lib/libfltk.so.1.3) ==2186== by 0x804F08A: main (dillo.cc:436)

2 1

problems with very long alt text
by jcid＠dillo.org Aug. 28, 2011

Aug. 28, 2011

On Fri, Aug 26, 2011 at 07:29:21PM +0200, hiro wrote: > With tinycorelinux and Xorg-7.5 i915 on 1024x768 I get a completely > garbled screen and unresponsive system, CPU ran hot and I had to > reboot. > I can reboot into VESA if you want. [later hiro wrote:] > Ok I don't get the bug with Xvesa, tinycorelinux' default Xserver. Excellent! This proves the point (AFAICS). Thank you very much. -- Cheers Jorge.-

1 1

margin-bottom counted twice
by johannes.hofmann＠gmx.de Aug. 27, 2011

Aug. 27, 2011

Hi, I found that current dillo uses twice the given margin-bottom: value. See attached margin_bottom.html. It is related to commented out code in textblock.cc. Attached patch tries to fix it while not breaking the test cases in http://www.dillo.org/test/img/ I'm not 100% sure about removing the "if" in Textblock::wordWrap(), but why shouldn't a widget's top margin count in the first line of a paragraph? Cheers, Johannes

4 5

What's pending for RC2
by jcid＠dillo.org Aug. 26, 2011

Aug. 26, 2011

Hi there, So far we have: [ok] moveToWidget(). Jorge has code for this. [ok] downloads dpi string escaping. [ok] alt text wrapping (which looks like a nice feature). [ok] remove large panel. [ok] https dialog storm (BUG#868). [ok] The findbar input doesn't know about Ctrl+{a,e,d,k}. [ok] something other than "ctab%d" for initial tab label. [ok] single close tab button with tooltip. [ok] valgrind logs: close window [ok] valgrind logs: cancel expected url [ok] segfault with Ctrl-s (web search) [ok] UI_TEMPORARILY_SHOW_PANELS [ok] leave link into a native FLTK widget [ok] remove pointeronlink(), panelmode_cb_i() [ok] the panel problems when resizing from very small. [ok] resize problems when tabs overflow [ok] URL_ILLEGAL_CHARS [ok] Fixed CustLightButton "light" color deactivation. [ok] hoverTooltip glitch (tooltip linger when going into a child DIV). [ok] tab overflow (code a handler) [ok] valgrind error for first tab close, then resize(). [ok] update the Changelog [ ] default search engine [ ] new-tab and window title [ ] CSS adjacent sibling selectors I'd like to comment some: [ok] tab overflow (code a handler) Please send feedback on how it works. Remember that polish and eye candy was left for the next release. [ok] update the Changelog Corvid and Johannes may have some pending items to add. BTW, at this point the Changelog has a summary of user-visible changes, which reduces clutter given that the full history is in hg. Please vote. me: +1 on keeping it this way. [ ] default search engine I've been using search_url="http://duckduckgo.com/lite/?kp=-1&q=%s" and it works quite well for me. Please vote. me: +1 for making it the default Also I'd like to add some default search engines. For instance, this one is very useful for me: search_url="Free Dictionary http://www.thefreedictionary.com/%s" Comments/suggestions? [ ] new-tab and window title This is a simple deatail that I'll look into quickly. [ ] CSS adjacent sibling selectors Johannes has to decide on this. -- Cheers Jorge.-

3 3

problems with very long alt text
by corvid＠lavabit.com Aug. 26, 2011

Aug. 26, 2011

With remote stylesheets off and images off, opening http://www.dillo.org/test/long_alt_text.html and dragging the horizontal scrollbar left and right for a little while will make my machine suddenly start to page madly. from what I could see from valgrind and top, nothing looked amiss in dillo or X, but even after dillo can finally close (if it can), too many resources are tied up somehow and I need to restart X to have a really usable system. I changed the alt text back to using drawText(), and it didn't matter much that I could see. I tried using a width of 16K instead of the 25K or so that was the real value. This didn't seem to help either.

6 17

Tab close button gets clobbered by tabs
by mr_semantics＠hotmail.com Aug. 24, 2011

Aug. 24, 2011

Hello all, This evening I discovered that if I open too many tabs, the close tab 'X' at the top right of the browser gets clobbered out of sight, until I close some tabs (with CTRL-w). On the old Dillo (2.x), you could open a fair amount of tabs, and the x would never be clobbered, but instead a new instance of Dillo would start, after (I presume) a predetermined amount of tabs were opened. On another note, while fiddling with the tabs on Dillo 3 and 2.x, I got comparing the two, and I think I like the look of the old version better. I have two reasons: 1) I think the old version tabs look nicer (I realize not everyone will think like me), and more like the tabs you would see in other browsers. 2) From a practical standpoint, the old tabs seemed to behave a little nicer, in that they would adjust their size a little bit to show more of the title of the page opened in it. As it stands with Dillo 3, you can't see much of the title text before it gets an ellipsis. Regards, Rob

6 14

Two bugs in the Dillo GUI (both Mac and ubuntu)
by corvid＠lavabit.com Aug. 22, 2011

Aug. 22, 2011

Back in June, reza wrote: > 4. This one is more of a feature request. Like most Mac applications, the > shortcuts should work with Command key instead of control key. (I know it's > low priority, and I might give it a shot myself later on). I'm wondering whether it would be helpful to mention in keysrc that meta = command, or would that already be known by (or at least exceedingly obvious to) everyone using osx? BTW, 1.3 has #ifdef __APPLE__ # define FL_COMMAND FL_META ///< An alias for FL_CTRL on WIN32 and X11, or FL_META on MacOS X # define FL_CONTROL FL_CTRL ///< An alias for FL_META on WIN32 and X11, or FL_CTRL on MacOS X #else # define FL_COMMAND FL_CTRL ///< An alias for FL_CTRL on WIN32 and X11, or FL_META on MacOS X # define FL_CONTROL FL_META ///< An alias for FL_META on WIN32 and X11, or FL_CTRL on MacOS X #endif // __APPLE__ although I don't think this can do us any good in our circumstances.

1 3

right_click_closes_tab
by corvid＠lavabit.com Aug. 22, 2011

Aug. 22, 2011

dillorc has: # A mouse's middle click over a tab closes the Tab. # With mousewheel mouses, right click feels way better (set to YES). #right_click_closes_tab=NO I'm thinking that if the default is no and we start by talking about middle click, it would make the most sense to call it middle_click_closes_tab.

1 0