- Dillo-dev - lists.mailman3.com

patch: fltkimgbuf memleak

by Johannes.Hofmann＠gmx.de

On Tue, Apr 29, 2008 at 01:24:49AM +0000, corvid wrote: > You should see a significant decrease in memory used. Wow, indeed! > > diff -pur dw2/dw/fltkimgbuf.cc dw2-cur/dw/fltkimgbuf.cc > --- dw2/dw/fltkimgbuf.cc 2008-01-16 02:57:23.000000000 +0000 > +++ dw2-cur/dw/fltkimgbuf.cc 2008-04-29 00:40:04.000000000 +0000 > @@ -119,11 +119,10 @@ FltkImgbuf::~FltkImgbuf () > //else > // printf("FltkImgbuf[root %p]: deleted\n", this); > > - if (!isRoot()) { > + if (!isRoot()) > root->detachScaledBuf (this); > - delete rawdata; > - } > > + delete rawdata; > delete copiedRows; > > if (scaledBuffers) > _______________________________________________ > Dillo-dev mailing list > Dillo-dev(a)dillo.org > http://lists.auriga.wearlab.de/cgi-bin/mailman/listinfo/dillo-dev

April 29, 2008

Future of dillo2

by 4winter＠informatik.uni-hamburg.de

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Johannes Hofmann wrote: > On Sun, Apr 27, 2008 at 05:33:33AM +0200, Justus Winter wrote: >> Somewhat related I would suggest to get rid of the location dialog (ctrl >> + l). As far as I can see it has offers nothing more than entering the >> address directly into the url bar, so whats the point? Make ctrl + l a >> shortcut that moves the cursor to the url bar instead. > > On the other hand the dialogs allow to use dillo without panel > (Ctrl-Space). That's a nice feature, indeed. What about getting rid of the dialog box and just showing the panel on demand in "without panel" mode? >> Flickering... there was a thread about fltk and double buffer >> something... there is no flickering when scrolling a page on my machine >> (there are some artifacts in the scrollbar, it looks like it is jumping >> around a little or changing its size) but if I move the whole window it >> flickers a lot. > > Both are fltk bugs I think. I also see this with fltk test > applications if I add clear_double_buffer(). > We should submit bug reports for that. If you have a small patch ready that introduces this behavior into one of their test programs I nominate you for this task :) >> And one more obscure bug I just noticed while playing around with the >> scrollbar: When I switch from one virtual desktop (my window manager is >> e17) to another with a dillo window on it and try to scroll the page it >> doesn't work. Most of the time it selects text on the page (it didn't >> when I recorded the movie, typical...). The problem goes away if I move >> or resize the window so I would guess that it has something to do with >> window coordinates / dimensions not being correct. I compiled dillo with >> fltk2 revision 6101. >> >> http://teythoon.cryptobitch.de/stuff/virtual-desktops.ogg > > I don't see this with dwm. Can you try whether fltk test > applications show this behaviour too? http://teythoon.cryptobitch.de/stuff/virtual-desktops-fltk2-problem.ogg You are right, this seems to be a fltk2 bug. I will file a bug report for this problem. Cheers, Justus - -- gpg key fingerprint: C82D 382A AB38 1A54 5290 19D6 A0F9 B035 686C 6996 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFIFyr6oPmwNWhsaZYRArM/AJ44W5nWh1xekGLtM1RrfmC9ujW3eACfTVbx OlU9NERtkjeQqraFwc96M2c= =pMct -----END PGP SIGNATURE-----

April 29, 2008

DO NOT: [PATCH]: call widget->layout_damage() from FltkWidgetView::allocateFltkWidget

by onepoint＠starurchin.org

D'oh! > This patch makes FltkWidgetView::allocateFltkWidget mark the FLTK > widget's layout as damaged. This fixes > dw::fltk::ui::FltkListResource (which is necessary to implement > MULTIPLE SELECT inputs in forms). And it also makes TextResources crash. Augh! Let me try and figure this out. Regards, Jeremy Henty

April 29, 2008

[PATCH]: call widget->layout_damage() from FltkWidgetView::allocateFltkWidget

by onepoint＠starurchin.org

This patch makes FltkWidgetView::allocateFltkWidget mark the FLTK widget's layout as damaged. This fixes dw::fltk::ui::FltkListResource (which is necessary to implement MULTIPLE SELECT inputs in forms). Currently a dw::fltk::ui::FltkListResource widget fails to display itself. (You can run test/dw-resource-test to see this). This is because the resource's fltk::MultiBrowser widget is created with a tiny size and later reallocated to the correct size. The viewport calls layout() each time, but this doesn't call widget->layout() unless the widget's layout is damaged. Unfortunately the layout is marked as damaged when the widget is created but *not* after it is resized, so MultiBrowser::layout() gets called when the widget is tiny, but not when it is the right size. Alas, MultiBrowser::draw() uses size information that was cached by MultiBrowser::layout() , so it behaves as though the widget is still tiny, concludes that none of the Items are visible and doesn't draw anything. This bug potentially bites all widgets, it's just sheer luck that only MultiBrowser is visibly affected. Other widgets are either simple enough to survive their layout() being called when they were a different size, or else smart enough to mark *themselves* as damaged so that the viewport calls their layout() at the right time. Regards, Jeremy Henty

April 29, 2008

patch: fltkimgbuf memleak

by corvid＠lavabit.com

You should see a significant decrease in memory used.

April 29, 2008

patch: motionNotify and LEAVE

by corvid＠lavabit.com

Since findtext has been in the news lately, I might as well at least provide this bit so that my code behaves better with dialogs.

April 28, 2008

Future of dillo2

by Johannes.Hofmann＠gmx.de

On Sun, Apr 27, 2008 at 05:33:33AM +0200, Justus Winter wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Johannes Hofmann wrote: > > On Mon, Apr 21, 2008 at 03:39:50PM +0000, corvid wrote: > >> I think the very first thing is to update the main dillo.org > >> page so that visitors will know that dillo shows signs of life. > > I agree. We should update the front page and make it a bit more > > optimistic :-). > +1 > > Also I'd like to see some infrastructure changes like a more advanced > bug tracker / modify the existing one (user comments come to mind, and > *please* get rid of the background image in the search results). > > > What is still missing for an official release apart from image maps? > I like the way firefox is getting rid of more and more dialog boxes. > Like the "find in page" command. In firefox the user enters the search > string into a input field embedded in a bar emerging from the bottom of > the screen. > > This has several advantages: > * avoid cluttering the desktop (this is even more important if the > screen is very small [think mobile phones]) > * avoid to cover the search result (this is possible if it isn't > possible to scroll the windows content in a way that the search result > is visible) > * you always know which search box belongs to which browser window > since the search bar is a part of that window (this gets even worse with > tabs) > > Somewhat related I would suggest to get rid of the location dialog (ctrl > + l). As far as I can see it has offers nothing more than entering the > address directly into the url bar, so whats the point? Make ctrl + l a > shortcut that moves the cursor to the url bar instead. On the other hand the dialogs allow to use dillo without panel (Ctrl-Space). > > Flickering... there was a thread about fltk and double buffer > something... there is no flickering when scrolling a page on my machine > (there are some artifacts in the scrollbar, it looks like it is jumping > around a little or changing its size) but if I move the whole window it > flickers a lot. Both are fltk bugs I think. I also see this with fltk test applications if I add clear_double_buffer(). We should submit bug reports for that. > > And one more obscure bug I just noticed while playing around with the > scrollbar: When I switch from one virtual desktop (my window manager is > e17) to another with a dillo window on it and try to scroll the page it > doesn't work. Most of the time it selects text on the page (it didn't > when I recorded the movie, typical...). The problem goes away if I move > or resize the window so I would guess that it has something to do with > window coordinates / dimensions not being correct. I compiled dillo with > fltk2 revision 6101. > > http://teythoon.cryptobitch.de/stuff/virtual-desktops.ogg I don't see this with dwm. Can you try whether fltk test applications show this behaviour too? Cheers, Johannes

April 28, 2008

[patch] phrase search - take 2

by Johannes.Hofmann＠gmx.de

Hi, here comes my second attempt on phrase search. As already described, I changed CharIterator back to not return \0s. Instead blanks are inserted when the space flag is set. This way CharIterator always returns a smooth stream of chars whether for normal html- or pre-text. Then there was a bug in DeepIterator::prev() that took me some time to find. I also adjusted charsEqual() to treat all space-chars equally. Finally I had to add a check in CharIterator::highlight() to avoid a crash when searching for the last word in pre-text with a blank appended. Cheers, Johannes

April 28, 2008

Since we've been talking about name clashes, what about "assert"

by onepoint＠starurchin.org

On Sun, Apr 27, 2008 at 06:58:22PM +0200, Johannes Hofmann wrote: > On Sun, Apr 27, 2008 at 08:52:31AM +0100, Jeremy Henty wrote: > > > > Because lout::misc defines an inline assert function, including > > dw/core.hh fails if you've already included anything that defines > > the standard assert macro (such the C++ iostream stuff). > > I also see this with gcc 3.4.6, but not with gcc 4.1.2. That's surprising (at least to me). I guess gcc4 must resolve names differently? Or maybe it's smart enough not to macro-expand anything that looks like a macro token but appears somewhere where no macro token should? Anyway, I have no axe to grind on this, I just raised it as a concern. Regards, Jeremy Henty -- What is Jewish decision theory? Maximize regrets. -- Sidney Morgenbesser

April 27, 2008

patch: Re: patch: Re: [Dillo-dev] start on image maps (dw-part)

by corvid＠lavabit.com

> /* BUG: if several ISMAP images follow each other without > * being separated with a word, only the first one is ISMAPed > */ PS I left this old comment in, but if I make a file containing a bunch of ismap images in a row, they certainly _seem_ ismapped. I click on them and dillo tries to go to http://[wherever]?x,y.

April 27, 2008