- Dillo-dev - lists.mailman3.com

Uncluttering dillo, part II: the findbar

by jcid＠dillo.org

On Wed, May 21, 2008 at 12:02:14AM +0200, Justus Winter wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Hey everyone :) Hi there! > attached is a patch that transforms dillos find in page dialog box into > a firefox style toolbar. Nice! Committed with some changes. See below. > There are some issues that I cannot put my > finger on... see the comments in the event handler. Maybe there's some > magic about fltks event model that I did not fully understand... then > again it might be fltks fault. Anyway, I'd be happy for some hints. Ok, the comment says: > /* BUG: this is acting funny > * > * 1. figure out why the first EscapeKey is only > * sent to the input widget if it is focused at the time > * 2. ctrl-f ; type foo ; select foo and type bar ; press esc > * => foo is inserted into the input box > */ Fixed both in the committed version. FLTK2 sends the event to handle() and its return value tells whether the event was handled or not. Here, the input widget was handling the event (and making the selection empty) so there was no need to propagate the event upwards, but when the selected text was already empty (second time) there was nothing to do and the event was propagated upwards. The patch was the same as for the location bar. A custom input widget that handles keys in a slightly different way that the native Input widget. In this case (MyInput) we don't handle the escape key and let FLTK pass it upwards, and thus get the desired effect. > Jorge, could you update your coding conventions with respect to c++ > specific language features? Oh yes. I need to find some time though. I've been giving priority to pending patches. The coding conventions, are clearly out of date. At least the C++ part. In this regard we're currently following dw2's style. Can you write a few rules down and send them to me to update the page. The idea is not to be pedantic on style issues, but to define a common style that we can adhere to. -- Cheers Jorge.-

May 22, 2008

memory statistics

by Johannes.Hofmann＠gmx.de

Hi, to see which part of dillo consumes which amount of memory, I hacked together a version of dlib.c that counts memory that was allocated via dMalloc(). A malloc_tag allows to assign the mallocs to specific parts in the code. In the patch below I use malloc_tag = 1 for cache stuff, malloc_tag = 2 for text strings that are passed dw2 and malloc_tag = 0 for everything else. It's fun to see how memory get's allocated and free'd while loading. Here is the output once the huge mysql manual page is loaded: ====> malloc size 5185114 26967623 6602392 0 So that's 6M for text strings, 26M cache and 5M other. Of course there is also all the dw2 stuff that is not taken into account here. BTW, textblock delete's the text strings that are allocated using malloc(). That's not allowed and might cause unexpected behaviour on some platforms. Cheers, Johannes

May 22, 2008

RFC: human interface guidelines

by jcid＠dillo.org

On Tue, May 20, 2008 at 11:55:59PM +0200, Justus Winter wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Hey everyone :) > > I was thinking about improving the dillo experience by adding some > functionality / shortcuts people have come to expect. But what people > expect is influenced by their background, their choice of desktop > environment and platform. True. > We could pick one / some user interface guideline(s) ([1] lists some) > and try to adhere to it / them. Or we could pick one popular browser and > copy its 'feel'. Some people love dillo's bindings! :-) The canonical solution is to provide configurable keybindings for the most used functions. Maybe worth trying? ;) -- Cheers Jorge.-

May 21, 2008

[patch] minor memory reduction

by jcid＠dillo.org

On Wed, May 21, 2008 at 05:14:25PM +0200, Johannes Hofmann wrote: > Ok, here comes a second attempt. It works with your test page and > avoids the memore leak. Please test. Committed. -- Cheers Jorge.-

May 21, 2008

Uncluttering dillos UI - getting rid of the dialog boxes

by jcid＠dillo.org

On Wed, May 21, 2008 at 10:10:59AM +0200, Justus Winter wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Justus Winter wrote: > > Jorge Arellano Cid wrote: > >> On Sat, May 17, 2008 at 05:46:26PM +0200, Justus Winter wrote: > >>> -----BEGIN PGP SIGNED MESSAGE----- > >>> Hash: SHA1 > >>> > >>> I thought about dillos fullscreen mode, which is not a fullscreen mode > >>> at all but a "get rid of those panels"-mode. So I decided to rename > >>> those functions and variables accordingly (00-*) > >> Committed. > > > >> With the addition that when the Location is already focused, > >> the text is also selected. This helps as a visual clue and also > >> to quickly get rid of the old URL. > > > >> Note: I used a trick in ::handle() to do it. If you know a > >> better way, please send a patch. > > > > Yes, that's a nice feature indeed. Reading the fltk documentation I came > > up with the attached patch. I am pretty sure that strlen is the wrong > > function when dealing with utf-8 character data, but you get the idea. > > > > Justus > > Looking at the fltk source I realized that Input::size() does exactly > what we want and is also much more efficient. Revised patch is attached. Committed (with the comment). -- Cheers Jorge.-

May 21, 2008

Keyboard shortcuts & fullscreen button

by jcid＠dillo.org

On Tue, May 20, 2008 at 11:28:54PM +0200, Justus Winter wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Jorge Arellano Cid wrote: > > On Sat, May 17, 2008 at 06:09:19PM +0200, Justus Winter wrote: > >> Also I noticed, that the way events are handled is somewhat lax with > >> respect to modifier keys. The code just checks if ctrl is pressed, not > >> if ctrl but none of the other modifiers are pressed. This makes ctrl-l > >> equivalent with ctrl-shift-alt-l. I'd like to hear your opinions, is > >> this a bug or a feature? > > > > Bug I think. > > > > The UI was coded in a hurry to get it running. Lax parts need > > polish. > > The attached patch addresses modifier issue. Committed. -- Cheers Jorge.-

May 21, 2008

Uncluttering dillo, part II: the findbar

by Johannes.Hofmann＠gmx.de

Hi Justus, On Wed, May 21, 2008 at 12:02:14AM +0200, Justus Winter wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Hey everyone :) > > attached is a patch that transforms dillos find in page dialog box into > a firefox style toolbar. There are some issues that I cannot put my > finger on... see the comments in the event handler. Maybe there's some > magic about fltks event model that I did not fully understand... then > again it might be fltks fault. Anyway, I'd be happy for some hints. This is really nice! For text search the popup menus are even more annoying than for entering an url. One minor thing. You can make the CheckButton and Input members of Findbar private if you use static methods for the callbacks instead of functions. Cheers, Johannes

May 21, 2008

[patch] minor memory reduction

by Johannes.Hofmann＠gmx.de

Ok, here comes a second attempt. It works with your test page and avoids the memore leak. Please test. Cheers, Johannes On Tue, May 20, 2008 at 05:50:35PM -0400, Jorge Arellano Cid wrote: > On Tue, May 20, 2008 at 07:16:45PM +0200, Johannes Hofmann wrote: > > Hi, > > > > attached is a small patch to improve memory usage. It creates a > > single word for a sequence of spaces in PRE mode instead of one word > > per character. > > Good idea, but it's not the same rendering. Check the offsets > with the attached file. > > BTW, there's a memory leak when len=0 (bacause Dstr allocates > for the string). While reviewing I thought why not just count the > spaces and use dStrnfill(num_spc, ' '). > > -- > Cheers > Jorge.- > <html> > <pre> > Bla bla bla. Blabla blabla bla blala bla > bla. Blabla blabla bla bla. > > Blala bla. Blabla blabla bla blaba bla bla. Bla > blabla bla bla. > > Blilly blilly bla bla bla. Blabla > blabla bla bla. > <HR> > Lallyla layllaylay layla layla lally lally. > </pre> > </html> > _______________________________________________ > Dillo-dev mailing list > Dillo-dev(a)dillo.org > http://lists.auriga.wearlab.de/cgi-bin/mailman/listinfo/dillo-dev

May 21, 2008

[patch] minor memory reduction

by Johannes.Hofmann＠gmx.de

On Tue, May 20, 2008 at 05:50:35PM -0400, Jorge Arellano Cid wrote: > On Tue, May 20, 2008 at 07:16:45PM +0200, Johannes Hofmann wrote: > > Hi, > > > > attached is a small patch to improve memory usage. It creates a > > single word for a sequence of spaces in PRE mode instead of one word > > per character. > > Good idea, but it's not the same rendering. Check the offsets > with the attached file. Ouch! Do you think this is a general problem or should I try to get this working somehow? > > BTW, there's a memory leak when len=0 (bacause Dstr allocates > for the string). While reviewing I thought why not just count the > spaces and use dStrnfill(num_spc, ' '). Yes, that would be better. Cheers, Johannes > > -- > Cheers > Jorge.- > <html> > <pre> > Bla bla bla. Blabla blabla bla blala bla > bla. Blabla blabla bla bla. > > Blala bla. Blabla blabla bla blaba bla bla. Bla > blabla bla bla. > > Blilly blilly bla bla bla. Blabla > blabla bla bla. > <HR> > Lallyla layllaylay layla layla lally lally. > </pre> > </html> > _______________________________________________ > Dillo-dev mailing list > Dillo-dev(a)dillo.org > http://lists.auriga.wearlab.de/cgi-bin/mailman/listinfo/dillo-dev

May 21, 2008

Uncluttering dillos UI - getting rid of the dialog boxes

by 4winter＠informatik.uni-hamburg.de

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Justus Winter wrote: > Jorge Arellano Cid wrote: >> On Sat, May 17, 2008 at 05:46:26PM +0200, Justus Winter wrote: >>> -----BEGIN PGP SIGNED MESSAGE----- >>> Hash: SHA1 >>> >>> I thought about dillos fullscreen mode, which is not a fullscreen mode >>> at all but a "get rid of those panels"-mode. So I decided to rename >>> those functions and variables accordingly (00-*) >> Committed. > >> With the addition that when the Location is already focused, >> the text is also selected. This helps as a visual clue and also >> to quickly get rid of the old URL. > >> Note: I used a trick in ::handle() to do it. If you know a >> better way, please send a patch. > > Yes, that's a nice feature indeed. Reading the fltk documentation I came > up with the attached patch. I am pretty sure that strlen is the wrong > function when dealing with utf-8 character data, but you get the idea. > > Justus Looking at the fltk source I realized that Input::size() does exactly what we want and is also much more efficient. Revised patch is attached. Sorry for the spam. 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 iD8DBQFIM9kSoPmwNWhsaZYRAhaDAKC7A/I3SYJJRYIJ/kzbw0VI/afG0QCgm+WZ VuXGU9/u8Y64FLaYQyP27o8= =hvqx -----END PGP SIGNATURE-----

May 21, 2008

RFC: human interface guidelines

by 4winter＠informatik.uni-hamburg.de

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hey everyone :) I was thinking about improving the dillo experience by adding some functionality / shortcuts people have come to expect. But what people expect is influenced by their background, their choice of desktop environment and platform. We could pick one / some user interface guideline(s) ([1] lists some) and try to adhere to it / them. Or we could pick one popular browser and copy its 'feel'. Attached are two patches mapping 'reload' to F5 and 'help' to F1. Let me know what you think of them and the whole interface thing. Justus 1: http://en.wikipedia.org/wiki/Human_Interface_Guidelines - -- 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 iD8DBQFIM0jvoPmwNWhsaZYRAveJAJ9ROk+E0VWLzK6G99qcTlpNjeB5igCeJ7r9 Grz4wtWQ+BIEbWrIY8PDCFg= =JnX2 -----END PGP SIGNATURE-----

May 21, 2008

Uncluttering dillo, part II: the findbar

by 4winter＠informatik.uni-hamburg.de

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hey everyone :) attached is a patch that transforms dillos find in page dialog box into a firefox style toolbar. There are some issues that I cannot put my finger on... see the comments in the event handler. Maybe there's some magic about fltks event model that I did not fully understand... then again it might be fltks fault. Anyway, I'd be happy for some hints. Jorge, could you update your coding conventions with respect to c++ specific language features? 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 iD8DBQFIM0pmoPmwNWhsaZYRAi1VAJwO/e+JWmCoMsjOF/hYuF4X1DUOWwCePB96 +X++Veu0Gyv854m0TxSwI7Q= =rc5Z -----END PGP SIGNATURE-----

May 21, 2008

[patch] minor memory reduction

by jcid＠dillo.org

On Tue, May 20, 2008 at 07:16:45PM +0200, Johannes Hofmann wrote: > Hi, > > attached is a small patch to improve memory usage. It creates a > single word for a sequence of spaces in PRE mode instead of one word > per character. Good idea, but it's not the same rendering. Check the offsets with the attached file. BTW, there's a memory leak when len=0 (bacause Dstr allocates for the string). While reviewing I thought why not just count the spaces and use dStrnfill(num_spc, ' '). -- Cheers Jorge.-

May 21, 2008

Uncluttering dillos UI - getting rid of the dialog boxes

by 4winter＠informatik.uni-hamburg.de

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Jorge Arellano Cid wrote: > On Sat, May 17, 2008 at 05:46:26PM +0200, Justus Winter wrote: >> -----BEGIN PGP SIGNED MESSAGE----- >> Hash: SHA1 >> >> I thought about dillos fullscreen mode, which is not a fullscreen mode >> at all but a "get rid of those panels"-mode. So I decided to rename >> those functions and variables accordingly (00-*) > > Committed. > > With the addition that when the Location is already focused, > the text is also selected. This helps as a visual clue and also > to quickly get rid of the old URL. > > Note: I used a trick in ::handle() to do it. If you know a > better way, please send a patch. Yes, that's a nice feature indeed. Reading the fltk documentation I came up with the attached patch. I am pretty sure that strlen is the wrong function when dealing with utf-8 character data, but you get the idea. 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 iD8DBQFIM0OmoPmwNWhsaZYRAi0KAJ9B9akYL9AN1c51U0HJoXe9imPcjwCfe7I2 oGA/oCVvuUyZ3+CYPBPsH5U= =SCqk -----END PGP SIGNATURE-----

May 20, 2008

Keyboard shortcuts & fullscreen button

by 4winter＠informatik.uni-hamburg.de

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Jorge Arellano Cid wrote: > On Sat, May 17, 2008 at 06:09:19PM +0200, Justus Winter wrote: >> Also I noticed, that the way events are handled is somewhat lax with >> respect to modifier keys. The code just checks if ctrl is pressed, not >> if ctrl but none of the other modifiers are pressed. This makes ctrl-l >> equivalent with ctrl-shift-alt-l. I'd like to hear your opinions, is >> this a bug or a feature? > > Bug I think. > > The UI was coded in a hurry to get it running. Lax parts need > polish. The attached patch addresses modifier issue. 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 iD8DBQFIM0KVoPmwNWhsaZYRAr9lAJ429TW3JBochKDUiHAZTK1n5gapZwCeOVGg Na+M4LYF8QqFGJfOaIBFkhI= =Xkri -----END PGP SIGNATURE-----

May 20, 2008

[patch] minor memory reduction

by Johannes.Hofmann＠gmx.de

Hi, attached is a small patch to improve memory usage. It creates a single word for a sequence of spaces in PRE mode instead of one word per character. The second part is just a minor simplification. Cheers, Johannes

May 20, 2008

Horizontal scrolling

by jcid＠dillo.org

On Sat, May 17, 2008 at 11:17:01PM +0200, Justus Winter wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Huhu there :) > > attached is a tiny patch that enables the horizontal scrolling of the > content of the main dillo widget. Unfortunately fltk supports this up to > now (r6129) only on OSX. I do have a patch for fltk that enables this on > x11, let's see if I can slip it in ;) Committed. BTW, please send patched with "diff -pru". -- Cheers Jorge.-

May 19, 2008

Keyboard shortcuts & fullscreen button

by jcid＠dillo.org

On Sat, May 17, 2008 at 06:09:19PM +0200, Justus Winter wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > The UI constructor creates a button for the fullscreen mode, but it is > not added to anything: > > // Make the full screen button (to be attached to the viewport later) > // TODO: attach to the viewport > FullScreen = new HighlightButton(0,0,15,15); > [...] > > Any idea why this was postponed? Should it be enabled now? The idea is that there's a gadget in the viewport (i.e. a small button in the lower right corner that shows even when in fullwindow mode). This was to be used to enable/disable full window mode. Now, thinking twice, maybe is not a bad idea to hook a right-click fullscreen to this button. Leaving left-click for fullwindow mode. ...but for that, the gadget code needs to be enabled. > Also I noticed, that the way events are handled is somewhat lax with > respect to modifier keys. The code just checks if ctrl is pressed, not > if ctrl but none of the other modifiers are pressed. This makes ctrl-l > equivalent with ctrl-shift-alt-l. I'd like to hear your opinions, is > this a bug or a feature? Bug I think. The UI was coded in a hurry to get it running. Lax parts need polish. -- Cheers Jorge.-

May 19, 2008

Uncluttering dillos UI - getting rid of the dialog boxes

by jcid＠dillo.org

On Sat, May 17, 2008 at 05:46:26PM +0200, Justus Winter wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > I thought about dillos fullscreen mode, which is not a fullscreen mode > at all but a "get rid of those panels"-mode. So I decided to rename > those functions and variables accordingly (00-*) Committed. With the addition that when the Location is already focused, the text is also selected. This helps as a visual clue and also to quickly get rid of the old URL. Note: I used a trick in ::handle() to do it. If you know a better way, please send a patch. BTW, it'd be great if we can get rid of the findtext window, and place it as an horizontal widget near the bottom (i.e as FF). > to make it possible to implement a real fullscreen mode (01-*). Not committed yet. More on this in my next post... -- Cheers Jorge.-

May 19, 2008

website FAQ update

by corvid＠lavabit.com

Jorge wrote: > > [...] (and dillorc2 -> dillorc) > > We need a way to distinguish among the rc for dillo1 and dillo2. > > Currently you can have both dillo1 and dillo2 installed, sharing > cookies and bookmarks (because we have a single ~/.dillo directory). > > Ideas are welcomed, including dillorc2 -> dillo2rc :-) I was thinking that one was just a superset of the other, but that's because I forgot that the fonts would be different.

May 19, 2008

Uncluttering dillos UI - getting rid of the dialog boxes

by 4winter＠informatik.uni-hamburg.de

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Jorge Arellano Cid wrote: > On Sun, May 18, 2008 at 07:15:11PM +0200, Johannes Hofmann wrote: >> shouldn't window resizing be done by the window manager? > > Yes, AFAIS. > >> Do we really need to implement this inside dillo? > > I'd let the WM do it. (e.g. ALT+F11 in Xfce). The problem with this is that we cannot switch to fullwindow mode the moment the windowmanager maximizes our window since fltk does not expose whether our window is in fullscreen mode or not. And we cannot do things like start in fullscreen mode (would be nice for presentations etc). In x11-land there is a NetWM property for this: $ xprop | grep _NET_WM_STATE _NET_WM_STATE(ATOM) = _NET_WM_STATE_FULLSCREEN We could - listen for XPropertyEvents inside of dillo (ugly) - modify fltk to expose those events (nice) But what about the two other major platforms? The homepage states that one of dillos objectives is to be platform independent. I doubt that there's a fullscreen functionality for windows apps, and I have no idea about osx. This seems like a lot of work for a functionality that works (at least it should work) across all three major platforms with a ~30 lines of code patch. What's the problem besides "on x11 the wm should take care of this"? It doesn't break anything, does it? 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 iD8DBQFIMbiVoPmwNWhsaZYRAs2MAJ955OOtsLhdC3NVyNRUC0kkOd6hbgCeMCnd Sd4ofspcJb/5MfbgF83xzTw= =aQ+g -----END PGP SIGNATURE-----

May 19, 2008

[patch] dillo and huge pages

by Johannes.Hofmann＠gmx.de

On Mon, May 19, 2008 at 01:18:01PM -0400, Jorge Arellano Cid wrote: > On Mon, May 19, 2008 at 06:23:11PM +0200, Johannes Hofmann wrote: > > Hi, > > > > to test how dillo behaves on huge pages, I downloaded the mysql > > manual as a single html page: > > http://downloads.mysql.com/docs/refman-5.1-en.html.tar.gz > > > > With this page I noticed that the drawRegion in FltkViewBase can > > grow pretty large which slows down loading the page. Attached patch > > limits the size of the drawRegion to 16 rectangles. > > > > With this patch (and HUGE_FILESIZE in cache.c increased to 16M) > > dillo works pretty well on this huge page. Better than dillo1 and > > firefox! > > Great! > > Committed. > > BTW, this is one of the important points whe should list in the > dillo2 features page. Please somebody remember this! > > > > One problem is that clicking on anchor links (to anchors within the > > page) reloads the page which take some time. > > Some time ago, we had anchors hooked with direct scroll jumps. > It was instantaneous, but the navigation stack wasn't modified > and thus BACK didn't go to the original place. > > It can be hooked to jump again, but in that case, the > navigation stack needs to be modified accordingly. This creates > two paths for the same logic operation inside dillo (which > increases the mantainability costs). That said, if there's a > simple way to do it, I agree. I will have a look at this. > > > > Also memory usage gets pretty high (250M with images disabled). > > I'm experimenting with reducing the size of struct Word in > > textblock.hh by moving the highlight stuff out of struct Word and > > into Textblock. This brings memory usage down to 199M. > > However, I don't know the highlighting stuff enough to be sure I > > don't break anything. This probabely falls into the > > "premature optimization" category anyway. > > Not at all. > > I'd say the important part is to know whether that optimization > can be done without breaking the underlying model for everything > else (a question Sebastian is the most knowledgeable person to > answer in the case of dw2). Ok, I will cleanup what I have and send a patch to the list for review. It would be really great, if Sebastian could have a look at it. Cheers, Johannes

May 19, 2008

[patch] dillo and huge pages

by jcid＠dillo.org

On Mon, May 19, 2008 at 06:23:11PM +0200, Johannes Hofmann wrote: > Hi, > > to test how dillo behaves on huge pages, I downloaded the mysql > manual as a single html page: > http://downloads.mysql.com/docs/refman-5.1-en.html.tar.gz > > With this page I noticed that the drawRegion in FltkViewBase can > grow pretty large which slows down loading the page. Attached patch > limits the size of the drawRegion to 16 rectangles. > > With this patch (and HUGE_FILESIZE in cache.c increased to 16M) > dillo works pretty well on this huge page. Better than dillo1 and > firefox! Great! Committed. BTW, this is one of the important points whe should list in the dillo2 features page. Please somebody remember this! > One problem is that clicking on anchor links (to anchors within the > page) reloads the page which take some time. Some time ago, we had anchors hooked with direct scroll jumps. It was instantaneous, but the navigation stack wasn't modified and thus BACK didn't go to the original place. It can be hooked to jump again, but in that case, the navigation stack needs to be modified accordingly. This creates two paths for the same logic operation inside dillo (which increases the mantainability costs). That said, if there's a simple way to do it, I agree. > Also memory usage gets pretty high (250M with images disabled). > I'm experimenting with reducing the size of struct Word in > textblock.hh by moving the highlight stuff out of struct Word and > into Textblock. This brings memory usage down to 199M. > However, I don't know the highlighting stuff enough to be sure I > don't break anything. This probabely falls into the > "premature optimization" category anyway. Not at all. I'd say the important part is to know whether that optimization can be done without breaking the underlying model for everything else (a question Sebastian is the most knowledgeable person to answer in the case of dw2). -- Cheers Jorge.-

May 19, 2008

website FAQ update

by jcid＠dillo.org

On Fri, May 16, 2008 at 09:01:18PM +0000, corvid wrote: > Jorge wrote: > > Please try to contact J?rgen. If he doesn't answer, leave dillo-dev > > as the only contact until we find somebody else. > > His dillo address redirects to an address where I get User Unknown. > I found a telia.com address, but that also gave me User Unknown. The let's point to "dillo-dev" until somebody volunteers. ;) Although the problem of dillo-dev is that it needs subscription. -- Cheers Jorge.-

May 19, 2008

website FAQ update

by jcid＠dillo.org

On Fri, May 16, 2008 at 09:54:06PM +0000, corvid wrote: > Justus wrote: > > Jorge Arellano Cid wrote: > > > On Fri, May 16, 2008 at 08:21:25PM +0000, corvid wrote: > > >> Are "dillo1" and "dillo2" official names? If so, I'll add in some text to > > >> make it clear what's what. > > > > > > They're not official, but good mnemonics! > > > > > > When there was GTK1 dillo only, there was no confusion. > > > Currently the CVS is using dillo-fltk, but I'd much prefer > > > to have a binary named "dillo2". > > > > > > Probably this is a good choice. Let's hear what dillo-dev > > > people say. Here's the proposal: > > > > > > * call the new binary "dillo2", and refer to the old 0.8.x series > > > as dillo1. > > > > I think it's problematic to have a digit in the official name (or in the > > name of the binary) since people will think it's the version. I'd > > suggest something along these lines: > > > > * Dillo is the name of the brand > > * dillo2 is the codename for the fltk branch > > * for marketing reasons sell any snapshots / upcoming releases of that > > branch as the all new dillo version 2.x.y with tons of new features > > * the binary is called dillo (it's what people expect to type in to > > start the lightweight web browser thingie [at least I tried to run > > $PREFIX/bin/dillo a couple of times now...]) > > I also prefer to have a binary named dillo. OK. > [...] (and dillorc2 -> dillorc) We need a way to distinguish among the rc for dillo1 and dillo2. Currently you can have both dillo1 and dillo2 installed, sharing cookies and bookmarks (because we have a single ~/.dillo directory). Ideas are welcomed, including dillorc2 -> dillo2rc :-) -- Cheers Jorge.-

May 19, 2008

website FAQ update

by jcid＠dillo.org

On Fri, May 16, 2008 at 11:34:57PM +0200, Justus Winter wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Jorge Arellano Cid wrote: > > On Fri, May 16, 2008 at 08:21:25PM +0000, corvid wrote: > >> Are "dillo1" and "dillo2" official names? If so, I'll add in some text to > >> make it clear what's what. > > > > They're not official, but good mnemonics! > > > > When there was GTK1 dillo only, there was no confusion. > > Currently the CVS is using dillo-fltk, but I'd much prefer > > to have a binary named "dillo2". > > > > Probably this is a good choice. Let's hear what dillo-dev > > people say. Here's the proposal: > > > > * call the new binary "dillo2", and refer to the old 0.8.x series > > as dillo1. > > I think it's problematic to have a digit in the official name (or in the > name of the binary) since people will think it's the version. I'd > suggest something along these lines: > > * Dillo is the name of the brand > * dillo2 is the codename for the fltk branch > * for marketing reasons sell any snapshots / upcoming releases of that > branch as the all new dillo version 2.x.y with tons of new features > * the binary is called dillo (it's what people expect to type in to > start the lightweight web browser thingie [at least I tried to run > $PREFIX/bin/dillo a couple of times now...]) Well, that's the canonical way of doing it (no pun intended). OK, let's do it that way then, and let's talk about "dillo2" in the FAQ. -- Cheers Jorge.-

May 19, 2008

patch: dw configure args: rtfl and doublebuffer

by jcid＠dillo.org

On Sun, May 18, 2008 at 08:34:38PM +0000, corvid wrote: > I saw in the bug tracker that --disable-rtfl still turns on rtfl msgs. > > Did some reading. > The third argument to AC_ARG_ENABLE is always run if the option was > specified, regardless of the value. If the option is not specified, > the fourth argument is run. > > --disable-FEATURE just becomes --enable-FEATURE=no. > > (I'll send a separate patch to Jorge to allow compilation with rtfl.) Committed. -- Cheers Jorge.-

May 19, 2008

[patch] dillo and huge pages

by Johannes.Hofmann＠gmx.de

Hi, to test how dillo behaves on huge pages, I downloaded the mysql manual as a single html page: http://downloads.mysql.com/docs/refman-5.1-en.html.tar.gz With this page I noticed that the drawRegion in FltkViewBase can grow pretty large which slows down loading the page. Attached patch limits the size of the drawRegion to 16 rectangles. With this patch (and HUGE_FILESIZE in cache.c increased to 16M) dillo works pretty well on this huge page. Better than dillo1 and firefox! One problem is that clicking on anchor links (to anchors within the page) reloads the page which take some time. Also memory usage gets pretty high (250M with images disabled). I'm experimenting with reducing the size of struct Word in textblock.hh by moving the highlight stuff out of struct Word and into Textblock. This brings memory usage down to 199M. However, I don't know the highlighting stuff enough to be sure I don't break anything. This probabely falls into the "premature optimization" category anyway. Cheers, Johannes

May 19, 2008

patch: dw configure args: rtfl and doublebuffer

by corvid＠lavabit.com

I saw in the bug tracker that --disable-rtfl still turns on rtfl msgs. Did some reading. The third argument to AC_ARG_ENABLE is always run if the option was specified, regardless of the value. If the option is not specified, the fourth argument is run. --disable-FEATURE just becomes --enable-FEATURE=no. (I'll send a separate patch to Jorge to allow compilation with rtfl.)

May 18, 2008

Uncluttering dillos UI - getting rid of the dialog boxes

by jcid＠dillo.org

On Sun, May 18, 2008 at 07:15:11PM +0200, Johannes Hofmann wrote: > On Sat, May 17, 2008 at 05:46:26PM +0200, Justus Winter wrote: > > -----BEGIN PGP SIGNED MESSAGE----- > > Hash: SHA1 > > > > I thought about dillos fullscreen mode, which is not a fullscreen mode > > at all but a "get rid of those panels"-mode. So I decided to rename > > those functions and variables accordingly (00-*) to make it possible to > > implement a real fullscreen mode (01-*). > > shouldn't window resizing be done by the window manager? Yes, AFAIS. > Do we really need to implement this inside dillo? I'd let the WM do it. (e.g. ALT+F11 in Xfce). > I agree on the name change however. Maybe "fullwindow" is a good name. -- Cheers Jorge.-

May 18, 2008

Bug that aborts Dillo

by jcid＠dillo.org

On Sat, May 17, 2008 at 02:17:55AM +0000, corvid wrote: > Jorge wrote: > > Hi there, > > > > I reviewed my personal notes on dillo segfaults and all of them > > but one are fixed now. This is: > > > > BUG: > > http://episteme.arstechnica.com/groupee/forums/a/tpc/f/96509133 > > /m/151006647731/r/264004847731 > > Aborts dillo. > > I tried it and it crashed. > Then I dug up the > > - requisition->width += getStyle()->boxDiffHeight (); > + requisition->width += getStyle()->boxDiffWidth (); > > in sizeRequestImpl from > http://lists.auriga.wearlab.de/pipermail/dillo-dev/2008-January/003690.html > and it no longer wanted to crash for me. Committed, also with the rest of the patch. BTW, when I skip one of these patches, please send me a reminder. I commented the patch to let Sebastian delete/fix the commented-out lines when working there. -- Cheers Jorge.-

May 18, 2008

Uncluttering dillos UI - getting rid of the dialog boxes

by Johannes.Hofmann＠gmx.de

On Sat, May 17, 2008 at 05:46:26PM +0200, Justus Winter wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > I thought about dillos fullscreen mode, which is not a fullscreen mode > at all but a "get rid of those panels"-mode. So I decided to rename > those functions and variables accordingly (00-*) to make it possible to > implement a real fullscreen mode (01-*). shouldn't window resizing be done by the window manager? Do we really need to implement this inside dillo? I agree on the name change however. Cheers, Johannes

May 18, 2008

Horizontal scrolling

by 4winter＠informatik.uni-hamburg.de

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Huhu there :) attached is a tiny patch that enables the horizontal scrolling of the content of the main dillo widget. Unfortunately fltk supports this up to now (r6129) only on OSX. I do have a patch for fltk that enables this on x11, let's see if I can slip it in ;) 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 iD8DBQFIL0tNoPmwNWhsaZYRAmE+AJ9LN79iT4/nyt1JWJYt7BP1PYcFlQCeLV/g BRZ6xTrW7ER5rbDm7V5E1Hk= =SWKS -----END PGP SIGNATURE-----

May 17, 2008

Keyboard shortcuts & fullscreen button

by 4winter＠informatik.uni-hamburg.de

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 The UI constructor creates a button for the fullscreen mode, but it is not added to anything: // Make the full screen button (to be attached to the viewport later) // TODO: attach to the viewport FullScreen = new HighlightButton(0,0,15,15); [...] Any idea why this was postponed? Should it be enabled now? Also I noticed, that the way events are handled is somewhat lax with respect to modifier keys. The code just checks if ctrl is pressed, not if ctrl but none of the other modifiers are pressed. This makes ctrl-l equivalent with ctrl-shift-alt-l. I'd like to hear your opinions, is this a bug or a feature? 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 iD8DBQFILwMvoPmwNWhsaZYRAmHYAJ49XUO6F6mBTG+CbWAzfN5yknkrKgCfQOjB QzlODEtqUdlztUo6tHuluUE= =hCHy -----END PGP SIGNATURE-----

May 17, 2008

Uncluttering dillos UI - getting rid of the dialog boxes

by 4winter＠informatik.uni-hamburg.de

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I thought about dillos fullscreen mode, which is not a fullscreen mode at all but a "get rid of those panels"-mode. So I decided to rename those functions and variables accordingly (00-*) to make it possible to implement a real fullscreen mode (01-*). 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 iD8DBQFILv3RoPmwNWhsaZYRAtAeAKCpvjp7U4n+NW7U25gfd1DJDpnGkACeJw2c PhKxtMxtFGQ1zPs+Qi/Wv8o= =jxiR -----END PGP SIGNATURE-----

May 17, 2008

Bug that aborts Dillo

by corvid＠lavabit.com

Jorge wrote: > Hi there, > > I reviewed my personal notes on dillo segfaults and all of them > but one are fixed now. This is: > > BUG: > http://episteme.arstechnica.com/groupee/forums/a/tpc/f/96509133 > /m/151006647731/r/264004847731 > Aborts dillo. I tried it and it crashed. Then I dug up the - requisition->width += getStyle()->boxDiffHeight (); + requisition->width += getStyle()->boxDiffWidth (); in sizeRequestImpl from http://lists.auriga.wearlab.de/pipermail/dillo-dev/2008-January/003690.html and it no longer wanted to crash for me.

May 17, 2008

Bug that aborts Dillo

by jcid＠dillo.org

Hi there, I reviewed my personal notes on dillo segfaults and all of them but one are fixed now. This is: BUG: http://episteme.arstechnica.com/groupee/forums/a/tpc/f/96509133 /m/151006647731/r/264004847731 Aborts dillo. If you wget that URL to a local file, it's easier to debug because the problem is with a requested image. A negative number is propagated down from image.cc:199 #13 0x0807d083 in dw::Image::sizeAllocateImpl (this=0x81eca98,allocation=0xbfc69fb8) at image.cc:201 (gdb) p *allocation $4 = {x = 366, y = 941, width = 3, ascent = 2, descent = 1} (gdb) p dx $5 = 8 (gdb) p dy $6 = 26607 (gdb) allocation->width - dx = -5. from there width=-5 wreaks havoc. Please someone more seasoned in the (dx,dy) offsets give it a look. -- Cheers Jorge.-

May 17, 2008

Dillo freeze

by jcid＠dillo.org

On Fri, May 16, 2008 at 04:28:12PM -0400, Jorge Arellano Cid wrote: > On Mon, May 12, 2008 at 09:29:11PM +0200, Johannes Hofmann wrote: > > On Mon, May 05, 2008 at 05:41:32PM -0400, Jorge Arellano Cid wrote: > > > Hi there, > > > > > > Dillo freezes on this page: > > > > > > http://www.kernel.org/pub/linux/kernel/v2.6/linux-2.6.24.6.tar.bz2.sign > > > > > > Dillo bug or FLTK2? > > > > Other than I initially thought it is a Dillo bug. The IOwatch for > > reading is not removed properly. As a result the select call in the > > fltk event loop keeps failing with EBADF. > > Attached is a fix for the problem, but I'm not sure whether it is > > correct. Jorge, can you please have a look at it? > > Committed. > > After a really long review. :-P > > Related CCC patches pending here... Ok, the pending patches are now committed. The Changelog reads: - CCC: added reentrancy control to the OpEnd and OpAbort operations. - CCC: enhanced the debug function and implemented OpAbort for dpi. This fixes: - A ftp problem with short files not being sent back. - Downloading short ftp files freezing the downloader. - Cancel on download dialog for ftp file aborting dillo (SEGFAULT). Maybe more... :) Please test. -- Cheers Jorge.-

May 17, 2008

website FAQ update

by corvid＠lavabit.com

Justus wrote: > Jorge Arellano Cid wrote: > > On Fri, May 16, 2008 at 08:21:25PM +0000, corvid wrote: > >> Are "dillo1" and "dillo2" official names? If so, I'll add in some text to > >> make it clear what's what. > > > > They're not official, but good mnemonics! > > > > When there was GTK1 dillo only, there was no confusion. > > Currently the CVS is using dillo-fltk, but I'd much prefer > > to have a binary named "dillo2". > > > > Probably this is a good choice. Let's hear what dillo-dev > > people say. Here's the proposal: > > > > * call the new binary "dillo2", and refer to the old 0.8.x series > > as dillo1. > > I think it's problematic to have a digit in the official name (or in the > name of the binary) since people will think it's the version. I'd > suggest something along these lines: > > * Dillo is the name of the brand > * dillo2 is the codename for the fltk branch > * for marketing reasons sell any snapshots / upcoming releases of that > branch as the all new dillo version 2.x.y with tons of new features > * the binary is called dillo (it's what people expect to type in to > start the lightweight web browser thingie [at least I tried to run > $PREFIX/bin/dillo a couple of times now...]) I also prefer to have a binary named dillo. (and dillorc2 -> dillorc)

May 17, 2008

website FAQ update

by 4winter＠informatik.uni-hamburg.de

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Jorge Arellano Cid wrote: > On Fri, May 16, 2008 at 08:21:25PM +0000, corvid wrote: >> Are "dillo1" and "dillo2" official names? If so, I'll add in some text to >> make it clear what's what. > > They're not official, but good mnemonics! > > When there was GTK1 dillo only, there was no confusion. > Currently the CVS is using dillo-fltk, but I'd much prefer > to have a binary named "dillo2". > > Probably this is a good choice. Let's hear what dillo-dev > people say. Here's the proposal: > > * call the new binary "dillo2", and refer to the old 0.8.x series > as dillo1. I think it's problematic to have a digit in the official name (or in the name of the binary) since people will think it's the version. I'd suggest something along these lines: * Dillo is the name of the brand * dillo2 is the codename for the fltk branch * for marketing reasons sell any snapshots / upcoming releases of that branch as the all new dillo version 2.x.y with tons of new features * the binary is called dillo (it's what people expect to type in to start the lightweight web browser thingie [at least I tried to run $PREFIX/bin/dillo a couple of times now...]) 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 iD8DBQFILf4BoPmwNWhsaZYRAor2AJ0U728787tvENHwVuxFW000e/DC/gCgpqZD JVaxk8o+/yROgKeKOapZlA4= =ym7X -----END PGP SIGNATURE-----

May 16, 2008

website FAQ update

by corvid＠lavabit.com

Jorge wrote: > Please try to contact J?rgen. If he doesn't answer, leave dillo-dev > as the only contact until we find somebody else. His dillo address redirects to an address where I get User Unknown. I found a telia.com address, but that also gave me User Unknown.

May 16, 2008

website FAQ update

by jcid＠dillo.org

On Fri, May 16, 2008 at 08:21:25PM +0000, corvid wrote: > Are "dillo1" and "dillo2" official names? If so, I'll add in some text to > make it clear what's what. They're not official, but good mnemonics! When there was GTK1 dillo only, there was no confusion. Currently the CVS is using dillo-fltk, but I'd much prefer to have a binary named "dillo2". Probably this is a good choice. Let's hear what dillo-dev people say. Here's the proposal: * call the new binary "dillo2", and refer to the old 0.8.x series as dillo1. -- Cheers Jorge.-

May 16, 2008

Dillo freeze

by jcid＠dillo.org

On Mon, May 12, 2008 at 09:29:11PM +0200, Johannes Hofmann wrote: > On Mon, May 05, 2008 at 05:41:32PM -0400, Jorge Arellano Cid wrote: > > Hi there, > > > > Dillo freezes on this page: > > > > http://www.kernel.org/pub/linux/kernel/v2.6/linux-2.6.24.6.tar.bz2.sign > > > > Dillo bug or FLTK2? > > Other than I initially thought it is a Dillo bug. The IOwatch for > reading is not removed properly. As a result the select call in the > fltk event loop keeps failing with EBADF. > Attached is a fix for the problem, but I'm not sure whether it is > correct. Jorge, can you please have a look at it? Committed. After a really long review. :-P Related CCC patches pending here... -- Cheers Jorge.-

May 16, 2008

website FAQ update

by corvid＠lavabit.com

Are "dillo1" and "dillo2" official names? If so, I'll add in some text to make it clear what's what. BTW, I misspelled 'publicly' :(

May 16, 2008

Uncluttering dillos UI - getting rid of the dialog boxes

by 4winter＠informatik.uni-hamburg.de

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Johannes Hofmann wrote: > Cool! I like it already better than the dialog box. > What do you mean by "on-blur" events? I like to be notified the moment the input box loses focus. For some reason the code works quite well as it is (a comment in the source code mentions this might be a fltk bug :-/), but it is possible to leave the location box without triggering the event handler. But relying on some undocumented and or buggy behavior is obviously not an option ;). I'd expect something like inputbox->when(WHEN_ENTER_KEY | UNFOCUS) but I have no experience with fltk at all and a quick look into the callback section of their documentation didn't help either. 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 iD8DBQFILei8oPmwNWhsaZYRAqidAJwMMweoNyvL+yMLbPXv6udUU7BiNQCgiSU3 VxfMpr6FsBJifzC/ecmqJ0o= =P3yn -----END PGP SIGNATURE-----

May 16, 2008

Uncluttering dillos UI - getting rid of the dialog boxes

by Johannes.Hofmann＠gmx.de

Hi, On Fri, May 16, 2008 at 09:20:17PM +0200, Justus Winter wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Hey everyone :) > > somewhere deep inside the "Future of Dillo" monster thread I suggested > to get rid of the location dialog box but Johannes pointed out that this > breaks the fullscreen mode. > > So, here is my proposed solution: > > * on ctrl-l: focus location bar > * if in fullscreen mode: temporarily show toolbars until the user > pressed enter > > I am not familiar with the fltk event model, so I just hijacked the > existing event handler which works reasonably well. It would be nice to > also capture any "on-blur" events, but I could use some hints (the fltk2 > documentation only mentions 6 WHEN_* events, though some googling > suggests that there might be UNFOCUS events... uh?). > > Anyway, the patch is attached. let me know what you think. Cool! I like it already better than the dialog box. What do you mean by "on-blur" events? Johannes

May 16, 2008

Uncluttering dillos UI - getting rid of the dialog boxes

by 4winter＠informatik.uni-hamburg.de

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hey everyone :) somewhere deep inside the "Future of Dillo" monster thread I suggested to get rid of the location dialog box but Johannes pointed out that this breaks the fullscreen mode. So, here is my proposed solution: * on ctrl-l: focus location bar * if in fullscreen mode: temporarily show toolbars until the user pressed enter I am not familiar with the fltk event model, so I just hijacked the existing event handler which works reasonably well. It would be nice to also capture any "on-blur" events, but I could use some hints (the fltk2 documentation only mentions 6 WHEN_* events, though some googling suggests that there might be UNFOCUS events... uh?). Anyway, the patch is attached. let me know what you think. 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 iD8DBQFILd5xoPmwNWhsaZYRApqEAJ9glSSkR3Q9oU0BmmUf9/ErG/ItfwCfVs5M asgEwaQAMk4ggsfzO2toTc0= =MHCi -----END PGP SIGNATURE-----

May 16, 2008

patch: some more window deletion

by jcid＠dillo.org

On Wed, May 14, 2008 at 10:42:57PM +0000, corvid wrote: > I thought I was going to have to remove the viewport from the UI > because both UI and layout would want to delete it, but it appears that > TopGroup gets informed of the deletion and everything turns out fine. Committed. It'd be good to add a comment if necessary. -- Cheers Jorge.-

May 16, 2008

website FAQ update

by jcid＠dillo.org

On Thu, May 15, 2008 at 06:49:11PM +0000, corvid wrote: > I made http://www.dillo.org/FAQ.corvid.html > The old FAQ is still at http://www.dillo.org/FAQ.html for now. > > So if you have comments/find mistakes/whatever.... Yes. > Q: I have an addition to the FAQ/website. Whom do I contact? > > Send all such additions, as well as general questions about the website, to > the webmaster J?rgen Viksell. better to use dillo-dev instead nowadays? Please try to contact J?rgen. If he doesn't answer, leave dillo-dev as the only contact until we find somebody else. > Q: Why isn't Dillo-0.8.x rendering some HTML entities (such as &euro)? Please remove this question as it is specific to dillo1. The web site, and all its information should focuse in dillo2. BTW, it'd be good to state in the FAQ that dillo1 is unmaintained. Something like: Q: What's the state of dillo1 (dillo-0.8.x series)? dillo-0.8.x is unmaintained. It depends on an obsolete library and dillo2 has achieved and surpassed all it had to offer. [In the future we may add a link to the Changelog and to the Plans page] > Q: How do I use a different charset with Dillo? > Internationalization and Localization (i18n & l10n) > > FLTK port > [...] With regard to FLTK: s/FLTK/FLTK2/g in the whole FAQ. Several people write me saying they can't compile dillo2 with FLTK1! Let's try to avoid that misunderstanding. And in this particular case, we can re-word it like this: "Since dillo was ported to FLTK2, the UTF-8 character encoding is used internally. Dillo uses the iconv ..." > Q: How do I enable the https (SSL) support plugin for dillo? Well, currently the CVS has it enabled, but it is definitely worth stating that it's a bare-bones plugin. Also, we can't make a release with HTTPS enabled until we solve the dialog spamming that happens on sites with multiple resources. This is, porting the https filter dpi plugin to a dpi server plugin. Before the release, something like: Q: What's the state of HTTPS support? First of all, beware that this is a prototype in alpha state. It will only provide for very basic web page retrievals, POST and GET. We need to port it internally fomr dpi filter to dpi server. If you want to help us to improve the plugin, please let us know! could do it. > Q: What's the status of the FLTK port? I'd say something like: Code using the FLTK2 GUI toolkit was made publically available in September 2007, and progress has since accelerated. It has become the main development branch, and the GTK1 based dillo1 (0.8.x series) went unmantained. Dillo2 provides much more than the old one. Developers are encouraged to try it out. Users are advised to wait until a release is made. _Experienced_ users may try it out. [a link to the CVS page with build directions is needed here] > Q: Why does the page X.Y look strange? It'd be good here to add something about the HTML bug meter: In the lower-right corner of Dillo you'll see a bug-shaped icon. Click on it to see the detected HTML errors. More info [add link to: http://www.dillo.org/help/bug_meter.html] That's it for me. Once done, let it be the new FAQ, and feel free to improve upon! -- Cheers Jorge.-

May 16, 2008

tabs - what are they?

by Johannes.Hofmann＠gmx.de

On Thu, May 15, 2008 at 05:10:28PM +0000, corvid wrote: > Johannes wrote: > > while thinking about tabs I realized that I'm not really sure > > what they are. > > > > * On the one hand they could be implemented just as a GUI alternative > > to multiple bowser windows. > > > > * On the other hand they could be something like temporary bookmarks > > that are easily accessible e.g. via buttons below the panel. That > > way one would only need one layout and view. > > The only issue here would be that one would expect a separate > > navigation stack for each "tab". > > This mail > http://lists.auriga.wearlab.de/pipermail/dillo-dev/2008-January/003685.html > says the latter. Right, Jorge already proposed that and it's really interesting to think about tabs as a navigation thing rather than a gui thing. Currently we have a linear navigation history. One can go back and forward. Perhaps tabs are nothing else than the possibility to fork navigation history at certain points. So it becomes a tree rather than a linear list. > > PS I read your subject line and thought, "They're eight spaces, of course." :) Yeah definately, what else :-)

May 15, 2008