May 2009 - Dillo-dev - lists.mailman3.com

whitespace entities inside words

by jcid＠dillo.org

On Fri, May 08, 2009 at 03:53:03AM +0000, corvid wrote: > Is a whitespace entity inside a word supposed to be > treated like normal whitespace? > > I thought it would be, but first I saw that > adjacent   ... aren't collapsed, > then I realized that we are leaving the > space inside the word instead of calling > addSpace, and then I saw that PrevWasSPC > isn't set if the space was at the end of the > word. > > If it's just something that no one ever got > around to dealing with, I'll make a patch. That hits into a deeper problem: the treatment of whitespace is not well defined inside the textblock yet. The same problem arises when handling NBSP. There's no clear definition on how to handle/flag it. If you're willing to study the textblock word wrapping, and the whitespace problem, great! I think the next big step for dillo is to handle DIV as a floating word (so it can be rendered side by side, as intended), and AFAIS that needs changes in the textblock. This is a task for 2.2. -- Cheers Jorge.-

May 8, 2009

css and font specification [Re: Dillo2 + FLTK2 + NxLib running on mips linux]

by Johannes.Hofmann＠gmx.de

On Fri, May 08, 2009 at 03:27:56PM +0000, corvid wrote: > ----- Forwarded message from ?? ? ----- > > I have try current version dillo + fltk base on X11 in X86 PC (Ubuntu) > > with LC_LANG=zh_TW.UTF-8. > > > > Using default font_xxxx I can see chinese text and write chinese characters to text editor also works. > > > > But, I cannot see any chinese text after google search ??? > > The same URL works on dillo + fltk base on nxlib. > > > > What I think is the FLTK already supports multi bytes characters well. > > The problem is caused by others (I mean fonts / X11 configuration etc.) > > Am I right ? > > (This came with screenshots showing that www.google.com.tw doesn't work.) > > If dillo doesn't parse embedded css, it works for me. > The "body,td,a,p,.h{font-family:arial,sans-serif}" in the page overrides > my font_* prefs. > So I tried to add a !important rule to set font-family in style.css, e.g. > * {font-family: "Bitstream Cyberbit" !important} Hm, this seems to work fine for me. Maybe you have other stuff in your style.css that confuses the CSS parser? > but either it's impossible or I was getting the syntax wrong somehow. Looks ok to me and works here. What does xlsfonts | grep -i cyberbit say? Cheers, Johannes

May 8, 2009

css and font specification [Re: Dillo2 + FLTK2 + NxLib running on mips linux]

by corvid＠lavabit.com

----- Forwarded message from ?? ? ----- > I have try current version dillo + fltk base on X11 in X86 PC (Ubuntu) > with LC_LANG=zh_TW.UTF-8. > > Using default font_xxxx I can see chinese text and write chinese characters to text editor also works. > > But, I cannot see any chinese text after google search ??? > The same URL works on dillo + fltk base on nxlib. > > What I think is the FLTK already supports multi bytes characters well. > The problem is caused by others (I mean fonts / X11 configuration etc.) > Am I right ? (This came with screenshots showing that www.google.com.tw doesn't work.) If dillo doesn't parse embedded css, it works for me. The "body,td,a,p,.h{font-family:arial,sans-serif}" in the page overrides my font_* prefs. So I tried to add a !important rule to set font-family in style.css, e.g. * {font-family: "Bitstream Cyberbit" !important} but either it's impossible or I was getting the syntax wrong somehow. (This is not necessarily the same problem he's having, but...)

May 8, 2009

Image invalid access problems

by onepoint＠starurchin.org

It's too early to be absolutely certain, but from the logging data I've been collecting it seems that invalid DilloImage accesses always come from this call sequence: a_Bw_stop_clients() -> a_Capi_stop_client() -> a_Cache_stop_client() -> Cache_client_dequeue() -> a_Web_free() -> a_Image_unref() -> Image_free() -> dFree() So the question is: why isn't this call sequence stopping the image callbacks from running before it dFree()s the image? When Cache_process_queue() sees that all the data has arrived for a client it executes "(Client->Callback)(CA_Close, Client);" before calling Cache_client_dequeue(). Perhaps the fix is for a_Cache_stop_client() to close the client too? Regards, Jeremy Henty PS: Interestingly, a_Bw_stop_clients() has a comment: /* * Stop the active clients of this bw's top page. * Note: rendering stops, but the cache continues to be fed. */ This seems to be talking about our problem: the cache is still running and therefore accessing the client, but the client has been freed.

May 8, 2009

[patch]: replace DW2TB with HT2TB

by onepoint＠starurchin.org

I noticed that the convenience macro DW2TB is always used as DW2TB(html->dw) so it makes sense to use an even more convenient macro HT2TB to reduce DW2TB(html->dw) to HT2TB(html). Patch attached. Regards, Jeremy Henty

May 8, 2009

whitespace entities inside words

by corvid＠lavabit.com

Is a whitespace entity inside a word supposed to be treated like normal whitespace? I thought it would be, but first I saw that adjacent   ... aren't collapsed, then I realized that we are leaving the space inside the word instead of calling addSpace, and then I saw that PrevWasSPC isn't set if the space was at the end of the word. If it's just something that no one ever got around to dealing with, I'll make a patch.

May 8, 2009

change FLTK scheme for in dillo2

by akhilman＠gmail.com

?? Fri, 8 May 2009 00:37:50 +0200 Johannes Hofmann <Johannes.Hofmann(a)gmx.de> ????????: > What do people think about reading X resources as in attached patch > (mainly copied from commented fltk2 code)? > It allows to specify e.g. > > dillo.background:red > dillo.foreground:white > > in your .Xdefaults Will be great for me. I'm using .Xdefaults to tweak colour scheme for urxvt and other programs such as xpdf and puredata. -- ?????? ??????????? aka AkhIL ??? ??? 8 09:24:47 KRAST 2009 Fri May 8 01:24:47 UTC 2009 ---------------------------------- Visit my home page http://akhilman.blogspot.com/ jabber: akhil(a)jabber.ru ---------------------------------- ??? ????????? ????????? ????????? ??, ??? ?? ??????? ?????????, ?????? ?? ???? ????, ??? ?? ???????????? ??? ?????????. (?.?????????) ---------------------------------- Linux artstation 2.6.28-gentoo-r5 #1 SMP Sat May 2 18:40:24 KRAST 2009 x86_64 AMD Phenom(tm) 9550 Quad-Core Processor AuthenticAMD GNU/Linux up 5 days, 14:19, 4 users, load average: 0.43, 0.50, 0.52

May 8, 2009

change FLTK scheme for in dillo2

by Johannes.Hofmann＠gmx.de

On Thu, May 07, 2009 at 03:44:19PM +0000, corvid wrote: > IL'dar wrote: > > Can I change colours of dialogues and scrollbar in dillo2? > > I would like to gets something less bright. > > Even something like fluid2's -bg option will be ok. > > Not currently. > > I would not necessarily be opposed to making ui_fg and ui_bg prefs and > adding a bunch of [class]::default_style->[whatever]color() > to dillo.cc, although I wouldn't be surprised if we ran into our > share of places where fltk hardcodes things to, say, GRAY75. What do people think about reading X resources as in attached patch (mainly copied from commented fltk2 code)? It allows to specify e.g. dillo.background:red dillo.foreground:white in your .Xdefaults Cheers, Johannes

May 8, 2009

Applause: image resizing fix

by onepoint＠starurchin.org

I spend a lot of time surfing the DoomWorld picture viewer[1] and so I'm used to the thumbnails at the bottom displaying random small chunks of the image instead of anything useful. But now I'm seeing proper previews, just as the site designers intended. It's *really* nice to see this working properly! Dillo just keeps getting better! Regards, Jeremy Henty [1]: eg. http://www.doomworld.com/php/screenie.php?dir=/images/newstuff/341/&filenam…

May 8, 2009

[patch]: more cleanups

by jcid＠dillo.org

On Mon, May 04, 2009 at 06:59:34AM +0100, Jeremy Henty wrote: > > Three more cleanup patches: > > 1: Inline Cache_client_make_key() into Cache_client_enqueue(). Why > have a separate function just to increment a variable? It's just > as clear (and less noisy) to increment a static variable inside > Cache_client_enqueue(). Committed. > 2: The width and height arguments of a_Image_new() do nothing. The > function itself ignores them and its callers always set them to 0. > Since neither the function or its callers seem to care what these > arguments are, why not remove them? Committed. > 3: a_Web_dispatch_by_type() has a strange convention for its return > type, it returns -1 to indicate failure. It's less confusing to > use the common convention of a boolean: 1 indicates success and 0 > indicates failure. It's not strange. Just like libc: 0 success, < 0 error, > 0 success with return code. -- Cheers Jorge.-

May 7, 2009