- Dillo-dev - lists.mailman3.com

Re: [Dillo-dev] Re: Some news...

by Ferdi Franceschini

On Mon, Jul 14, 2003 at 02:18:49AM +0200, Diego S?enz wrote: > I have see the dpi code and think that it is a very good approach. Only > one think: how do you integrate a plug-in without patch dillo? I known > that dpi is not finished, but i am impatient. If the PIs can add menu > entries to dillo menus it can be easy to start/call a PI when a menu > entrie is clicked and the PI can ask some dillo state(like url below the > mouse) or open a save dialog or etc using a dpi tag. A cfg file can > asociate prefix(ftp://), sufix(.swf) and menu entries with PIs. It can > be modified on PI install or be generated. There must be some code for > dinamics menus in dillo for history yet. About PIs that render to a zone > in a web page it can be done like Sylpheed-claws integrate dillo in > it(GTK X sockets or something that i think). I hope to help something > even if nobody ask. At the moment there is no means for adding dpi menu entries or buttons to dillo, and it may be a long time before this is implemented. In the meantime maybe we could create a plugins page of dpi urls to launch plugins and then simply provide a link to it on the splash screen. Cheers Ferdi

July 14, 2003

*** Lowest Rate Mortgages - Personal Service *** vmigiij r

by w2E1R786gI＠process4unow.com

July 14, 2003

Selection segfault again

by Andrew A. Gill

[I'm not subscribed to this list, but I do regularly check the archives. Lets see if this gets through.] Sebastian Geerken: > > Something I forgot: The fix does not work perfectly, when, in the test > page, you start select within the text, and then move the pointer > below the text, the selection starts at the beginning. This is a known > bug, and it won't be fixed for 0.7.3. If you're referring to what I think you're referring to, I prefer to think of that as a feature. Well, now that I test it, it looks like a different feature. Go to: <http://www.planetnintendo.com/ff1/> Friggin' black-text-on-black-background! Select within the text and drag the cursor to the lower right corner of the screen. Voila! All the text within the page is selected! Now you can read it without changing the dillorc! Which reminds me--maybe there should be a way to change page layout per page. -- |Andrew A. Gill |<superluser(a)frontiernet.net>

July 14, 2003

dillo size is 1.2mb ?

by Nagaraja

Hi All, I got dillo compiled on arm with natively build toolchain, when I check the size of dillo it is 1.2mb. my configure option is: ./configure --with-gtk-prefix=/etc/gtk --x-includes=/usr/X11R6/include --x-libraries=/opt/toolchain/arm-linux/lib/X11 --with-jpeg-lib=/opt/toolchain/arm-linux/lib when I do file dillo ....... I am getting. dillo: ELF 32-bit LSB executable, Advanced RISC Machines ARM, version 1, dynamically linked (uses shared libs), stripped nothing seems satically linked. could any one please tell me what could be the cause..... Thanks and regards Nags

July 14, 2003

Re: [Dillo-dev] Re: Some news...

by TheStar

On Mon, 14 Jul 2003 10:18 am, Diego Sáenz wrote: > I have see the dpi code and think that it is a very good approach. Only > one think: how do you integrate a plug-in without patch dillo? I known > that dpi is not finished, but i am impatient. If the PIs can add menu > entries to dillo menus it can be easy to start/call a PI when a menu > entrie is clicked and the PI can ask some dillo state(like url below the > mouse) or open a save dialog or etc using a dpi tag. A cfg file can > asociate prefix(ftp://), sufix(.swf) and menu entries with PIs. It can > be modified on PI install or be generated. There must be some code for > dinamics menus in dillo for history yet. About PIs that render to a zone > in a web page it can be done like Sylpheed-claws integrate dillo in > it(GTK X sockets or something that i think). I hope to help something > even if nobody ask. > > Diego. > Also, would the dpi code also allow shockwave, java, client-side scripts, and other plugins to be added, if re-written (ie, utilising kaffe for java). I know that the plugins themselves will take a LOT of work, but I'm asking about the actual dpi...

July 14, 2003

Re: [Dillo-dev] Re: Some news...

by Diego Sáenz

I have see the dpi code and think that it is a very good approach. Only one think: how do you integrate a plug-in without patch dillo? I known that dpi is not finished, but i am impatient. If the PIs can add menu entries to dillo menus it can be easy to start/call a PI when a menu entrie is clicked and the PI can ask some dillo state(like url below the mouse) or open a save dialog or etc using a dpi tag. A cfg file can asociate prefix(ftp://), sufix(.swf) and menu entries with PIs. It can be modified on PI install or be generated. There must be some code for dinamics menus in dillo for history yet. About PIs that render to a zone in a web page it can be done like Sylpheed-claws integrate dillo in it(GTK X sockets or something that i think). I hope to help something even if nobody ask. Diego. El Mon, 7 Jul 2003 15:35:30 -0400 (CLT) Jorge Arellano Cid <jcid(a)softhome.net> escribio: > > Hi! > > > > > This has been a very busy week. I didn't anticipate how much > > work the dpid (dpi daemon) integration could take. Don't panic > > though!, the code that binds it all is running (to some extent) > > in my machine. The time consuming problem is that while making > > the integration, a lot of design matters spring in, and they take > > a lot of time to solve. > > > > The good news is that it all starts to shape into a more clear > > design (the code is still hackish at parts, but it is clear what, > > where and how to fix it). > > > > While working the gory details of it, I realized it'd be a good > > idea to make the 0.7.3 release with what we have now (plus a > > little bug fix we're working on with Sebastian). That way, > > there'll be less pressure on us while polishing the new dpid > > implementation, and hey!, it's been a long time since our last > > release. > > > > I'll keep the CVS free of the dpid code until the 0.7.3 release > > is done. For the meanwhile, I'll provide a tarball to the new > > code. (and if I can't wrap the new code before Monday, I'll > > provide a link to the latest dpid testing tarball I have). > > > First, the final code for the 0.7.3 is on the CVS (unless you > find a bug in it). It will be released by Wednesday, so please > give it a hard test today and tomorrow! > > With regard to 0.8.0-pre (aka dillo with dpid), here's the good > news: > > * Finally I have a tarball that does bookmarks and downloads > with dpid (using wget)! It can handle multiple plugins and is > working smoothly. > > I want to add one more plugin (at least) and update the > documentation before commiting it to the CVS (after releasing > 0.7.3 of course). > > For those very interested in studying it, please review: > > http://dillo.org/download/dpid4.tgz > > (not my tarball but the dpid testing suite itself) > > Cheers > Jorge.- > > _______________________________________________ > Dillo-dev mailing list > Dillo-dev(a)lists.auriga.wearlab.de > http://lists.auriga.wearlab.de/cgi-bin/mailman/listinfo/dillo-dev

July 14, 2003

Re: [Dillo-dev] Another delay for 0.7.3 :(

by Sebastian Geerken

On Sun, Jul 13, Joe Crayne wrote: > Does this fix it? > > > --- source.cvs/src/dw_page.c 2003-07-12 14:17:12.000000000 -0400 > +++ source.fix/src/dw_page.c 2003-07-13 00:16:32.000000000 -0400 > @@ -451,7 +451,7 @@ static void Dw_page_get_extremes (DwWidg > /* no rewrap necessary -> values in lines are up to date */ > line = &page->lines[page->num_lines - 1]; > if (page->words[line->first_word].style->nowrap) > - extremes->min_width = MAX (line->max_word_min, line->par_min); > + extremes->min_width = MAX (line->max_line_width, line->par_min); > else > extremes->min_width = line->max_word_min; > extremes->max_width = MAX (line->max_par_max, line->par_max); No, you cannot compare minimal widths (line->par_min) with actual widths (line->max_line_width), the problem is another one. I've just committed a fix, I'll summarize it here. The first part is: @@ -450,10 +450,11 @@ static void Dw_page_get_extremes (DwWidg } else if (page->wrap_ref == -1) { /* no rewrap necessary -> values in lines are up to date */ line = &page->lines[page->num_lines - 1]; - if (page->words[line->first_word].style->nowrap) - extremes->min_width = MAX (line->max_word_min, line->par_min); - else - extremes->min_width = line->max_word_min; + /* Historical note: The former distinction between lines with and without + * words[first_word]->nowrap set is no longer necessary, since + * Dw_page_real_word_wrap sets max_word_min to the correct value in any + * case. */ + extremes->min_width = line->max_word_min; extremes->max_width = MAX (line->max_par_max, line->par_max); } else { /* Calculate the extremes, based on the values in the line from Why it is not neccessary anymore, is clear from the second part (simplified): @@ -907,9 +916,15 @@ static void Dw_page_real_word_wrap (DwPa if (!new_par) last_line->par_max += last_space; if (word->style->nowrap) { last_line->par_min += word_extremes.min_width; + /* This may also increase the accumulated minimum word width. */ + last_line->max_word_min = + MAX (last_line->max_word_min, last_line->par_min); + /* NOTE: Most code relies on that all values of nowrap are equal for all + * words within one line. */ } else /* Simple case. */ last_line->max_word_min = MAX (last_line->max_word_min, word_extremes.min_width); Actually, this bug has existed for a longer time, what I assume (but did not investigate further) is that the fixed code was never used before, i.e. instead of handling this case: } else if (page->wrap_ref == -1) { /* no rewrap necessary -> values in lines are up to date */ [...] this case: } else { /* Calculate the extremes, based on the values in the line from where a rewrap is necessary. */ [...] was *always* handled, due to the rewrapping bug I fixed some time ago. So this bug was hid by another one. There is still another bug: At http://bbs.archlinux.org/viewtopic.php?t=771 the text is not rewapped sometimes. I'm working on it. Sebastain

July 13, 2003

Re: [Dillo-dev] Another delay for 0.7.3 :(

by Joe Crayne

On Thu, Jul 10, 2003 at 01:27:48PM -0400, Jorge Arellano Cid wrote: > > So the 0.7.3 release will be held until I have some news about > this bug. (Jorge, sorry for the double emails, I meant to send to the list before.) Does this fix it? --- source.cvs/src/dw_page.c 2003-07-12 14:17:12.000000000 -0400 +++ source.fix/src/dw_page.c 2003-07-13 00:16:32.000000000 -0400 @@ -451,7 +451,7 @@ static void Dw_page_get_extremes (DwWidg /* no rewrap necessary -> values in lines are up to date */ line = &page->lines[page->num_lines - 1]; if (page->words[line->first_word].style->nowrap) - extremes->min_width = MAX (line->max_word_min, line->par_min); + extremes->min_width = MAX (line->max_line_width, line->par_min); else extremes->min_width = line->max_word_min; extremes->max_width = MAX (line->max_par_max, line->par_max);

July 13, 2003

Re: [Dillo-dev] selection segfault again

by Sebastian Geerken

I wrote: > FYI: I seem to have fixed the bug, expect it to be committed > today. Those interested should look at the differences in the code, it > is only a simple coding bug, nothing very interesting. (Has been committed.) Something I forgot: The fix does not work perfectly, when, in the test page, you start select within the text, and then move the pointer below the text, the selection starts at the beginning. This is a known bug, and it won't be fixed for 0.7.3. Sebastian

July 12, 2003

Re: [Dillo-dev] selection segfault again

by Sebastian Geerken

On Sat, Jul 12, Jamin W. Collins wrote: > On Sat, Jul 12, 2003 at 04:58:42PM +0200, Sebastian Geerken wrote: > > > Since the </table> tag is mandatory, we are not obliged to handle > > this. Anyway, a missing </table> tag will leave a well-defined widget > > tree, which the selection must be able to handle. > > While not required to handle it, the application should *never* segfault > when provided bad data. Dillo does not segfault anymore ;-) (Fix has been committed.) The problem is unrelated to the missing </table>, there may be *valid* pages which would have caused dillo to segfault due to the same bug. Sebastian

July 12, 2003