Hi, Now that dillo 0.8.1 is done, I want to clarify what happened with the Tabs/Frames/KeybNav patch. Sometime has passed since, but at the time we decided to not stir an already ungrateful situation. Here I'm forwarding a couple of emails that should be enough to do it (dillo-dev archives holds the rest). Sincerely Jorge.- -------------------------------------------------------------------------- From akemnade@informatik.uni-bremen.de Mon May 17 08:44:51 2004 Date: Sat, 27 Mar 2004 14:25:31 +0100 From: Andreas Kemnade <akemnade@informatik.uni-bremen.de> To: Sebastian Geerken <s.geerken@ping.de> Cc: jcid@dillo.org Subject: Re: [Dillo-dev]Frame support in CVS Hi Sebastian, Hi Jorge, On Sun, 7 Mar 2004 22:14:49 +0100 Sebastian Geerken <s.geerken@ping.de> wrote:
Hi Andreas.
(I'm writing this in English to make forwarding etc. simpler.)
Jorge and I would indeed like to incorporate Frank's work into the main trunk, but unfortunately Frank refused to clean up the code, and finally, he quit the project, and plans to make an own version of dillo. I've a mail of 8 Jan, which I can forward you, if you are interested.
Please send me that mail. Ok, perhaps I was not deep enough in the archives. I just did not find any statement about what has to be done with the patch. I think other readers of dillo-dev do not know about that, too. Perhaps it would be better if that was more clarified but I also understand that you do not want to repeat yourself too often.
If you want to commit some work, you may clean up the patch, especially split it up into several parts. Most important, there are three parts: DilloDoc (which I'd like to be renamed), frames, and tabs. They are of course related, but it would be handy to have single patches for them (because of the over-linear correlation of patch size and review time).
Most the time I was just happy to be able to apply the frames patch. I was not forced to look at the code. But with the massive conflicts the patch produces when I applied it to 0.8 I really had to start to look at the patch. Sometimes being not a maintainer has really big advances... If I checked the patch size before fixing the conflicts I guess I would have given up. I have never noticed before how big the patch was. If splitting does not work, then there is something really wrong with the code. So let's divide and conquer. I think an important part is to remove nice (but not very important) features from the patch which can be added separetely later on. I think I can find time in the next month to split and clean up the frame patch. Greetings Andreas Kemnade -------------------------------------------------------------------------- From s.geerken@ping.de Mon May 17 08:44:37 2004 Date: Sat, 10 Jan 2004 13:04:11 +0100 From: Sebastian Geerken <s.geerken@ping.de> To: Jorge Arellano Cid <jcid@dillo.org> Subject: Fwd: Re: Notes on keyboard navigation FYI. [ Part 2: "Included Message" ] Date: Thu, 08 Jan 2004 22:31:10 +0100 From: Frank de Lange <frank@unternet.org> To: Sebastian Geerken <s.geerken@ping.de> Subject: Re: Notes on keyboard navigation Sebastian Geerken wrote:
Hi Frank.
First of all, a happy new year!
Several notes on keyboard navigation:
1. Jorge and I decided, that we do not immediately support keyboard configuration, also because it is rather unclear what widget toolkit will be used in the future. So the version of keyboard navigation, which will be integrated the first time, should have hard-coded keys.
2. Some days ago, I committed some changes to iterators, you may be interested in. There is some description within doc/Dw.txt, section "Anchors and Scrolling", subsection "Scrolling". Especially, scrolling has now two dimensions, there is a new scrolling position, DW_[HV]POS_INTO_VIEW, and DwIterator::scroll_to has been replaced by the more general DwIterator::get_allocation.
3. I noticed some performance problems with stress tests, I'm not sure they play a role in real life. Anyway, you wrote on the list that you are already working on in.
Sebastian
Happy new year as well... and thanks for the notice. For keyboard navigation to work in a sensible way x coördinate support is useful... so useful that I had already implemented it. The performance problems have been solved in my tree as well, as I have stated in my last message to the list. They are caused by the use of g_list_search_custom to find the focusable in focus.c. Using a hash table solved this. But... I have decided to discontinue support to the Dillo project as of the 'infringer' message to the list, so I will no longer post or update patches to Dillo. Instead, I am working on a new browser which should fill the same niche as Dillo (small, fast, enough features for most sites so it can be used as main browser). Starting point for this is my patched version of Dillo, with additional changes: - dpi and dpid and related things are gone (DONE) - library plugins for everything dpi did, and more (mime decoders, schemes, etc) (DONE) - ccc is on the way out, to be replaced by something less convoluted and more general (WORK IN PROGRESS) - DOM support with new xml parser (WORK IN PROGRESS) The project currently uses GTK+ as toolkit, but might switch to something else (FLTK is a candidate). A side goal for this project is to be developer-friendly. Patches are welcome, they will get their own space on the site (which will be on savannah, which is still not completely recovered from the break-in yet... hence the delay). I may post a short notice to the list if/when the project is up and ready for testing. Good luck with Dillo. Cheers//Frank -- WWWWW ________________________ ## o o\ / Frank de Lange \ }# \| / +46-734352015 \ \ `--| _/ <Hacker for Hire> \ `---' \ +31-640037120 / \ frank@unternet.org / `------------------------' [ "Omnis enim res, quae dando non deficit, dum habetur et non datur, nondum habetur, quomodo habenda est." ]