- Dillo-dev - lists.mailman3.com

revision control and other project infrastructure

by dillo-dev＠toykeeper.net

Hello. I noticed some recent messages about revision control and possibly switching to a DVCS. This makes me happy. :) So, I was wondering about general project infrastructure... things like bug trackers, todo lists, mailing lists, help forums, release management, news, patch queues and code review, code repositories, etc. What I've found so far is the following: dillo.org: - general project info - release tarballs - bug list - FAQ auriga.wearlab.de: - mailing lists - cvs freshmeat.net: release announcements freenode: IRC (inactive?) http://misc.andi.de1.cc/dillo/ : Packaging for Debian, Ubuntu, Openmoko http://www.hyperborea.org/software/dillo/ : Packaging for Red Hat http://freehg.org/u/dillo/css-prototype : Prototype CSS branch Is that about right? Did I miss anything? -- Scott

Dec. 17, 2008

dillo_dbg_rendering

by Johannes.Hofmann＠gmx.de

On Tue, Dec 16, 2008 at 12:11:47AM +0000, corvid wrote: > I'm finally bored enough to copy the cmdline code from dillo1. > > Does anyone care about the dillo_dbg_rendering option? > It forces borders in table.cc. I would remove it - at least once we add css support. table, td {border: 1px solid black !important} in the user stylesheet has the same effect. Cheers, Johannes

Dec. 16, 2008

dillo_dbg_rendering

by corvid＠lavabit.com

I'm finally bored enough to copy the cmdline code from dillo1. Does anyone care about the dillo_dbg_rendering option? It forces borders in table.cc.

Dec. 16, 2008

shaded colors and css

by Johannes.Hofmann＠gmx.de

On Sun, Dec 14, 2008 at 08:00:48PM +0000, corvid wrote: > Johannes wrote: > > On Sun, Nov 23, 2008 at 04:13:56AM +0000, corvid wrote: > > > Johannes wrote: > > > > mapping css properties to dw::core::style::Style members went very > > > > smooth, because Style was obviously implemented with css in mind. > > > > > > > > Just for the shaded color thing I can't find a corresponding concept > > > > in css. > > > > What exactly is the definition of a shaded color? Is it just a bit > > > > darker than the normal one? > > > > Does anyone know the corresponding concept in css? > > > > > > I had figured the shaded colors were just for giving > > > tables and rulers a three-dimensional appearance. > > > > Right, it seems to be used in drawBorder() in style.cc. > > Only now I noticed that the only difference between > > shaded colors and normal colors is that for shaded ones two more > > variations (SHADING_DARK and SHADING_LIGHT) are computed in addition > > to SHADING_NORMAL and SHADING_INVERSE. > > So it seems that it's an optimization to avoid calls to > > Color::shadeColor(). > > > > I'd like to get rid of the two types of colors and use always shaded > > colors. > > The current scheme adds complexity and can lead to errors > > (e.g. a normal color is passed to drawBorder(). > > > > In addition we can get more sharing of colors if they don't have a > > type member. > > > > If needed we could call Color::shadeColor() only on demand, but hide > > that optimization in the Color class. > > > > Any opinions? > > No objections here. > Ok, I pushed a corresponding change to css-prototype. Cheers, Johannes

Dec. 15, 2008

shaded colors and css

by corvid＠lavabit.com

Johannes wrote: > On Sun, Nov 23, 2008 at 04:13:56AM +0000, corvid wrote: > > Johannes wrote: > > > mapping css properties to dw::core::style::Style members went very > > > smooth, because Style was obviously implemented with css in mind. > > > > > > Just for the shaded color thing I can't find a corresponding concept > > > in css. > > > What exactly is the definition of a shaded color? Is it just a bit > > > darker than the normal one? > > > Does anyone know the corresponding concept in css? > > > > I had figured the shaded colors were just for giving > > tables and rulers a three-dimensional appearance. > > Right, it seems to be used in drawBorder() in style.cc. > Only now I noticed that the only difference between > shaded colors and normal colors is that for shaded ones two more > variations (SHADING_DARK and SHADING_LIGHT) are computed in addition > to SHADING_NORMAL and SHADING_INVERSE. > So it seems that it's an optimization to avoid calls to > Color::shadeColor(). > > I'd like to get rid of the two types of colors and use always shaded > colors. > The current scheme adds complexity and can lead to errors > (e.g. a normal color is passed to drawBorder(). > > In addition we can get more sharing of colors if they don't have a > type member. > > If needed we could call Color::shadeColor() only on demand, but hide > that optimization in the Color class. > > Any opinions? No objections here.

Dec. 14, 2008

shaded colors and css

by Johannes.Hofmann＠gmx.de

On Sun, Nov 23, 2008 at 04:13:56AM +0000, corvid wrote: > Johannes wrote: > > mapping css properties to dw::core::style::Style members went very > > smooth, because Style was obviously implemented with css in mind. > > > > Just for the shaded color thing I can't find a corresponding concept > > in css. > > What exactly is the definition of a shaded color? Is it just a bit > > darker than the normal one? > > Does anyone know the corresponding concept in css? > > I had figured the shaded colors were just for giving > tables and rulers a three-dimensional appearance. Right, it seems to be used in drawBorder() in style.cc. Only now I noticed that the only difference between shaded colors and normal colors is that for shaded ones two more variations (SHADING_DARK and SHADING_LIGHT) are computed in addition to SHADING_NORMAL and SHADING_INVERSE. So it seems that it's an optimization to avoid calls to Color::shadeColor(). I'd like to get rid of the two types of colors and use always shaded colors. The current scheme adds complexity and can lead to errors (e.g. a normal color is passed to drawBorder(). In addition we can get more sharing of colors if they don't have a type member. If needed we could call Color::shadeColor() only on demand, but hide that optimization in the Color class. Any opinions? Cheers, Johannes

Dec. 14, 2008

Plans for the near future

by francis＠daoine.org

On Thu, Dec 11, 2008 at 10:24:38AM +0000, Jeremy Henty wrote: > On Wed, Dec 10, 2008 at 04:17:02PM -0300, Jorge Arellano Cid wrote: > > 2.- Basic authentication > > a) I have to review this patch after working on point 1.- > > b) This is a very requested feature after CSS. > > It'd be great to have it working. > > It works locally on various web servers (Apache and others) and also > on the LinuxFromScratch wiki. So (subject to any further comments I > get) I think it is working. My test machine has the wrong autotools, so I'm afraid I can't built and test this myself, but: does it work as expected if an img on a page requires authentication? As far as I recall (hazy memory) the unofficial patches for dillo1 didn't. If this one does, then great! Cheers, f -- Francis Daly francis(a)daoine.org

Dec. 12, 2008

Plans for the near future

by jcid＠dillo.org

On Thu, Dec 11, 2008 at 10:24:38AM +0000, Jeremy Henty wrote: > On Wed, Dec 10, 2008 at 04:17:02PM -0300, Jorge Arellano Cid wrote: > > > 2.- Basic authentication > > a) I have to review this patch after working on point 1.- > > b) This is a very requested feature after CSS. > > It'd be great to have it working. > > It works locally on various web servers (Apache and others) and also > on the LinuxFromScratch wiki. So (subject to any further comments I > get) I think it is working. If anyone has an account they'd like me > to test then I'll try it out. Good. > > 5.- Close icon in Tab label > > a) This is much needed as anyone using Tabs has experienced. > > b) Not too hard to implement, but I have to work on 1 and 2. > > I'd like to do this but I don't see how to get a button into the tab > label. It just displays the widget's label(). I wondered if there > was some Symbol magic that might do it but Symbols don't seem to > respond to events. Am I missing something? Maybe the close button > will have to go in the tab body. Perhaps to the right of the web > search bar? (Top-right seems to be the traditional location.) My original idea was to use a '+' symbol rotated 45 degrees, then to override the virtual handle method to react when the click was in its "zone". Just an idea, I haven't coded anything. There was an aqua-like tab widget somewhere. I was suggested to look at it for insight. I don't have the reference now. :( -- Cheers Jorge.-

Dec. 12, 2008

Plans for the near future

by onepoint＠starurchin.org

On Wed, Dec 10, 2008 at 04:17:02PM -0300, Jorge Arellano Cid wrote: > 2.- Basic authentication > a) I have to review this patch after working on point 1.- > b) This is a very requested feature after CSS. > It'd be great to have it working. It works locally on various web servers (Apache and others) and also on the LinuxFromScratch wiki. So (subject to any further comments I get) I think it is working. If anyone has an account they'd like me to test then I'll try it out. > 5.- Close icon in Tab label > a) This is much needed as anyone using Tabs has experienced. > b) Not too hard to implement, but I have to work on 1 and 2. I'd like to do this but I don't see how to get a button into the tab label. It just displays the widget's label(). I wondered if there was some Symbol magic that might do it but Symbols don't seem to respond to events. Am I missing something? Maybe the close button will have to go in the tab body. Perhaps to the right of the web search bar? (Top-right seems to be the traditional location.) > 6.- Switch to mercurial > a) From comments it looks like switching to Hg is a good idea. > b) Yes, I'm only updating Mercurial these days as an experiment. > c) Switching as soon as I have some time to do it. Probably after > 1.- and 2.- Cool! Jeremy Henty

Dec. 11, 2008

Dillo and Gmail

by daniel.victoria＠gmail.com

Wow, that's a lot of things for a newbie to do... I guess I'll just keep firefox along with dillo. One for gmail and the other for quick things Thanks for your help anyway Daniel On Fri, Dec 5, 2008 at 11:56 AM, Francis Daly <francis(a)daoine.org> wrote: > On Thu, Dec 04, 2008 at 10:17:00PM -0200, Daniel Victoria wrote: > > Hi there, > >> The only thing I'm missing is Gmail. Does Dillo work with gmail? Do I >> need to do any special tricks? > > I have it "working" with dillo1, and don't see why it shouldn't work in > the same way with dillo2. > > "Working" means "I can read and write mail, http only" -- but only after > I've got the authentication cookie with a different browser. (My dillo1 > doesn't do https.) > > If that's good enough for you, then you can try the same with dillo2 > and report on whether it still works. > > My cookiesrc says: > > DEFAULT DENY > mail.google.com ACCEPT > > I use a mozilla browser to sign in, and then I don't sign out so as > not to invalidate the cookie on the server. It seems to remain valid > for a couple of weeks, so I occasionally have to redo this part. Find > the cookie called GX, and write it to a file in the correct format -- > 7 tab-separated fields. My current one is a file ~/gmcookie which looks > like this (apart from the value of the cookie, which has been modified): > > mail.google.com TRUE /mail FALSE 2000000000 GX DQAAAIIAAACY9ABCDEFABCDEFABCDEFABCDEFABCDEFABCDEFABCDEFABCDEFABCDEFABCDEFABCDEFHELLOKITTYEFABCDEFABCDEFABCDEFABCDEFABCDEFABCDEFABCDEFABCDEFABCDEFABCDEFABCDEFABCDEFABCDEFABCDEFABCDEFABCDEFABCDEF__p_abcdef > > Then, to load or reload the cookie into dillo, I do > > pkill -HUP cookie > pkill cookie > cat $HOME/gmcookie >> $HOME/.dillo/cookies.txt > > And then I go to my bookmarked gmail page at > > http://mail.google.com/mail/?ui=html&zy=n > > and it all Just Works. If I get redirected to a https page, that indicates > that the cookie is absent or invalid, so I put it back or get a fresh one. > > No guarantees that it will keep working, of course. The gmail folk can > change their authentication requirements at any stage. And you miss out on > the javascript goodness that is a large part of the gmail experience. And > there was a stage when the html returned was broken in a way that dillo > couldn't usefully handle, although that has since been changed back. But for > straightforward mail tasks, it works well enough for me. > > Couple it with a lightweight gmail-notifier script, and you don't have > to hit reload to see the new news. > > Good luck, > > f > -- > Francis Daly francis(a)daoine.org >

Dec. 11, 2008