- Dillo-dev - lists.mailman3.com

Re: [Dillo-dev]0.8.3rc1 test

by Florent BERANGER

Hi, see the http://fr.wikipedia.org web page for exemple. Best regards, Florent ---------- Initial Header ----------- From : dillo-dev-admin(a)lists.auriga.wearlab.de To : dillo-dev(a)auriga.wearlab.de Cc : Date : Fri, 1 Oct 2004 20:03:46 -0400 Subject : Re: [Dillo-dev]0.8.3rc1 test > > What's bad about that code? HTML 4 is, in the absence of headers or > > meta tags indicating otherwise, assumed to be ISO-8859-1. The é is > > perfectly legal there. > > If you give Dillo an ISO-8859-1 'é', it will work fine. > > Now, if you give it in some other encoding, it will fail. > > > > > Given that they show up correctly when I look at that site here, I > > suspect it's a font problem on the display, not an HTML or Dillo problem. > > Yes, it could be. BTW, if the source page has an ISO-8859-1 'é' > this is surely the problem. Sorry, I should have said: "if the source page has a NON ISO-8859-1 'é', this is surely the problem" -- Cheers Jorge.- _______________________________________________ Dillo-dev mailing list Dillo-dev(a)lists.auriga.wearlab.de http://lists.auriga.wearlab.de/cgi-bin/mailman/listinfo/dillo-dev ************************ ADSL ILLIMITE TISCALI + TELEPHONE GRATUIT ************************ Surfez 40 fois plus vite pour 30EUR/mois seulement ! Et téléphonez partout en France gratuitement, vers les postes fixes (hors numéros spéciaux). Tarifs très avantageux vers les mobiles et l'international ! Pour profiter de cette offre exceptionnelle, cliquez ici : http://register.tiscali.fr/adsl (voir conditions sur le site)

Oct. 2, 2004

Re: [Dillo-dev]0.8.3rc1 test

by Jorge Arellano Cid

> > What's bad about that code? HTML 4 is, in the absence of headers or > > meta tags indicating otherwise, assumed to be ISO-8859-1. The é is > > perfectly legal there. > > If you give Dillo an ISO-8859-1 'é', it will work fine. > > Now, if you give it in some other encoding, it will fail. > > > > > Given that they show up correctly when I look at that site here, I > > suspect it's a font problem on the display, not an HTML or Dillo problem. > > Yes, it could be. BTW, if the source page has an ISO-8859-1 'é' > this is surely the problem. Sorry, I should have said: "if the source page has a NON ISO-8859-1 'é', this is surely the problem" -- Cheers Jorge.-

Oct. 2, 2004

Re: [Dillo-dev]0.8.3rc1 test

by Jorge Arellano Cid

On Fri, Oct 01, 2004 at 07:33:32PM +0200, Florent BERANGER wrote: > Hi, Hi. > > I'm Florent BERANGER, from Flonix company > (http://www.flonix.com) > > We like Dillo because it's small and fast :) > I've just tested the 0.8.3rc1, the rendering is really better :) Good you like it. ;) > > Here 2 remaining bugs I've found : > - some hardcoded accented characters (for exemple use "é" in > the place of "é" in HTML code) aren't properly displayed. > - the pages background must be white by default (actually grey). (Answered in a previous email) > We currently working to port the "crystal" icon set to Dillo. > Since done, you'll can add it in icons page :) OK. > We are very interested by Tabs support feature. > I saw some patchs for this feature. How can we compile Dillo > 0.8.3 with them ? A good way to get TABS is to have a window manager that does it. I'm using fluxbox since a long time and TABS work great here for me. Even better than with native support because fluxbox allows me to set the tab labels on the information bar at the top of the window, allowing me to maximize the viewable space. Notwithstanding that I believe the WM to be the right place to implement TABS, maybe in the future we'll have to add some native support for them. Unless the most popular WMs or freedesktop project promote them as a standard, users will lack of TABS in Dillo and asking them to change the WM is too much. Now, once again, with limited manpower all these tasks become a procrastinated forever in the priority list. :( > > Thanks for your good works, Did you plan at some point to include Dillo in Flonix? -- Cheers Jorge.-

Oct. 2, 2004

Re: [Dillo-dev]0.8.3rc1 test

by Jorge Arellano Cid

On Fri, Oct 01, 2004 at 04:28:07PM -0700, Kelson Vibber wrote: > higuita wrote: > >On Fri, 1 Oct 2004 19:33:32 +0200, "Florent BERANGER" > ><cosmic.flo(a)libertysurf.fr> wrote: > >>Here 2 remaining bugs I've found : > >>- some hardcoded accented characters (for exemple use "é" in > >>the place of "é" in HTML code) aren't properly displayed. > >> > > i think that its the dev team stand point to not support bad > > html, as this breaks even more the standard and makes the > > browsers bigger without reason... > > What's bad about that code? HTML 4 is, in the absence of headers or > meta tags indicating otherwise, assumed to be ISO-8859-1. The é is > perfectly legal there. If you give Dillo an ISO-8859-1 'é', it will work fine. Now, if you give it in some other encoding, it will fail. > > Given that they show up correctly when I look at that site here, I > suspect it's a font problem on the display, not an HTML or Dillo problem. Yes, it could be. BTW, if the source page has an ISO-8859-1 'é' this is surely the problem. > As for the default background: that's up to the browser, not the spec. A > basic web design rule going back to 1996 is: never assume what the > default colors are going to be -- if you want something specific, set it > in your code. Unfortunately, Dillo doesn't support CSS yet, which means > you have to resort to setting these in the <body> tag. The background color can be defined in dillorc (either the global one, or on a per user basis). HTH. -- Cheers Jorge.-

Oct. 1, 2004

Re: [Dillo-dev](no subject)

by Jorge Arellano Cid

On Fri, Oct 01, 2004 at 11:12:29PM +0200, rootag(a)terra.es wrote: > Hello, > I had an incompatible library, for this reason I need remove bugs: > 596, > 597, > 599, > 600, > I tried remove it in the page but don't run. Anyone can remove this bugs? Thanks for the reports. Removed. BTW, BUG#598 is because Dillo doesn't support floating objects yet. -- Cheers Jorge.-

Oct. 1, 2004

Re: [Dillo-dev]0.8.3rc1 test

by Kelson Vibber

higuita wrote: > On Fri, 1 Oct 2004 19:33:32 +0200, "Florent BERANGER" <cosmic.flo(a)libertysurf.fr> wrote: >>Here 2 remaining bugs I've found : >>- some hardcoded accented characters (for exemple use "é" in >>the place of "é" in HTML code) aren't properly displayed. > > > i think that its the dev team stand point to not support bad > html, as this breaks even more the standard and makes the > browsers bigger without reason... What's bad about that code? HTML 4 is, in the absence of headers or meta tags indicating otherwise, assumed to be ISO-8859-1. The é is perfectly legal there. Given that they show up correctly when I look at that site here, I suspect it's a font problem on the display, not an HTML or Dillo problem. As for the default background: that's up to the browser, not the spec. A basic web design rule going back to 1996 is: never assume what the default colors are going to be -- if you want something specific, set it in your code. Unfortunately, Dillo doesn't support CSS yet, which means you have to resort to setting these in the <body> tag. -- Kelson Vibber www.hyperborea.org

Oct. 1, 2004

Re: [Dillo-dev]0.8.3rc1 test

by higuita

On Fri, 1 Oct 2004 19:33:32 +0200, "Florent BERANGER" <cosmic.flo(a)libertysurf.fr> wrote: > Here 2 remaining bugs I've found : > - some hardcoded accented characters (for exemple use "é" in > the place of "é" in HTML code) aren't properly displayed. i think that its the dev team stand point to not support bad html, as this breaks even more the standard and makes the browsers bigger without reason... but i'm just one user, the dev team will say something for sure but i write because you can workaround that... install the privoxy filtering proxy and configure it to swap those bad characters with the correct html codes... that is, use privoxy to fix bad html pages, it works well, unless the connection is encripted > We are very interested by Tabs support feature. > I saw some patchs for this feature. How can we compile Dillo > 0.8.3 with them ? the tab patch was too big and needed lot of work to break it in small pieces so it could get into dillo... the patch creator stopped updating its patch so now the patchs fails in alot of places when trying to use it in 0.8.3 there is someone that is trying to update it, but like i said, the patchs is really big cya higuita -- Naturally the common people don't want war... but after all it is the leaders of a country who determine the policy, and it is always a simple matter to drag the people along, whether it is a democracy, or a fascist dictatorship, or a parliament, or a communist dictatorship. Voice or no voice, the people can always be brought to the bidding of the leaders. That is easy. All you have to do is tell them they are being attacked, and denounce the pacifists for lack of patriotism and exposing the country to danger. It works the same in every country. -- Hermann Goering, Nazi and war criminal, 1883-1946

Oct. 1, 2004

[Dillo-dev](no subject)

by rootag＠terra.es

Hello, I had an incompatible library, for this reason I need remove bugs: 596, 597, 599, 600, I tried remove it in the page but don't run. Anyone can remove this bugs?

Oct. 1, 2004

[Dillo-dev]0.8.3rc1 test

by Florent BERANGER

Hi, I'm Florent BERANGER, from Flonix company (http://www.flonix.com) We like Dillo because it's small and fast :) I've just tested the 0.8.3rc1, the rendering is really better :) Here 2 remaining bugs I've found : - some hardcoded accented characters (for exemple use "é" in the place of "é" in HTML code) aren't properly displayed. - the pages background must be white by default (actually grey). We currently working to port the "crystal" icon set to Dillo. Since done, you'll can add it in icons page :) We are very interested by Tabs support feature. I saw some patchs for this feature. How can we compile Dillo 0.8.3 with them ? Thanks for your good works, Best regards, Florent http://www.flonix.com ************************ ADSL ILLIMITE TISCALI + TELEPHONE GRATUIT ************************ Surfez 40 fois plus vite pour 30EUR/mois seulement ! Et téléphonez partout en France gratuitement, vers les postes fixes (hors numéros spéciaux). Tarifs très avantageux vers les mobiles et l'international ! Pour profiter de cette offre exceptionnelle, cliquez ici : http://register.tiscali.fr/adsl (voir conditions sur le site)

Oct. 1, 2004

Re: [Dillo-dev]Dillo on Cygwin

by Jorge Arellano Cid

On Tue, Sep 21, 2004 at 11:32:43AM -0700, Kelson Vibber wrote: > A few days ago I noticed that GTK+ and GLib are now available through > the Cygwin installer, so I started reworking my directions for building > Dillo on Cygwin (currently at > http://www.hyperborea.org/software/dillo/cygwin.html ). I hadn't tried > it since Dillo 0.7.3, so when I built 0.8.2 with the new built-in > libraries, I discovered some problems with DPI. > > First, since Cygwin knows it's running under Windows, it appends .exe to > any executable files it builds. (When you run a command in Cygwin, it > will look for the file as written, then look for it with .exe added -- > so the compiler is really gcc.exe but you can still run it by typing > "gcc".) This includes the DPI plugins, which end up being > bookmarks.dpi.exe, downloads.dpi.exe, etc. This means dpid will not > find them when it looks for available plugins unless you rename them. > > Second, even after renaming them, I can't get any DPI plugins to > actually work. Here is the output from clicking on the bookmarks button: > > Nav_open_url: Url=>dpi:/bm/< > url_str = dpi:/bm/ > Dpi_check_uds: Connection refused /tmp/kvibber-5re3lm/dpid.srs > debug_msg - init_sockdir: The socket directory /tmp/kvibber-5re3lm > exists and is OK > dpid started > Capi_dpi_conn_timeout:: try 1 > main.c:153: get_command: dpid tag is NULL > : main.c:326: main: get_command failed > : bookmarks.dpi (v.13): accepting connections... > [accept]: Software caused connection abort This error message from accept() is weird. > > File permissions on the sockets look the same as they do on a Linux box > (srw-------) > > It may just be something about the way Cygwin handles sockets. Most probably as this scheme works for the *BSDs, GNU/Linux, etc. > Unfortunately I have very little experience with sockets on *any* > platform (and very little experience with C, for that matter) > > Anyway, I figure Cygwin support probably isn't a priority, but I thought > I'd point out the issues in case anyone was interested in tracking them > down. I don't have Cygwin (nor time) here to test it, but surely the Cygwin developers have a clue of what's happening. They probably would appreciate the report too, as they aim to provide the most POSIX like environment I guess. The best way is to test it by stages. The implementation of dpid is not that complex: there's an external daemon process 'dpid' which listens on named sockets and that launches the real dpis when there's activity in the respective named socket. When this happens, the new dpi process "takes over" the socket until it finishes. The details are in doc/Dpid.txt under the sections: "How does it work?" and "dpi service process diagram". Hope this helps. -- Cheers Jorge.-

Sept. 30, 2004