- Dillo-dev - lists.mailman3.com

[patch]: enable the "--enable-ssl" configure flag

by onepoint＠starurchin.org

D'oh!. Here's the patch!

Oct. 16, 2008

[patch]: enable the "--enable-ssl" configure flag

by onepoint＠starurchin.org

It seems a little silly to implement a configure flag in configure.in to define ENABLE_SSL , only for dpi/https.c to ignore it! This patch: * changes --disable-ssl to --enable-ssl * changes dpi/https.c to respect --enable-ssl The default is to disable SSL. The configure script will issue a stern warning about alpha code if --enable-ssl is used. The DPI warning when SSL is disabled has been changed to explain about --enable-ssl . Regards, Jeremy Henty

Oct. 16, 2008

[patch] fltkviewbase cleanup

by Johannes.Hofmann＠gmx.de

Hi, attached patch garbage collects some unused code and renames one of the drawRectangle() methods to just draw(). Both drawRectangle() methods were doing completly different things. Cheers, Johannes

Oct. 16, 2008

cross-site request forgery

by corvid＠lavabit.com

I wrote some code because I said I was going to write some code, but it turns out that I dislike cookies and advertisements too much to give it any decent testing. If anybody ever wants to play with it, though, here it is: http://www.dillo.org/test/csrf-20081015.patch requests not made by user: 1) Cache_redirect() calling a_Nav_push with the URL from the Location HTTP header, which calls Nav_open_url, which can stuff the requesting URL into the Web structure. 2) a_Html_add_new_image() for automatically-loaded images, calling Html_load_image(), which can stuff the requesting URL into the Web structure. Http_send_query() gets the requesting URL from the Web structure and passes it to a_Http_make_query_str(), which calls a_Cookies_get_query. If cookiesrc shows that the requesting URL is not trusted with cookies, I decided to deny it. Otherwise, the URL is sent off with the request to the cookies dpi. In the dpi, the arguments find their way to Cookies_get and then Cookies_requester_authorized(), which tries to decide whether it wants to send a particular cookie on behalf of the requester.

Oct. 15, 2008

[patch]: Ignore the "HttpOnly" cookie attribute

by jcid＠dillo.org

On Wed, Oct 15, 2008 at 12:17:26PM +0100, Jeremy Henty wrote: > > I submitted this before, but it vanished without comment. The > Wikipedia login sets a cookie with an "HttpOnly" attribute, which > Dillo rejects. This means that vanilla Dillo cannot login to > Wikipedia. This patch fixes this by silently ignoring the attribute. Committed. -- Cheers Jorge.-

Oct. 15, 2008

dillo-2.0 is released!

by corvid＠lavabit.com

Diego wrote: > http://es.wikipedia.org/wiki/Dillo is dis-updated?? It could be worse -- in the case of the English article, I did update it some, but it got reverted (presumably because I mostly ripped out old stuff and left the task of blathering on about current directions and such to whoever came by next. I wasn't feeling like stretching "oh, CSS, probably" to fill up much space.) Whatever. So much for wikipedia. Better to write patches. Somewhat related: I trimmed some things out of our FAQ yesterday now that dillo1 is history, and it definitely has room for more questions if anyone thinks something belongs there.

Oct. 15, 2008

[rjn@moxtrait.com: 2.0]

by jcid＠dillo.org

Enjoy! -- Cheers Jorge.-

Oct. 15, 2008

[patch]: Ignore the "HttpOnly" cookie attribute

by onepoint＠starurchin.org

I submitted this before, but it vanished without comment. The Wikipedia login sets a cookie with an "HttpOnly" attribute, which Dillo rejects. This means that vanilla Dillo cannot login to Wikipedia. This patch fixes this by silently ignoring the attribute. See the dillo-dev discussion at http://lists.auriga.wearlab.de/pipermail/dillo-dev/2008-October/005042.html Regards, Jeremy Henty

Oct. 15, 2008

[patch]: change the default tab label to "Dillo"

by onepoint＠starurchin.org

Just a nitpick, but I think it looks a little sad when a new tab pops up with the pointless label "Label", so I changed it to "Dillo". Regards, Jeremy Henty

Oct. 15, 2008

dillo-2.0 is released!

by darkspirit5000＠gmail.com

2008/10/14 Kelson Vibber <kelson(a)pobox.com> > Jorge Arellano Cid wrote: > >> The freshmeat submission was made (waiting for review), the >> web site is updated (AFAIS), we have some packages too. >> > > Woo-hoo! :D > > I've submitted stories to Slashdot, Digg, and Reddit. The Slashdot story > seems to be getting some attention on the Firehose. > > http://digg.com/linux_unix/Dillo_Web_Browser_Updated_At_Last > > http://www.reddit.com/r/programming/comments/77353/dillo_web_browser_update… > http://slashdot.org/firehose.pl?op=view&id=1300555 Submitted to http://barrapunto.com (waiting to see it). http://es.wikipedia.org/wiki/Dillo is dis-updated?? Diego.

Oct. 15, 2008

Donations www-page

by fgeek＠hack.fi

Someone should update http://www.dillo.org/donations.html - Henri

Oct. 15, 2008

dillo-2.0 is released!

by Johannes.Hofmann＠gmx.de

On Tue, Oct 14, 2008 at 05:36:44PM -0300, Jorge Arellano Cid wrote: > On Tue, Oct 14, 2008 at 08:30:22PM +0200, Johannes Hofmann wrote: > > On Tue, Oct 14, 2008 at 12:10:55PM -0300, Jorge Arellano Cid wrote: > > > Hi, > > > > > > The freshmeat submission was made (waiting for review), the > > > web site is updated (AFAIS), we have some packages too. > > > > What about an entry in the fltk applications wiki on www.fltk.org? > > Good. > > Can you make it? Done (waiting for review). Cheers, Johannes

Oct. 14, 2008

dillo-2.0 is released!

by jcid＠dillo.org

On Tue, Oct 14, 2008 at 08:30:22PM +0200, Johannes Hofmann wrote: > On Tue, Oct 14, 2008 at 12:10:55PM -0300, Jorge Arellano Cid wrote: > > Hi, > > > > The freshmeat submission was made (waiting for review), the > > web site is updated (AFAIS), we have some packages too. > > What about an entry in the fltk applications wiki on www.fltk.org? Good. Can you make it? -- Cheers Jorge.-

Oct. 14, 2008

dillo-2.0 is released!

by kelson＠pobox.com

Jorge Arellano Cid wrote: > The freshmeat submission was made (waiting for review), the > web site is updated (AFAIS), we have some packages too. Woo-hoo! :D I've submitted stories to Slashdot, Digg, and Reddit. The Slashdot story seems to be getting some attention on the Firehose. http://digg.com/linux_unix/Dillo_Web_Browser_Updated_At_Last http://www.reddit.com/r/programming/comments/77353/dillo_web_browser_update… http://slashdot.org/firehose.pl?op=view&id=1300555 -- Kelson Vibber Hyperborea.org - SpeedForce.org - AlternativeBrowserAlliance.com

Oct. 14, 2008

dillo-2.0 is released!

by Johannes.Hofmann＠gmx.de

On Tue, Oct 14, 2008 at 12:10:55PM -0300, Jorge Arellano Cid wrote: > Hi, > > The freshmeat submission was made (waiting for review), the > web site is updated (AFAIS), we have some packages too. What about an entry in the fltk applications wiki on www.fltk.org? Cheers, Johannes

Oct. 14, 2008

patch: remove image-middle-click-loads-all

by jcid＠dillo.org

On Tue, Oct 14, 2008 at 05:32:07PM +0000, corvid wrote: > Yes, hurray for everything :) ! > I doubt anyone uses this to load page images. > I only press it by accident once in a while. > Committed. -- Cheers Jorge.-

Oct. 14, 2008

patch: remove image-middle-click-loads-all

by corvid＠lavabit.com

Yes, hurray for everything :) I doubt anyone uses this to load page images. I only press it by accident once in a while.

Oct. 14, 2008

dillo-2.0 is released!

by fgeek＠hack.fi

On Tue, 14 Oct 2008 12:10:55 -0300 Jorge Arellano Cid <jcid(a)dillo.org> wrote: > Hi, > > The freshmeat submission was made (waiting for review), the > web site is updated (AFAIS), we have some packages too. > > Great! > > I want to thank all the people that made it possible. > > > Wheeeeeeee! > Cangratulations! :) -- Henri Salo <fgeek at hack.fi> +358407705733 GPG ID: 2EA46E4F fp: 14D0 7803 BFF6 EFA0 9998 8C4B 5DFE A106 2EA4 6E4F

Oct. 14, 2008

dillo-2.0 is released!

by Johannes.Hofmann＠gmx.de

On Tue, Oct 14, 2008 at 12:10:55PM -0300, Jorge Arellano Cid wrote: > Hi, > > The freshmeat submission was made (waiting for review), the > web site is updated (AFAIS), we have some packages too. > > Great! > > I want to thank all the people that made it possible. > > > Wheeeeeeee! > Hooray!

Oct. 14, 2008

dillo-2.0 is released!

by onepoint＠starurchin.org

On Tue, Oct 14, 2008 at 12:10:55PM -0300, Jorge Arellano Cid wrote: > I want to thank all the people that made it possible. Me too. Dillo-2 is a great achievement and its team (and its team manager) are a great team. I've been very glad to take part and I hope everyone else feels the same way. Jeremy Henty

Oct. 14, 2008

dillo-2.0 is released!

by jcid＠dillo.org

Hi, The freshmeat submission was made (waiting for review), the web site is updated (AFAIS), we have some packages too. Great! I want to thank all the people that made it possible. Wheeeeeeee! -- Cheers Jorge.-

Oct. 14, 2008

patch: fix numeric ipv6 address handling

by jcid＠dillo.org

On Tue, Oct 14, 2008 at 04:40:27PM +0200, Justus Winter wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Jorge Arellano Cid wrote: > >> 2. joining hostname and port for the Host: header > >> > >> a_Http_make_query_str used to build the Host: header by concatenating > >> the url->hostname with the port number in case it's not the default > >> port. Unfortunately this breaks with numeric ipv6 addresses since an > >> ipv6 literal has to be enclosed with square brackets when used in urls > >> and Host: header fields (according to my experiments and my > >> interpretation of rfc 2616 section 14.23 ("Host header") and 3.2.2 > >> ("http URL")). > >> > >> I would therefore suggest to use the original authority string obtained > >> from the url. Patch is attached. > > > > AFAIR, the "remove default port" part was added because some sites > > answer 404 when receiving it. > Yes, that's what your comment says. But my proposal does not break those > cases in general. > > The current behavior is to construct the authority string for the host > header using the parsed hostname and the port number in case it differs > from 80. > > I proposed to use the authority part from the url the user entered. This > may break some sites _if_ the user explicitly entered ":80" in the url. That's the point. > Btw, I just checked firefox and it seems to do it this way. > > Entering "http://localhost:80" in firefox url bar yields the following > request: > > root@thinkbox:/var/tmp# nc -l -p 80 > GET / HTTP/1.1 > Host: localhost:80 > [...] Good. If FF does it currently, then it should be less of a problem. I'll commit as it is and let's see what happens. -- Cheers Jorge.-

Oct. 14, 2008

patch: fix numeric ipv6 address handling

by 4winter＠informatik.uni-hamburg.de

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Jorge Arellano Cid wrote: >> 2. joining hostname and port for the Host: header >> >> a_Http_make_query_str used to build the Host: header by concatenating >> the url->hostname with the port number in case it's not the default >> port. Unfortunately this breaks with numeric ipv6 addresses since an >> ipv6 literal has to be enclosed with square brackets when used in urls >> and Host: header fields (according to my experiments and my >> interpretation of rfc 2616 section 14.23 ("Host header") and 3.2.2 >> ("http URL")). >> >> I would therefore suggest to use the original authority string obtained >> from the url. Patch is attached. > > AFAIR, the "remove default port" part was added because some sites > answer 404 when receiving it. Yes, that's what your comment says. But my proposal does not break those cases in general. The current behavior is to construct the authority string for the host header using the parsed hostname and the port number in case it differs from 80. I proposed to use the authority part from the url the user entered. This may break some sites _if_ the user explicitly entered ":80" in the url. Btw, I just checked firefox and it seems to do it this way. Entering "http://localhost:80" in firefox url bar yields the following request: root@thinkbox:/var/tmp# nc -l -p 80 GET / HTTP/1.1 Host: localhost:80 [...] > I'd prefer a patch that handles IPv6 like you suggest, and IPv4 as now. While this is certainly doable, I can't see anything gained by assembling the authority string besides stripping the default port in case the user explicitly entered it. Justus -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkj0r1oACgkQoPmwNWhsaZYwrACdGJanS3yoXudr/Dir/3q9jR2U w2wAoJJiFj6E5zGmO+ILNTcLzhxIfyJR =uX/b -----END PGP SIGNATURE-----

Oct. 14, 2008

patch: fix numeric ipv6 address handling

by jcid＠dillo.org

On Sun, Oct 05, 2008 at 01:49:29AM +0200, Justus Winter wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Hey all :) > > there were two problems preventing the use of numeric ipv6 literals: > > 1. splitting the authority in hostname and port > > the attached patch fixes the handling of numeric ipv6 addresses entered > into the url bar. a_Url_hostname got the colons delimiting the 8 groups > of hexadecimal digits of an ipv6 address confused with a port number. Committed to CVS (not in the already made release). > 2. joining hostname and port for the Host: header > > a_Http_make_query_str used to build the Host: header by concatenating > the url->hostname with the port number in case it's not the default > port. Unfortunately this breaks with numeric ipv6 addresses since an > ipv6 literal has to be enclosed with square brackets when used in urls > and Host: header fields (according to my experiments and my > interpretation of rfc 2616 section 14.23 ("Host header") and 3.2.2 > ("http URL")). > > I would therefore suggest to use the original authority string obtained > from the url. Patch is attached. AFAIR, the "remove default port" part was added because some sites answer 404 when receiving it. I'd prefer a patch that handles IPv6 like you suggest, and IPv4 as now. -- Cheers Jorge.-

Oct. 14, 2008

Dillo Packages

by akemnade＠tzi.de

On Tue, 14 Oct 2008 09:56:40 -0300 Jorge Arellano Cid <jcid(a)dillo.org> wrote: > Hi Andreas, > > On Mon, Oct 13, 2008 at 11:09:25PM +0200, Andreas Kemnade wrote: > > Hi, > > > > On Mon, 13 Oct 2008 15:27:19 -0300 > > Jorge Arellano Cid <jcid(a)dillo.org> wrote: > > > > > Hi, > > > > > > I'd like to have packages for Debian and some RPMs. > > > Kelson already committed himself with the RPMs, and although > > > I can make i32 and i64 Debian packages, I'm a newbie on it. > > > > > I played a bit with the deb stuff and created a .deb for debian testing. > > I have put a .deb file in http://misc.andi.de1.cc/dillo > > I installed the correct fltk snapshot before using the usual ./configure && make && make install, > > and built the .deb after that. This handling of the fltk dependancy is > > not very clean... I guess the cleaner one would be to have > > fltk2 packages, but as fltk2 is linked statically against dillo, > > at least the .deb package should work. > > How do I get an etch .deb from your work? I have uploaded one to the place where the other .deb is. If you want to do it yourself: Unpack the sources, apply the debian patch cd into that directory, dpkg-buildpackage -rfakeroot you have to install the packages debhelper, debmake, devscripts and fakeroot (I hope I don't miss one) Greetings Andreas Kemnade

Oct. 14, 2008

Dillo Packages

by jcid＠dillo.org

On Mon, Oct 13, 2008 at 02:22:49PM -0700, Kelson Vibber wrote: > Jorge Arellano Cid wrote: >> I'd like to have packages for Debian and some RPMs. >> Kelson already committed himself with the RPMs, and although >> I can make i32 and i64 Debian packages, I'm a newbie on it. > > Here's the status on the RPMs. I've built the following: > > Fedora 9, x86_64 > RHEL 3, x86 > RHEL 4, x86 and x86_64 > RHEL 5, x86 and x86_64 Excellent. Thanks a lot. -- Cheers Jorge.-

Oct. 14, 2008

Dillo Packages

by jcid＠dillo.org

Hi Andreas, On Mon, Oct 13, 2008 at 11:09:25PM +0200, Andreas Kemnade wrote: > Hi, > > On Mon, 13 Oct 2008 15:27:19 -0300 > Jorge Arellano Cid <jcid(a)dillo.org> wrote: > > > Hi, > > > > I'd like to have packages for Debian and some RPMs. > > Kelson already committed himself with the RPMs, and although > > I can make i32 and i64 Debian packages, I'm a newbie on it. > > > I played a bit with the deb stuff and created a .deb for debian testing. > I have put a .deb file in http://misc.andi.de1.cc/dillo > I installed the correct fltk snapshot before using the usual ./configure && make && make install, > and built the .deb after that. This handling of the fltk dependancy is > not very clean... I guess the cleaner one would be to have > fltk2 packages, but as fltk2 is linked statically against dillo, > at least the .deb package should work. How do I get an etch .deb from your work? -- Cheers Jorge.-

Oct. 14, 2008

*bump*

by jcid＠dillo.org

On Tue, Oct 14, 2008 at 01:49:22AM +0200, Justus Winter wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Justus Winter wrote: > > Hey all :) > > > > there were two problems preventing the use of numeric ipv6 literals: > > > > 1. splitting the authority in hostname and port > > [...] > > 2. joining hostname and port for the Host: header > Any comments? Any chances of getting those two in for the release? I do > consider ipv6 support somewhat important these days... ;) It will go after the release. Changing the query string in this moment gives me shudders! :-o -- Cheers Jorge.-

Oct. 14, 2008

aclocal.m4 and autom4te.cache in CVS

by onepoint＠starurchin.org

On Mon, Oct 13, 2008 at 04:19:59PM +0200, Johannes Hofmann wrote: > On Mon, Oct 13, 2008 at 02:06:32PM +0100, Jeremy Henty wrote: > > BTW, is anyone else using git like this? I thought I saw some > > patches on the list that looked git-generated. > > I'm using mercurial, which uses a similar patch format for exporting > patches. Ah, it was probably your patches I was thinking of. Jeremy

Oct. 14, 2008

*bump*

by 4winter＠informatik.uni-hamburg.de

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Justus Winter wrote: > Hey all :) > > there were two problems preventing the use of numeric ipv6 literals: > > 1. splitting the authority in hostname and port > [...] > 2. joining hostname and port for the Host: header Any comments? Any chances of getting those two in for the release? I do consider ipv6 support somewhat important these days... ;) Justus -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFI896CoPmwNWhsaZYRAj/cAKCSkIWOj2HpIMnaJPfLGEczSe+SpQCeNINO 1pqXTIFLO41SZi/4XLBCWsA= =BbKT -----END PGP SIGNATURE-----

Oct. 14, 2008

Dillo Packages

by kelson＠pobox.com

Jorge Arellano Cid wrote: > I'd like to have packages for Debian and some RPMs. > Kelson already committed himself with the RPMs, and although > I can make i32 and i64 Debian packages, I'm a newbie on it. Here's the status on the RPMs. I've built the following: Fedora 9, x86_64 RHEL 3, x86 RHEL 4, x86 and x86_64 RHEL 5, x86 and x86_64 Unfortunately I don't have access to any 32-bit Fedora boxes right now, but I should be able to build it with mock if I can figure out how to convince it to install the fltk package I've built. My old build system is in storage (way in the back, under a bunch of boxes, so not easy to get at), so it will be a while before I can build/test on any other distributions. If anyone here has a 32-bit Fedora 9 box and would like to contribute an RPM, or another RPM-based distribution (SuSE, Mandriva, etc.), here are the source RPMs I'm using. http://www.hyperborea.org/software/dillo/rpms/fltk2-2.0.x.r6403-0.1.src.rpm http://www.hyperborea.org/software/dillo/rpms/dillo-2.0-1.src.rpm Directions for building the Dillo RPM: 1. Ideally, set up an RPM build tree in your home directory so you can build as yourself and not as root. (I think directions are up at rpm.org) 2. Build the fltk RPM: rpmbuild --rebuild fltk2-2.0.x.r6403-0.1.src.rpm 3. Install the fltk RPM (as root, or sudo) rpm -ivh /path/to/rpm/file 4. Build the dillo RPM: rpmbuild --rebuild dillo-2.0-1.src.rpm 5. Install the dillo RPM: rpm -ivh /path/to/rpm/file Possible issues with these SRPMS; - I don't have all the build dependencies set properly. - Distro-specific code on the Dillo SRPM may be out of date. - If you already have FLTK2 installed, you can skip the FLTK RPM and add --nodeps to the commandline in step 4. -- Kelson Vibber Hyperborea.org - SpeedForce.org - AlternativeBrowserAlliance.com

Oct. 13, 2008

Dillo Packages

by jcid＠dillo.org

On Mon, Oct 13, 2008 at 11:09:25PM +0200, Andreas Kemnade wrote: > Hi, > > On Mon, 13 Oct 2008 15:27:19 -0300 > Jorge Arellano Cid <jcid(a)dillo.org> wrote: > > > Hi, > > > > I'd like to have packages for Debian and some RPMs. > > Kelson already committed himself with the RPMs, and although > > I can make i32 and i64 Debian packages, I'm a newbie on it. > > > I played a bit with the deb stuff and created a .deb for debian testing. > I have put a .deb file in http://misc.andi.de1.cc/dillo > I installed the correct fltk snapshot before using the usual ./configure && make && make install, > and built the .deb after that. This handling of the fltk dependancy is > not very clean... I guess the cleaner one would be to have > fltk2 packages, but as fltk2 is linked statically against dillo, > at least the .deb package should work. > I basically took the old .diff.gz for dillo 0.8.6 from deb-lenny and deleted the i18n stuff. Great Andreas. I added the link to that page. Would you mind packing it with "make install-strip"? -- Cheers Jorge.-

Oct. 13, 2008

Dillo Packages

by akemnade＠tzi.de

Hi, On Mon, 13 Oct 2008 15:27:19 -0300 Jorge Arellano Cid <jcid(a)dillo.org> wrote: > Hi, > > I'd like to have packages for Debian and some RPMs. > Kelson already committed himself with the RPMs, and although > I can make i32 and i64 Debian packages, I'm a newbie on it. > I played a bit with the deb stuff and created a .deb for debian testing. I have put a .deb file in http://misc.andi.de1.cc/dillo I installed the correct fltk snapshot before using the usual ./configure && make && make install, and built the .deb after that. This handling of the fltk dependancy is not very clean... I guess the cleaner one would be to have fltk2 packages, but as fltk2 is linked statically against dillo, at least the .deb package should work. I basically took the old .diff.gz for dillo 0.8.6 from deb-lenny and deleted the i18n stuff. Greetings Andreas Kemnade

Oct. 13, 2008

Altavista search does not work

by francis＠daoine.org

On Mon, Oct 13, 2008 at 05:06:23PM +0000, corvid wrote: > Francis wrote: > > On Fri, Oct 10, 2008 at 12:43:59PM +0000, corvid wrote: > > > Christopher wrote: > > > > Doing web searches using www.altavista.com does not work. It is as > > > > if the search terms are not submitted. > > > > Searching for "Images", "MP3/Audio", or "Video" instead of "Web" > > > > or "News" works well. > > It's a problem with altavista, in that their HTML is invalid, and > > therefore they rely on browser-fixups to get the result they probably > > wanted. > > > > It's a problem with dillo, in that the submit button (presumably) works > > for most other browsers. > > > > More specifically, the main difference between the (working) images page > > and the (failing) web page is the presence of an explicit close-table-cell > > </td> tag. > > > > If you remove the </td> on line 68 of the altavista index.html page, > > then the submit button works. > > > > So presumably dillo is handling "explicit </td> before <td>" differently > > from "no </td> immediately before <td>", at least in this case. > > I think dillo sees <td> ... <form> ... </td> > and closes the form, which seems like the right behavior. Yes; it is "a" right behaviour. Because the markup is incorrect, pretty much anything dillo does is right. But what dillo does, is it sees <table><tr><td><form><input></td><td> and (correctly) closes the form when it sees "</td>". And on the "images" search page, dillo sees the (I think) exactly equivalent <table><tr><td><form><input><tr> and (also correctly) does not close the form when it sees "<tr>" which implies "</td>" and "</tr>". That was the difference I was trying to point out. The html is broken in each case. Dillo's handling of it is valid in each case. But the second example results in "friendlier" response to the user (in this case. Under other circumstances, showing the brokenness may be "friendlier"). f -- Francis Daly francis(a)daoine.org

Oct. 13, 2008

Dillo Packages

by jcid＠dillo.org

Hi, I'd like to have packages for Debian and some RPMs. Kelson already committed himself with the RPMs, and although I can make i32 and i64 Debian packages, I'm a newbie on it. Volunteers? -- Cheers Jorge.-

Oct. 13, 2008

Altavista search does not work

by jcid＠dillo.org

On Mon, Oct 13, 2008 at 05:41:20PM +0100, Francis Daly wrote: > On Fri, Oct 10, 2008 at 12:43:59PM +0000, corvid wrote: > > Christopher wrote: > > Hi there, > > > > Doing web searches using www.altavista.com does not work. It is as > > > if the search terms are not submitted. The following HTML errors > > > are detected: > > > > > > HTML warning: line 2, the required DOCTYPE declaration is missing (or invalid) > > > HTML warning: line 68, - forcing close of open tag: <form> > > > HTML warning: line 68, FORM lacks a Submit button > > > HTML warning: line 94, unexpected closing tag: </form>. -- expected </table> > > > > > > Searching for "Images", "MP3/Audio", or "Video" instead of "Web" > > > or "News" works well. > > > > > > Is this a problem with altavista rather than with dillo? > > > > It looks like altavista just threw together their form and table tags > > rather than nesting them. > > It's a problem with altavista, in that their HTML is invalid, and > therefore they rely on browser-fixups to get the result they probably > wanted. Yes. Both validators WDG and W3C (bug meter's right-click menu), explain this bug. In simple terms, the form declaration should be outside the table. Otherwise, following HTML rules, the form is closed. -- Cheers Jorge.-

Oct. 13, 2008

Altavista search does not work

by corvid＠lavabit.com

Francis wrote: > On Fri, Oct 10, 2008 at 12:43:59PM +0000, corvid wrote: > > Christopher wrote: > > Hi there, > > > > Doing web searches using www.altavista.com does not work. It is as > > > if the search terms are not submitted. The following HTML errors > > > are detected: > > > > > > HTML warning: line 2, the required DOCTYPE declaration is missing (or invalid) > > > HTML warning: line 68, - forcing close of open tag: <form> > > > HTML warning: line 68, FORM lacks a Submit button > > > HTML warning: line 94, unexpected closing tag: </form>. -- expected </table> > > > > > > Searching for "Images", "MP3/Audio", or "Video" instead of "Web" > > > or "News" works well. > > > > > > Is this a problem with altavista rather than with dillo? > > > > It looks like altavista just threw together their form and table tags > > rather than nesting them. > > It's a problem with altavista, in that their HTML is invalid, and > therefore they rely on browser-fixups to get the result they probably > wanted. > > It's a problem with dillo, in that the submit button (presumably) works > for most other browsers. > > More specifically, the main difference between the (working) images page > and the (failing) web page is the presence of an explicit close-table-cell > </td> tag. > > If you remove the </td> on line 68 of the altavista index.html page, > then the submit button works. > > So presumably dillo is handling "explicit </td> before <td>" differently > from "no </td> immediately before <td>", at least in this case. I think dillo sees <td> ... <form> ... </td> and closes the form, which seems like the right behavior. > (dillo1 acts the same, except it doesn't show the input field or the > search button at all in the first case.) > > It doesn't look like a significant v1-v2 regression to me, apart from > the fact that it now looks like it might submit something.

Oct. 13, 2008

Altavista search does not work

by francis＠daoine.org

On Fri, Oct 10, 2008 at 12:43:59PM +0000, corvid wrote: > Christopher wrote: Hi there, > > Doing web searches using www.altavista.com does not work. It is as > > if the search terms are not submitted. The following HTML errors > > are detected: > > > > HTML warning: line 2, the required DOCTYPE declaration is missing (or invalid) > > HTML warning: line 68, - forcing close of open tag: <form> > > HTML warning: line 68, FORM lacks a Submit button > > HTML warning: line 94, unexpected closing tag: </form>. -- expected </table> > > > > Searching for "Images", "MP3/Audio", or "Video" instead of "Web" > > or "News" works well. > > > > Is this a problem with altavista rather than with dillo? > > It looks like altavista just threw together their form and table tags > rather than nesting them. It's a problem with altavista, in that their HTML is invalid, and therefore they rely on browser-fixups to get the result they probably wanted. It's a problem with dillo, in that the submit button (presumably) works for most other browsers. More specifically, the main difference between the (working) images page and the (failing) web page is the presence of an explicit close-table-cell </td> tag. If you remove the </td> on line 68 of the altavista index.html page, then the submit button works. So presumably dillo is handling "explicit </td> before <td>" differently from "no </td> immediately before <td>", at least in this case. (dillo1 acts the same, except it doesn't show the input field or the search button at all in the first case.) It doesn't look like a significant v1-v2 regression to me, apart from the fact that it now looks like it might submit something. f -- Francis Daly francis(a)daoine.org

Oct. 13, 2008

aclocal.m4 and autom4te.cache in CVS

by 4winter＠informatik.uni-hamburg.de

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Jeremy Henty wrote: > Yes you're right, they're not in CVS/Entries . I've started managing > my dillo development by tracking the CVS checkout in one git > repository (the "remote") and developing in a cloned repository (the > "local"). So the remote repository should stay pristine, but it > didn't. Looks like I screwed up and ran autogen.sh in the remote at > some point. Apologies for the noise. > > BTW, is anyone else using git like this? I thought I saw some patches > on the list that looked git-generated. I am using bzr with a similar setup. Justus -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkjzaCIACgkQoPmwNWhsaZYENgCfSNnEZutmw0RoUbS7u4W+6r1P 8woAnjIPqNcBtZ5tAIY3jm4wW6mRa17p =LtQk -----END PGP SIGNATURE-----

Oct. 13, 2008

dillo-2.0 rc1 is in CVS

by jcid＠dillo.org

On Mon, Oct 13, 2008 at 06:45:23AM -0700, bf wrote: > The message about re-enabling https support (on line 667) in revision 1.3 of > dpi/https.c should refer to line 72, not line 65. Changed. > Ironically, > > http://cvs.auriga.wearlab.de/cgi-bin/cvsweb.cgi/~checkout~/dillo/dillo2/dpi… > > fails with "Error: Unsupported content type: text/x-c++src". Maybe some > of dillo's type-checking needs to be revisited. Also problematic are some > other kinds of text files. Please post some other URLs. To be reviewed in the release after this one... -- Cheers Jorge.-

Oct. 13, 2008

aclocal.m4 and autom4te.cache in CVS

by Johannes.Hofmann＠gmx.de

On Mon, Oct 13, 2008 at 02:06:32PM +0100, Jeremy Henty wrote: > On Mon, Oct 13, 2008 at 09:17:37AM -0300, Jorge Arellano Cid wrote: > > On Sun, Oct 12, 2008 at 09:43:26AM +0100, Jeremy Henty wrote: > > > > > > Why have aclocal.m4 and autom4te.cache been added to CVS? > > > > Just checked in CVS Web and also a clean CVS checkout, and don't > > see those files added. > > Yes you're right, they're not in CVS/Entries . I've started managing > my dillo development by tracking the CVS checkout in one git > repository (the "remote") and developing in a cloned repository (the > "local"). So the remote repository should stay pristine, but it > didn't. Looks like I screwed up and ran autogen.sh in the remote at > some point. Apologies for the noise. > > BTW, is anyone else using git like this? I thought I saw some patches > on the list that looked git-generated. I'm using mercurial, which uses a similar patch format for exporting patches. Regards, Johannes

Oct. 13, 2008

dillo-2.0 rc1 is in CVS

by bf2006a＠yahoo.com

The message about re-enabling https support (on line 667) in revision 1.3 of dpi/https.c should refer to line 72, not line 65. Ironically, http://cvs.auriga.wearlab.de/cgi-bin/cvsweb.cgi/~checkout~/dillo/dillo2/dpi… fails with "Error: Unsupported content type: text/x-c++src". Maybe some of dillo's type-checking needs to be revisited. Also problematic are some other kinds of text files. Regards, b.

Oct. 13, 2008

dillo-2.0 rc1 is in CVS

by jcid＠dillo.org

Hi, On Sat, Oct 11, 2008 at 02:03:50PM -0700, Kelson Vibber wrote: > Seems to compile and run okay on Fedora 9 x86_64. I uploaded the final dillo-2.0 to the website. It will be released tomorrow. It's rc1 with: * Release date in the splash page, * Directions on where to get the FLTK2 library in the README. It'd be good to have some packages for Debian and Fedora. PS: http://www.dillo.org/download/dillo-2.0.tar.bz2 -- Cheers Jorge.-

Oct. 13, 2008

aclocal.m4 and autom4te.cache in CVS

by onepoint＠starurchin.org

On Mon, Oct 13, 2008 at 09:17:37AM -0300, Jorge Arellano Cid wrote: > On Sun, Oct 12, 2008 at 09:43:26AM +0100, Jeremy Henty wrote: > > > > Why have aclocal.m4 and autom4te.cache been added to CVS? > > Just checked in CVS Web and also a clean CVS checkout, and don't > see those files added. Yes you're right, they're not in CVS/Entries . I've started managing my dillo development by tracking the CVS checkout in one git repository (the "remote") and developing in a cloned repository (the "local"). So the remote repository should stay pristine, but it didn't. Looks like I screwed up and ran autogen.sh in the remote at some point. Apologies for the noise. BTW, is anyone else using git like this? I thought I saw some patches on the list that looked git-generated. Regards, Jeremy Henty

Oct. 13, 2008

aclocal.m4 and autom4te.cache in CVS

by jcid＠dillo.org

On Sun, Oct 12, 2008 at 09:43:26AM +0100, Jeremy Henty wrote: > > Why have aclocal.m4 and autom4te.cache been added to CVS? Just checked in CVS Web and also a clean CVS checkout, and don't see those files added. > Or am I missing something? I hope. :-) -- Cheers Jorge.-

Oct. 13, 2008

dillo-2.0 rc1 is in CVS

by jcid＠dillo.org

On Sat, Oct 11, 2008 at 02:03:50PM -0700, Kelson Vibber wrote: > Seems to compile and run okay on Fedora 9 x86_64. Good. > Just a thought: Should the README mention where to get FLTK2, and perhaps > which snapshot to use? Done. -- Cheers Jorge.-

Oct. 13, 2008

dillo-2.0 rc1 is in CVS

by Johannes.Hofmann＠gmx.de

On Sat, Oct 11, 2008 at 12:58:57PM +0200, Joerg Sonnenberger wrote: > On Fri, Oct 10, 2008 at 07:31:08PM -0400, Jorge Arellano Cid wrote: > > http://www.dillo.org/test/dillo-2.0rc1.tar.bz2 > > Consider to replace the AC_CHECK_SIZEOF() point and d_size.h magic > with AC_TYPE_UINT16_T and friends. THose use either the system macros or > figure out the rest (just include inttypes.h and sys/types.h if > present). > > I don't really get your socklen_t test. I'm not aware of any platform > that lacks socklen_t and where the third argument is not unsigned int. Do you think the current setup causeis problems or can we delay such a cleanup until after the release? I'm a bit worried to break something... Cheers, Johannes

Oct. 12, 2008

aclocal.m4 and autom4te.cache in CVS

by onepoint＠starurchin.org

Why have aclocal.m4 and autom4te.cache been added to CVS? These are generated by the build so version controlling them only adds noise. Or am I missing something? Jeremy Henty

Oct. 12, 2008

dillo-2.0 rc1 is in CVS

by kelson＠pobox.com

Seems to compile and run okay on Fedora 9 x86_64. Just a thought: Should the README mention where to get FLTK2, and perhaps which snapshot to use? -- Kelson Vibber www.hyperborea.org and SpeedForce.org

Oct. 11, 2008

Re: dillo-2.0 rc1 is in CVS

by jcid＠dillo.org

Hi, I have to go out of the city, so most probably I'll review what's left this Monday and release on Tuesday. Thanks for all the feedback. Please keep reviewing rc1, and if the local dpi installation can be sorted out, great. -- Cheers Jorge.-

Oct. 11, 2008