- Dillo-dev - lists.mailman3.com

Request for the adding of a Keyboard Shortcut

by Zero3K＠myopera.com

There should be a keyboard shortcut to refresh the page by pressing F5 added to it.

March 23, 2013

Reporting broken sites

by Zero3K＠myopera.com

I think that an option to do so should be added to it.

March 23, 2013

Dialog box for Cookie management

by Zero3K＠myopera.com

It would be nice if one that allowed the allowing/disallowing of domains was added.

March 23, 2013

Ad Blocker

by Zero3K＠myopera.com

On Sat, 23 Mar 2013 23:51:24 +0400 p37sitdu at lavabit.com wrote: > On Sat, Mar 23, 2013 at 02:35:09PM -0400, Bryan Kirk wrote: > > It would be nice if one that supports using ABP lists was added to Dillo. > > domainrc support will be added to the next version. It allows even > more effective approach than ABP lists: you can use white lists. > First, you deny all external requests and then you only allow requests > that you need, like this: > > # domainrc > default deny > > google.com .gstatic.com > > # Wikimedia > wikidata.org .wikimedia.org > wikibooks.org .wikimedia.org > wikidata.org .wikimedia.org > wikinews.org .wikimedia.org > wikipedia.org .wikimedia.org > wikisource.org .wikimedia.org > wikiversity.org .wikimedia.org > wikivoyage.org .wikimedia.org > wiktionary.org .wikimedia.org > > freecode.com .fsdn.com > slashdot.org .fsdn.com > sourceforge.net .fsdn.com > > wordpress.com .wp.com > > This way lists are much shorter. Only ads that are on the same domain > are missed, but they are rare. > > > _______________________________________________ > Dillo-dev mailing list > Dillo-dev at dillo.org > http://lists.auriga.wearlab.de/cgi-bin/mailman/listinfo/dillo-dev Will the white/empty space be removed when the domains are blocked from loading content?

March 23, 2013

Javascript Support

by Zero3K＠myopera.com

On Sat, 23 Mar 2013 23:45:49 +0400 p37sitdu at lavabit.com wrote: > On Sat, Mar 23, 2013 at 02:34:28PM -0400, Bryan Kirk wrote: > > How much longer till it is available in Dillo? > > From http://www.dillo.org/funding/objectives.html: "Not allowing > browser programming from the outside." > > > _______________________________________________ > Dillo-dev mailing list > Dillo-dev at dillo.org > http://lists.auriga.wearlab.de/cgi-bin/mailman/listinfo/dillo-dev Its in the "Current Plans" though.

March 23, 2013

Dillo is missing three functions found in all other browsers

by p37sitdu＠lavabit.com

On Sat, Mar 23, 2013 at 03:46:59PM -0400, Bryan Kirk wrote: > CTRL+C and CTRL+P doesn't work as I expect them to. You copy text by selecting it and paste text with the middle mouse button.

March 23, 2013

Ad Blocker

by p37sitdu＠lavabit.com

On Sat, Mar 23, 2013 at 02:35:09PM -0400, Bryan Kirk wrote: > It would be nice if one that supports using ABP lists was added to Dillo. domainrc support will be added to the next version. It allows even more effective approach than ABP lists: you can use white lists. First, you deny all external requests and then you only allow requests that you need, like this: # domainrc default deny google.com .gstatic.com # Wikimedia wikidata.org .wikimedia.org wikibooks.org .wikimedia.org wikidata.org .wikimedia.org wikinews.org .wikimedia.org wikipedia.org .wikimedia.org wikisource.org .wikimedia.org wikiversity.org .wikimedia.org wikivoyage.org .wikimedia.org wiktionary.org .wikimedia.org freecode.com .fsdn.com slashdot.org .fsdn.com sourceforge.net .fsdn.com wordpress.com .wp.com This way lists are much shorter. Only ads that are on the same domain are missed, but they are rare.

March 23, 2013

Dillo is missing three functions found in all other browsers

by Zero3K＠myopera.com

On Sat, 23 Mar 2013 23:42:39 +0400 p37sitdu at lavabit.com wrote: > On Sat, Mar 23, 2013 at 02:53:48PM -0400, Bryan Kirk wrote: > > 1. The ability to copy selected text. > > 2. The ability to paste text copied from it/another program. > > These features are not missing, but PRIMARY selection is used instead > of CLIPBOARD. > > > _______________________________________________ > Dillo-dev mailing list > Dillo-dev at dillo.org > http://lists.auriga.wearlab.de/cgi-bin/mailman/listinfo/dillo-dev CTRL+C and CTRL+P doesn't work as I expect them to.

March 23, 2013

Javascript Support

by p37sitdu＠lavabit.com

On Sat, Mar 23, 2013 at 02:34:28PM -0400, Bryan Kirk wrote: > How much longer till it is available in Dillo? >From http://www.dillo.org/funding/objectives.html: "Not allowing browser programming from the outside."

March 23, 2013

Dillo is missing three functions found in all other browsers

by p37sitdu＠lavabit.com

On Sat, Mar 23, 2013 at 02:53:48PM -0400, Bryan Kirk wrote: > 1. The ability to copy selected text. > 2. The ability to paste text copied from it/another program. These features are not missing, but PRIMARY selection is used instead of CLIPBOARD.

March 23, 2013

Dillo is missing three functions found in all other browsers

by Zero3K＠myopera.com

1. The ability to copy selected text. 2. The ability to paste text copied from it/another program. 3. The ability to select all text in the page/address bar.

March 23, 2013

Ad Blocker

by Zero3K＠myopera.com

It would be nice if one that supports using ABP lists was added to Dillo.

March 23, 2013

Javascript Support

by Zero3K＠myopera.com

How much longer till it is available in Dillo?

March 23, 2013

dillo-3.0.3-rc1 available

by Hole.destructor＠gmx.de

Hi. > Please give it a try on as many platforms as possible. I compiled and ran it on Debian Squeeze and Scientific Linux 6 without problems. Even the installation of the hyphenation pattern files works fine on these operating systems. However, there is no remark about the installation of the hyphenation pattern files in README or INSTALL. Since some users might not know this new feature, I guess many will not install the pattern files. So, I recommend to add an example call of `dillo-install-hyphenation' to the "Dillo 3" section in README. Best, Alex

March 21, 2013

dillo-3.0.3-rc1 available

by Johannes.Hofmann＠gmx.de

Hi all, finally there is a first release candidate for the upcoming 3.0.3 release available: http://www.dillo.org/download/dillo-3.0.3-rc1.tar.bz2 Please give it a try on as many platforms as possible. It would be cool if package maintainers could also give their feedback. Cheers, Johannes

March 18, 2013

upcoming 3.0.3 release

by Johannes.Hofmann＠gmx.de

Hi, yesterday I made some changes to the README, ChangeLog, and splash screen in preparation for the upcoming 3.0.3 release. Please have a look at the texts and check for spelling errors or features I missed. I would like to make a first release candidate soon. Cheers, Johannes

March 14, 2013

release preperations

by sgeerken＠dillo.org

> > > > Also we would need to install the install-hyphenation script, so > > > > that it becomes part of binary packages. > > > > > > There would also some changes neccessary, since the script searches > > > "Makefile". > > > > > > This is an aspect I have not yet thought about: whether the script is > > > part of the installation process (as now) or the installation itself > > > (as you suggest). > > > > If we install it it will probabely end up in binary packages and be > > available for end users. > > If we don't install it, we rely on packagers to deal with the > > problem (e.g. to create a separate hyphenation pattern package). > > > > > > If we do that it would have to be renamed to > > > > dillo-install-hyphenation or something like that. > > > > Does this sound ok? > > I'll try to modify the script, so it can be installed to > ${prefix}/bin. In any case, this is certainly useful. Done, plus some polishing. Call dillo-install-hyphenation either from the tarball directory or from /usr/local/bin. Sebastian

March 8, 2013

Missing dillo.1.in

by onepoint＠starurchin.org

Johannes Hofmann wrote: > On Thu, Mar 07, 2013 at 08:06:13PM +0100, Johannes Hofmann wrote: > > On Thu, Mar 07, 2013 at 06:36:17PM +0000, Jeremy Henty wrote: > > > > > > Did someone forget to add a file? (Build log attached.) > > > > The file seems to be there: > > http://hg.dillo.org/dillo/file/86f817232a5f/doc/dillo.1.in > > I think the problem is related to the separate "Build" directory in > > your case. How did you set this up? > > I'd like to fix that case. > > Should be fixed now. Please give it a try. Yes, it's fixed. Thanks! Jeremy Henty

March 8, 2013

Missing dillo.1.in

by Johannes.Hofmann＠gmx.de

On Thu, Mar 07, 2013 at 08:06:13PM +0100, Johannes Hofmann wrote: > On Thu, Mar 07, 2013 at 06:36:17PM +0000, Jeremy Henty wrote: > > > > Did someone forget to add a file? (Build log attached.) > > The file seems to be there: > http://hg.dillo.org/dillo/file/86f817232a5f/doc/dillo.1.in > I think the problem is related to the separate "Build" directory in > your case. How did you set this up? > I'd like to fix that case. Should be fixed now. Please give it a try. Cheers, Johannes

March 7, 2013

Missing dillo.1.in

by Johannes.Hofmann＠gmx.de

On Thu, Mar 07, 2013 at 06:36:17PM +0000, Jeremy Henty wrote: > > Did someone forget to add a file? (Build log attached.) The file seems to be there: http://hg.dillo.org/dillo/file/86f817232a5f/doc/dillo.1.in I think the problem is related to the separate "Build" directory in your case. How did you set this up? I'd like to fix that case. Regards, Johannes > > Regards, > > Jeremy Henty > make all-recursive > make[1]: Entering directory `/home/jeremy/Personal/Geeky/Dillo/Build' > Making all in lout > make[2]: Entering directory `/home/jeremy/Personal/Geeky/Dillo/Build/lout' > make[2]: Nothing to be done for `all'. > make[2]: Leaving directory `/home/jeremy/Personal/Geeky/Dillo/Build/lout' > Making all in dw > make[2]: Entering directory `/home/jeremy/Personal/Geeky/Dillo/Build/dw' > make[2]: Nothing to be done for `all'. > make[2]: Leaving directory `/home/jeremy/Personal/Geeky/Dillo/Build/dw' > Making all in dlib > make[2]: Entering directory `/home/jeremy/Personal/Geeky/Dillo/Build/dlib' > make[2]: Nothing to be done for `all'. > make[2]: Leaving directory `/home/jeremy/Personal/Geeky/Dillo/Build/dlib' > Making all in dpip > make[2]: Entering directory `/home/jeremy/Personal/Geeky/Dillo/Build/dpip' > make[2]: Nothing to be done for `all'. > make[2]: Leaving directory `/home/jeremy/Personal/Geeky/Dillo/Build/dpip' > Making all in src > make[2]: Entering directory `/home/jeremy/Personal/Geeky/Dillo/Build/src' > Making all in IO > make[3]: Entering directory `/home/jeremy/Personal/Geeky/Dillo/Build/src/IO' > make[3]: Nothing to be done for `all'. > make[3]: Leaving directory `/home/jeremy/Personal/Geeky/Dillo/Build/src/IO' > make[3]: Entering directory `/home/jeremy/Personal/Geeky/Dillo/Build/src' > make[3]: Nothing to be done for `all-am'. > make[3]: Leaving directory `/home/jeremy/Personal/Geeky/Dillo/Build/src' > make[2]: Leaving directory `/home/jeremy/Personal/Geeky/Dillo/Build/src' > Making all in doc > make[2]: Entering directory `/home/jeremy/Personal/Geeky/Dillo/Build/doc' > sed 's%/usr/local%/usr%g' < dillo.1.in > dillo.1 > /bin/sh: dillo.1.in: No such file or directory > make[2]: *** [dillo.1] Error 1 > make[2]: Leaving directory `/home/jeremy/Personal/Geeky/Dillo/Build/doc' > make[1]: *** [all-recursive] Error 1 > make[1]: Leaving directory `/home/jeremy/Personal/Geeky/Dillo/Build' > make: *** [all] Error 2 > _______________________________________________ > Dillo-dev mailing list > Dillo-dev at dillo.org > http://lists.auriga.wearlab.de/cgi-bin/mailman/listinfo/dillo-dev

March 7, 2013

Missing dillo.1.in

by onepoint＠starurchin.org

Did someone forget to add a file? (Build log attached.) Regards, Jeremy Henty

March 7, 2013

release preperations

by Johannes.Hofmann＠gmx.de

On Wed, Mar 06, 2013 at 10:28:34PM +0100, Sebastian Geerken wrote: > On Wed, Mar 06, Johannes Hofmann wrote: > > On Tue, Mar 05, 2013 at 11:35:02PM +0100, Sebastian Geerken wrote: > > > "man dillo" shows > > > > > > In order to use the hyphenation feature, pattern files from CTAN need > > > to be installed to ${prefix}/lib/dillo/hyphenation/. This can be done > > > with the script install-hyphenation. Call it with ISO-639-1 language > > > codes as arguments, or without arguments to get more help. > > > > > > "${prefix}" should be replaced here. > > > > Yes. Does someone know how this is best done with automake? > > Move doc/dillo.1 to doc/dillo.1.in and add this to doc/Makefile.am: > > diff -r f37ab8cb8930 doc/Makefile.am > --- a/doc/Makefile.am Tue Mar 05 19:47:09 2013 +0100 > +++ b/doc/Makefile.am Wed Mar 06 22:24:33 2013 +0100 > @@ -39,3 +39,6 @@ > Selection.txt \ > Dpid.txt \ > README > + > +dillo.1: > + sed 's%.{prefix}%${prefix}%g' < dillo.1.in > dillo.1 > > This replaces ${prefix} by the actual value. I could not get the "$" > into the sed expression (therefore the "."), so this is not perfect. Thanks, I pushed it. Johannes

March 6, 2013

release preperations

by sgeerken＠dillo.org

On Wed, Mar 06, Johannes Hofmann wrote: > On Tue, Mar 05, 2013 at 11:35:02PM +0100, Sebastian Geerken wrote: > > "man dillo" shows > > > > In order to use the hyphenation feature, pattern files from CTAN need > > to be installed to ${prefix}/lib/dillo/hyphenation/. This can be done > > with the script install-hyphenation. Call it with ISO-639-1 language > > codes as arguments, or without arguments to get more help. > > > > "${prefix}" should be replaced here. > > Yes. Does someone know how this is best done with automake? Move doc/dillo.1 to doc/dillo.1.in and add this to doc/Makefile.am: diff -r f37ab8cb8930 doc/Makefile.am --- a/doc/Makefile.am Tue Mar 05 19:47:09 2013 +0100 +++ b/doc/Makefile.am Wed Mar 06 22:24:33 2013 +0100 @@ -39,3 +39,6 @@ Selection.txt \ Dpid.txt \ README + +dillo.1: + sed 's%.{prefix}%${prefix}%g' < dillo.1.in > dillo.1 This replaces ${prefix} by the actual value. I could not get the "$" into the sed expression (therefore the "."), so this is not perfect. > > > Also we would need to install the install-hyphenation script, so > > > that it becomes part of binary packages. > > > > There would also some changes neccessary, since the script searches > > "Makefile". > > > > This is an aspect I have not yet thought about: whether the script is > > part of the installation process (as now) or the installation itself > > (as you suggest). > > If we install it it will probabely end up in binary packages and be > available for end users. > If we don't install it, we rely on packagers to deal with the > problem (e.g. to create a separate hyphenation pattern package). > > > > If we do that it would have to be renamed to > > > dillo-install-hyphenation or something like that. > > > Does this sound ok? I'll try to modify the script, so it can be installed to ${prefix}/bin. In any case, this is certainly useful. > [...] > > OTOH, if someone installs dillo from the tarball, and does not notice > > that she should download pattern files. Since hyphenation is inactive, > > when there is no pattern file installed, this feature will be > > invisible. > > > > That's why I thought to put it into the Makefile, so he sees the > > notice when he calls "make install". That's what I meant with my > > question. > > We can of course also add the notice to make install, that should > not harm anyone. Although I must admit that this approach only works for one message. Sebastian

March 6, 2013

release preperations

by Johannes.Hofmann＠gmx.de

On Tue, Mar 05, 2013 at 11:35:02PM +0100, Sebastian Geerken wrote: > Hi, > > I hope I am, as the developer of hyphenation, not biased on this > topic. :-) > > > On Sat, Mar 02, 2013 at 07:57:26PM +0100, Sebastian Geerken wrote: > > > > > As for the way to inform the user, see attached patch, which will let > > > > > "make install" (and friends) print a message. Comments? > > > > > > > > Nice, but most users will get precompiled packages from their > > > > distributions. > > > > > > > > > How is the other work on the release going on? What is still missing? > > > > > > > > I started a couple of times to add some lines to the man page that > > > > explain how to get and install the pattern files, but never came up > > > > with anything good. Should we copy the lines from your patch below? > > > > > > How many people compiling dillo from the tarball are actually reading > > > README and the man page? > > > > At least the man page get's installed with most binary packages. > > I tried to include some info about the pattern files and how > > they can be downloaded - please give it a review. > > "man dillo" shows > > In order to use the hyphenation feature, pattern files from CTAN need > to be installed to ${prefix}/lib/dillo/hyphenation/. This can be done > with the script install-hyphenation. Call it with ISO-639-1 language > codes as arguments, or without arguments to get more help. > > "${prefix}" should be replaced here. Yes. Does someone know how this is best done with automake? > > > Also we would need to install the install-hyphenation script, so > > that it becomes part of binary packages. > > There would also some changes neccessary, since the script searches > "Makefile". > > This is an aspect I have not yet thought about: whether the script is > part of the installation process (as now) or the installation itself > (as you suggest). If we install it it will probabely end up in binary packages and be available for end users. If we don't install it, we rely on packagers to deal with the problem (e.g. to create a separate hyphenation pattern package). > > > If we do that it would have to be renamed to > > dillo-install-hyphenation or something like that. > > Does this sound ok? > > I'd leave the question, how binary packages are handled, to the > maintainers of the binary packages, since they know best how to handle > this. (E. g., Debian provides packages texlive-lang-... for each > language.) > > Perhaps package maintainers may comment on this. That would be really helpful. > > OTOH, if someone installs dillo from the tarball, and does not notice > that she should download pattern files. Since hyphenation is inactive, > when there is no pattern file installed, this feature will be > invisible. > > That's why I thought to put it into the Makefile, so he sees the > notice when he calls "make install". That's what I meant with my > question. We can of course also add the notice to make install, that should not harm anyone. Johannes

March 6, 2013

release preperations

by sgeerken＠dillo.org

Hi, I hope I am, as the developer of hyphenation, not biased on this topic. :-) > On Sat, Mar 02, 2013 at 07:57:26PM +0100, Sebastian Geerken wrote: > > > > As for the way to inform the user, see attached patch, which will let > > > > "make install" (and friends) print a message. Comments? > > > > > > Nice, but most users will get precompiled packages from their > > > distributions. > > > > > > > How is the other work on the release going on? What is still missing? > > > > > > I started a couple of times to add some lines to the man page that > > > explain how to get and install the pattern files, but never came up > > > with anything good. Should we copy the lines from your patch below? > > > > How many people compiling dillo from the tarball are actually reading > > README and the man page? > > At least the man page get's installed with most binary packages. > I tried to include some info about the pattern files and how > they can be downloaded - please give it a review. "man dillo" shows In order to use the hyphenation feature, pattern files from CTAN need to be installed to ${prefix}/lib/dillo/hyphenation/. This can be done with the script install-hyphenation. Call it with ISO-639-1 language codes as arguments, or without arguments to get more help. "${prefix}" should be replaced here. > Also we would need to install the install-hyphenation script, so > that it becomes part of binary packages. There would also some changes neccessary, since the script searches "Makefile". This is an aspect I have not yet thought about: whether the script is part of the installation process (as now) or the installation itself (as you suggest). > If we do that it would have to be renamed to > dillo-install-hyphenation or something like that. > Does this sound ok? I'd leave the question, how binary packages are handled, to the maintainers of the binary packages, since they know best how to handle this. (E. g., Debian provides packages texlive-lang-... for each language.) Perhaps package maintainers may comment on this. OTOH, if someone installs dillo from the tarball, and does not notice that she should download pattern files. Since hyphenation is inactive, when there is no pattern file installed, this feature will be invisible. That's why I thought to put it into the Makefile, so he sees the notice when he calls "make install". That's what I meant with my question. Sebastian

March 5, 2013

release preperations

by Johannes.Hofmann＠gmx.de

Hi, On Sat, Mar 02, 2013 at 07:57:26PM +0100, Sebastian Geerken wrote: > On Sat, Mar 02, Johannes Hofmann wrote: > > > I've pushed a new script, which uses the text versions (and adds the > > > licence information as comments). Should it download directly from > > > CTAN? From mirrors.dotsrc.org? Ask for a mirror? > > > > If mirrors.dotsrc.org works, that's probabely nicer - though I doubt > > we create any significant load. > > Mirrors.dotsrc.org works; I've changed the script. (Look at the > beginning.) > > > > As for the way to inform the user, see attached patch, which will let > > > "make install" (and friends) print a message. Comments? > > > > Nice, but most users will get precompiled packages from their > > distributions. > > > > > How is the other work on the release going on? What is still missing? > > > > I started a couple of times to add some lines to the man page that > > explain how to get and install the pattern files, but never came up > > with anything good. Should we copy the lines from your patch below? > > How many people compiling dillo from the tarball are actually reading > README and the man page? At least the man page get's installed with most binary packages. I tried to include some info about the pattern files and how they can be downloaded - please give it a review. Also we would need to install the install-hyphenation script, so that it becomes part of binary packages. If we do that it would have to be renamed to dillo-install-hyphenation or something like that. Does this sound ok? Cheers, Johannes PS: For me the documentation issue for hyphenation patterns is the only thing left before we could make a release candidate, or am I missing something?

March 5, 2013

https related crash

by jcid＠dillo.org

On Mon, Mar 04, 2013 at 10:07:21PM +0100, Johannes Hofmann wrote: > On Fri, Mar 01, 2013 at 01:37:13PM -0300, Jorge Arellano Cid wrote: > > Hi Johannes, > > > > On Thu, Feb 28, 2013 at 10:44:15PM +0100, Johannes Hofmann wrote: > > > Hi, > > > > > > from time to time I get the crash below. > > > Would a simple NULL check in a_Chain_check() be the correct fix? > > > > I gave it a first review, and it looks like the connection gets closed > > before the dialog sends an answer. So, at resume time, the sending branch > > is gone. It'd be great to have a test case. > > Sorry, I can't reproduce it atm. I keep trying. > > > > > Anyway, in the interim, please try the following patch: > > Will do, once I can reproduce the crash. AFAIR there're a few obscure corner cases not explicitly handled by the CCC (ATM not worth the coding effort), and those are silently caught by the CCC's check code. From a distance, and with the proposed patch, it doesn't look like a show stopper at all. -- Cheers Jorge.-

March 5, 2013

https related crash

by Johannes.Hofmann＠gmx.de

On Fri, Mar 01, 2013 at 01:37:13PM -0300, Jorge Arellano Cid wrote: > Hi Johannes, > > On Thu, Feb 28, 2013 at 10:44:15PM +0100, Johannes Hofmann wrote: > > Hi, > > > > from time to time I get the crash below. > > Would a simple NULL check in a_Chain_check() be the correct fix? > > I gave it a first review, and it looks like the connection gets closed > before the dialog sends an answer. So, at resume time, the sending branch > is gone. It'd be great to have a test case. Sorry, I can't reproduce it atm. I keep trying. > > Anyway, in the interim, please try the following patch: Will do, once I can reproduce the crash. Thanks, Johannes

March 4, 2013

Profiling dillo

by p37sitdu＠lavabit.com

On Sat, Mar 02, 2013 at 11:36:16PM +0000, corvid wrote: > Sebastian wrote: > > (According to the man page, "gmon.out" is created in the current > > working directory, when exit() is called.) > > There's a chdir("/tmp") hiding in Paths::init(). > > Why don't remove it? >From ChangeLog: - Fixed Dillo and its dpis to work from "/tmp" (for easy device unmount). There was a thread discussing it: http://lists.auriga.wearlab.de/pipermail/dillo-dev/2011-December/009221.html http://lists.auriga.wearlab.de/pipermail/dillo-dev/2011-December/009223.html http://lists.auriga.wearlab.de/pipermail/dillo-dev/2011-December/009225.html Because of this, open dialog starts in /tmp, and it is a bug. Usually you start dillo from your home directory, why would you want to unmount it? And if you really plan to unmount it, you can change working directory yourself.

March 3, 2013

Profiling dillo

by corvid＠lavabit.com

Sebastian wrote: > (According to the man page, "gmon.out" is created in the current > working directory, when exit() is called.) There's a chdir("/tmp") hiding in Paths::init().

March 2, 2013

Profiling dillo

by sgeerken＠dillo.org

On Sat, Mar 02, Johannes Hofmann wrote: > On Fri, Mar 01, 2013 at 10:49:30AM +0100, Sebastian Geerken wrote: > > Has anyone tried to use gprof for profiling dillo? Using > > "--enable-gprof" for configure seems to work, and calling the tests > > (e. g.. test/dw-example) create the file "gmon.out", but not so > > does starting src/dillo. The option "-pg" is set everywhere, so I do > > not know what the difference is. > > > > Any ideas? > > Did you look in /tmp for the gmon.out file? Cores also end up there. Ah, yes, it is there. Thanks! (According to the man page, "gmon.out" is created in the current working directory, when exit() is called.) Sebastian

March 2, 2013

Profiling dillo

by Johannes.Hofmann＠gmx.de

Hi, On Fri, Mar 01, 2013 at 10:49:30AM +0100, Sebastian Geerken wrote: > Hi! > > Has anyone tried to use gprof for profiling dillo? Using > "--enable-gprof" for configure seems to work, and calling the tests > (e. g.. test/dw-example) create the file "gmon.out", but not so > does starting src/dillo. The option "-pg" is set everywhere, so I do > not know what the difference is. > > Any ideas? Did you look in /tmp for the gmon.out file? Cores also end up there. Johannes

March 2, 2013

https related crash

by jcid＠dillo.org

Hi Johannes, On Thu, Feb 28, 2013 at 10:44:15PM +0100, Johannes Hofmann wrote: > Hi, > > from time to time I get the crash below. > Would a simple NULL check in a_Chain_check() be the correct fix? I gave it a first review, and it looks like the connection gets closed before the dialog sends an answer. So, at resume time, the sending branch is gone. It'd be great to have a test case. Anyway, in the interim, please try the following patch: diff -r 8c8975054b06 src/capi.c --- a/src/capi.c Wed Jan 30 10:04:00 2013 +0100 +++ b/src/capi.c Fri Mar 01 13:32:38 2013 -0300 @@ -571,10 +571,13 @@ int a_Capi_dpi_send_data(const DilloUrl /* Re-use an open connection */ conn = Capi_conn_find(server); if (conn) { - /* found */ - dbuf = a_Chain_dbuf_new(data, data_sz, 0); - a_Capi_ccc(OpSend, 1, BCK, conn->InfoSend, dbuf, NULL); - dFree(dbuf); + if (conn->InfoSend) { + /* found & operative*/ + dbuf = a_Chain_dbuf_new(data, data_sz, 0); + a_Capi_ccc(OpSend, 1, BCK, conn->InfoSend, dbuf, NULL); + dFree(dbuf); + } else + MSG(" ERROR: [a_Capi_dpi_send_data] Connection not operative\n"); } else { MSG(" ERROR: [a_Capi_dpi_send_data] No open connection found\n"); } diff -r 8c8975054b06 src/chain.c --- a/src/chain.c Wed Jan 30 10:04:00 2013 +0100 +++ b/src/chain.c Fri Mar 01 13:32:38 2013 -0300 @@ -189,7 +189,10 @@ int a_Chain_check(char *FuncStr, int Op, /* Show status information */ Chain_debug_msg(FuncStr, Op, Branch, Dir, Info); - if (Info->Flags & (CCC_Ended + CCC_Aborted)) { + if (!Info) { + MSG_WARN("CCC: call on a NULL node.\n" + "Caught as last resort; Most probably a BUG.\n"); + } else if (Info->Flags & (CCC_Ended + CCC_Aborted)) { /* CCC is not operative */ MSG_WARN("CCC: call on already finished chain. Flags=%s%s\n", Info->Flags & CCC_Ended ? "CCC_Ended " : "", -- Cheers Jorge.-

March 1, 2013

Profiling dillo

by sgeerken＠dillo.org

Hi! Has anyone tried to use gprof for profiling dillo? Using "--enable-gprof" for configure seems to work, and calling the tests (e. g.. test/dw-example) create the file "gmon.out", but not so does starting src/dillo. The option "-pg" is set everywhere, so I do not know what the difference is. Any ideas? Sebastian

March 1, 2013

https related crash

by Johannes.Hofmann＠gmx.de

Hi, from time to time I get the crash below. Would a simple NULL check in a_Chain_check() be the correct fix? Cheers, Johannes nated with signal 11, Segmentation fault. #0 0x0805d359 in a_Chain_check (FuncStr=0x81135b3 "a_Capi_ccc", Op=2, Branch=1, Dir=2, Info=0x0) at chain.c:192 192 if (Info->Flags & (CCC_Ended + CCC_Aborted)) { (gdb) p Info $1 = (ChainLink *) 0x0 (gdb) bt #0 0x0805d359 in a_Chain_check (FuncStr=0x81135b3 "a_Capi_ccc", Op=2, Branch=1, Dir=2, Info=0x0) at chain.c:192 #1 0x08062f03 in a_Capi_ccc (Op=2, Branch=1, Dir=2, Info=0x0, Data1=0x28c83670, Data2=0x0) at capi.c:622 #2 0x0806372f in a_Capi_dpi_send_data (url=0x0, bw=0x2871da20, data=0x29563748 "<cmd='answer' to_cmd='dialog' msg='1' '>", data_sz=40, server=0x2940ee70 "proto.https", flags=<optimized out>) at capi.c:576 #3 0x0806381a in a_Capi_dpi_send_cmd (url=0x0, bw=0x2871da20, cmd=0x29563748 "<cmd='answer' to_cmd='dialog' msg='1' '>", server=0x2940ee70 "proto.https", flags=0) at capi.c:593 #4 0x08085bed in Dpiapi_dialog_answer_cb (answer=<optimized out>, bw=<optimized out>) at dpiapi.c:43 #5 a_Dpiapi_dialog (bw=0x2871da20, server=0x2940ee70 "proto.https", dpip_tag=0x286612d0 "<cmd='dialog' title='Dillo HTTPS: Missing issuer certificate!' msg='Unable to get local issuer certificate. The issuer certificate of an untrusted certificate cannot be found.' alt1='Continue' alt2='C"...) at dpiapi.c:74 #6 0x0805d21f in a_Chain_fcb (Op=684209776, Info=0x2878cec0, Data1=0x286612d0, Data2=0x29ecb9d0) at chain.c:114 #7 0x0808b533 in Dpi_parse_token (conn=<optimized out>) at dpi.c:228 #8 Dpi_process_dbuf (conn=<optimized out>, Data1=<optimized out>, Op=<optimized out>) at dpi.c:326 #9 a_Dpi_ccc (Op=2, Branch=2, Dir=1, Info=0x2878cec0, Data1=0x29e55470, Data2=0x0) at dpi.c:713 #10 0x0805d21f in a_Chain_fcb (Op=684209776, Info=0x28788fc0, Data1=0x29e55470, Data2=0x0) at chain.c:114 #11 0x0808ba6e in a_IO_ccc (Op=2, Branch=2, Dir=1, Info=0x28788fc0, Data1=0x292f6d40, Data2=0x0) at IO.c:425 #12 0x0808bc80 in IO_read (io=0x292f6d40) at IO.c:194 #13 0x0808bd88 in IO_callback (io=0x0) at IO.c:259 #14 0x0808be47 in IO_fd_read_cb (fd=6, data=0x6d7) at IO.c:280 #15 0x080e7d15 in fl_wait(double) () #16 0x080c36fc in Fl::wait(double) () #17 0x080c377b in Fl::run() () #18 0x0804e1f1 in main (argc=1, argv=0xbfbff8c4) at dillo.cc:502 (gdb)

Feb. 28, 2013

text-transform and hyphenation

by sgeerken＠dillo.org

On Sat, Feb 16, corvid wrote: > 'text-transform: capitalize' and hyphenation do not play well together > (example attached). The fragments are capitalized, and the word is > sometimes split as "Cap-" "Talization", losing the 'i'. Fixed. Sebastian

Feb. 16, 2013

text-transform and hyphenation

by corvid＠lavabit.com

'text-transform: capitalize' and hyphenation do not play well together (example attached). The fragments are capitalized, and the word is sometimes split as "Cap-" "Talization", losing the 'i'.

Feb. 16, 2013

form input button and margin

by sgeerken＠dillo.org

On Sun, Feb 02, corvid wrote: > I wrote: > > In the second one, the words and the button overlap. I'm not immediately > > sure what's special enough about that case to cause that. > > Here's some quick code made by looking at how the image code is working. > I would not be surprised if there are still complications lurking > somewhere in textblock, but it's at least something to play with... First of all, two remarks: (i) the border should also be added in getExtremesImpl; (ii) Embed::draw should either work similar to Image::draw (testing for intersection), or as currently, which is probably preferrably, since simpler, and drawing only some pixels more than needed. > Aside: It seems that, at least for firefox, the border is the edge of a > button or text entry widget or whatever, and padding goes inside that. > Of course we're not in any position to do that, but it's interesting. I do not have specs at hand, but I assume that this is correct. For this reason, I'd rather let the resource, not the embed widget, handle border and padding. For a first version, the standard borders (outset for button, inset for input fields, etc.) could be drawn; for an improvement, custom Fltk widgets would be neccessary. For reasons of uniformity, I'd let the resource als handle margin, which should then be implemented somewhere in flkt::ui. BTW: Johannes and I are planning to cleanup size handling; I'd like to get rid of size hints, and so also of the distinction "uses size hints". This will make it simpler to add <button>'s inline. Sebastian

Feb. 8, 2013

IO/proto.c

by p37sitdu＠lavabit.com

File is empty except for comment. Shouldn't it be removed?

Feb. 7, 2013

form input button and margin

by corvid＠lavabit.com

I wrote: > In the second one, the words and the button overlap. I'm not immediately > sure what's special enough about that case to cause that. Here's some quick code made by looking at how the image code is working. I would not be surprised if there are still complications lurking somewhere in textblock, but it's at least something to play with... Aside: It seems that, at least for firefox, the border is the edge of a button or text entry widget or whatever, and padding goes inside that. Of course we're not in any position to do that, but it's interesting.

Feb. 3, 2013

form input button and margin

by corvid＠lavabit.com

In the second one, the words and the button overlap. I'm not immediately sure what's special enough about that case to cause that. <button type="submit" style="font-size: 200%; margin-bottom: 1em; width: 100%;">button submit margin bottom</button> words words <p> <input type="submit" value="input submit margin bottom BAD" style="font-size: 200%; margin-bottom: 1em; width: 100%;"> words words <p> <input type="submit" value="input submit" style="font-size: 200%; width: 100%;"> words words <p> <div style="font-size: 200%; margin-bottom: 1em; width: 100%; border-width: 1px; border-style: solid">div margin bottom</div> words words <p> <img src=file:/tmp/whatever border=1 style="margin-bottom: 1em; width: 100%"> words words

Feb. 3, 2013

[patch] remove old comment from src/IO/IO.c

by p37sitdu＠lavabit.com

There is an outdated comment in src/IO/IO.c. hg blame src/IO/IO.c also shows that it is older than the code. This patch removes comment. #defines are similar to SHUT_RD, SHUT_WR and SHUT_RDWR, but shutdown is not used: code just dCloses descriptor when both sides are asked to be closed.

Feb. 2, 2013

chain.c cleanup

by p37sitdu＠lavabit.com

I have updated patch and split it into 7 pathes so it is easier to check it. Description for each patch is embedded. http://www.inventati.org/noname/dillo/chain/

Feb. 1, 2013