- Dillo-dev - lists.mailman3.com

trimming system includes
by corvid＠lavabit.com June 22, 2009

June 22, 2009

Johannes wrote: > On Mon, Jun 22, 2009 at 02:08:01AM +0000, corvid wrote: > > Johannes wrote: > > > I'm all for it and will be happy to test it on DragonFly BSD. > > > > Does this come reasonably close to compiling? > > Works just fine here. Well, if it works for Johannes... :) Anyway, I looked through the cvs archive a little, and it was interesting to see the prehistoric origins of some of this stuff, dating as far back as gzilla presumably. One is unavoidably reminded of how a living thing today bears the mark of billions of years of evolutionary past.

2 1

trimming system includes
by corvid＠lavabit.com June 21, 2009

June 21, 2009

Would anyone have strong objections if I go through and make some patches to remove system includes where I can't see any evidence that any code now in that file still needs that header?

3 3

experimental patch: nowrap
by corvid＠lavabit.com June 20, 2009

June 20, 2009

Currently dillo does not handle the case where certain regions of a line can be wrapped and other regions cannot. A prominent example is the list of browsers at the bottom of http://en.wikipedia.org/wiki/Dillo Here is a patch that minimizes memory use (adds one int per textblock) at some cost of recalculating things http://www.dillo.org/test/nowrap.patch When whitespace isn't WHITE_SPACE_NORMAL, Textblock::add* functions still add words to the word list, but postpone wrapping until the entire nowrap/pre chunk has been added or a break has been seen. This is so that we know whether to put it all on a new line. Some possibilities: Putting some kind of spanWidth in each Word, calculating it in addNowrapWords and sticking this nonzero value into the beginning word. We'd have to remember to deal with it if things in the span changed size, or if dillo got on-the-fly display:none someday or whatever. Taking the time to stick the whole chunk into one word when the style is all the same. (Similar difficulties as the above.) I'm not sure whether that would make life more difficult if dillo got bidirectional text someday. Putting the nowrap Words into a Something, so that what is now a Word list would contain Word Word Word Something Word Something Word... That sounds complicated. Let the add* functions put the nowrap Words through wordWrap, and keep track of shifts between normal/nowrap in a line. When a line is completed, set up rewrap to do it over. This would be in an effort to "push down the bubble" of ickiness around add*/addNowrapWords/rewrap, although the resulting new bubble of ickiness that would pop up next to it would perhaps be similar. Small things that might or might not be improvements: make breaks have normal whitespace, and make lines with only one word have normal whitespace (text/plain).

1 0

dillo-2.1.tar.bz2
by jcid＠dillo.org June 18, 2009

June 18, 2009

Hi, Working on: 1.- Solve the OpenSSL license issue and upload rc3. On Thu, Jun 18, 2009 at 05:24:25PM +0200, Devid Antonio Filoni wrote: > On Wed, Jun 17, 2009 at 11:45 PM, Jorge Arellano Cid<jcid(a)dillo.org> wrote: > > On Wed, Jun 17, 2009 at 10:01:12PM +0200, Devid Antonio Filoni wrote: > > [...] > > ?Devid: please send me the .deb and its directions when done. > DONE! You can find it at https://launchpad.net/~d.filoni/+archive/ppa > . However I found an error checking my package using lintian: > > E: dillo: possible-gpl-code-linked-with-openssl > N: > N: This package appears to be covered by the GNU GPL but depends on the > N: OpenSSL libssl package and does not mention a license exemption or > N: exception for OpenSSL in its copyright file. The GPL (including version > N: 3) is incompatible with some terms of the OpenSSL license, and therefore > N: Debian does not allow GPL-licensed code linked with OpenSSL libraries > N: unless there is a license exception explicitly permitting this. > N: > N: If only the Debian packaging, or some other part of the package not > N: linked with OpenSSL, is covered by the GNU GPL, please add a lintian > N: override for this tag. Lintian currently has no good way of > N: distinguishing between that case and problematic packages. > N: > N: Severity: serious, Certainty: wild-guess > > Can you please fix it? AFAIS, the only program that links to OpenSSL is the https dpi. i.e. The ssl-enabled dillo binary doesn't link to it. https.filter.dpi links to it. IANAL, but it looks like giving permission for this dpi is enough, and it's already done in its sources (dpi/https.c): * [...] * As a special exception permission is granted to link the code of * the https dillo plugin with the OpenSSL project's "OpenSSL" * library, and distribute the linked executables, without including * the source code for OpenSSL in the source distribution. You must * obey the GNU General Public License, version 3, in all respects * for all of the code used other than "OpenSSL". * */ Please let me know if this is _really_ enough or if some extra legal formula must be added somewhere else. TIA. -- Cheers Jorge.-

2 2

dillo-2.1.tar.bz2
by jcid＠dillo.org June 18, 2009

June 18, 2009

On Thu, Jun 18, 2009 at 05:24:25PM +0200, Devid Antonio Filoni wrote: > On Wed, Jun 17, 2009 at 11:45 PM, Jorge Arellano Cid<jcid(a)dillo.org> wrote: > > [...] > > ?BTW, the rc2 version of dillo-2.1 (hopefully final) was > > just uploaded to http://www.dillo.org/download/ so packagers > > can make the respective files. > > > > ?The idea is to make the announcement tomorrow in the morning. > > > > ?Devid: please send me the .deb and its directions when done. > DONE! You can find it at https://launchpad.net/~d.filoni/+archive/ppa > . However I found an error checking my package using lintian: > > E: dillo: possible-gpl-code-linked-with-openssl > N: > N: This package appears to be covered by the GNU GPL but depends on the > N: OpenSSL libssl package and does not mention a license exemption or > N: exception for OpenSSL in its copyright file. The GPL (including version > N: 3) is incompatible with some terms of the OpenSSL license, and therefore > N: Debian does not allow GPL-licensed code linked with OpenSSL libraries > N: unless there is a license exception explicitly permitting this. > N: > N: If only the Debian packaging, or some other part of the package not > N: linked with OpenSSL, is covered by the GNU GPL, please add a lintian > N: override for this tag. Lintian currently has no good way of > N: distinguishing between that case and problematic packages. > N: > N: Severity: serious, Certainty: wild-guess > > Can you please fix it? OK guys, don't hurry! ---------------------------------------------------- Let's make the announcement early this Saturday 20. ---------------------------------------------------- That way we have until tomorrow to sort out the packaging problem and this license issue. I'd proceed in this order: 1.- Solve the OpenSSL license issue and upload rc3. 2.- To enable-ssl or not enable-ssl (define what to do). 3.- Get to the less painful package-system for the user given the current situation. (most probably distribution repositories) 4.- Future planning: how to get dillo2 into distros to enlarge its user base. FLTK2 vs FLTK-1.3 * I have plenty of info here, but will hold on until the previous points are done. Thanks for all the help. -- Cheers Jorge.-

1 0

dillo-2.1.tar.bz2
by jcid＠dillo.org June 17, 2009

June 17, 2009

On Wed, Jun 17, 2009 at 03:33:56AM +0000, corvid wrote: > I just remembered that the README needs to be updated > since it talks about all of the new things that 2.0 brought. I just updated the repo. Please give it a review and fix the grammar where necessary. The patch for working keyboard under numlock was also committed (overlooked by me). Those having trouble with rc1, please confirm the repo (rc2) works OK now. When I get both OKs, I'll upload a new tarball for making the respective packages. If there's anything else pending, speak now! -- Cheers Jorge.-

1 0

keyboard shortcuts non-functioning under KDE 3.5.10
by paul＠claws-mail.org June 17, 2009

June 17, 2009

Hi, I've got recent code from mercurial, June 1st. No keyboard shortcuts are working for me using KDE 3.5.10. I was speaking to Johannes Hofmann about this problem, he asked me what my locale was (it's en-GB.UTF-8) and also asked me to report the problem here. Do you need more information from me? Whilst I'm here I would like to say thanks to Johannes for sorting out the --xid option, Claws Mail now has a dillo plugin again! with regards Paul -- "Thanks to his green candle, the 'Pataphysician can lighten our darkness."

6 10

dillo-2.1.tar.bz2
by jcid＠dillo.org June 16, 2009

June 16, 2009

On Tue, Jun 16, 2009 at 09:34:43PM +0200, Devid Antonio Filoni wrote: > I'm working on the Debian/Ubuntu package. I'm the Debian developer of > dillo so I know what I do, however I will not upload this in Debian > until fltk doesn't fix the license (it should be December 2001 in > order to get uploaded in Debian). I will send you the package > tomorrow. Excellent. -- Cheers Jorge.-

2 1

Limiting number connections made to server
by dennisn＠dennisn.dyndns.org June 16, 2009

June 16, 2009

Currently, I don't think dillo places any kind of regulation on the number of connections it tries to make with an http server, and this causes bad things to happen with some -- ie. if it tries to open a page with hundreds of thumbnails, it can crash some http servers. Is there a way to limit the number of connections made?

3 3

dillo-2.1.tar.bz2
by jcid＠dillo.org June 16, 2009

June 16, 2009

Hi there, We're approaching quickly to the release of dillo-2.1 ! :) The idea is to make the official announcement this Thursday 18 (June) I've uploaded the source tarball: http://www.dillo.org/download/ We need the main .deb and .rpm before Thursday (e.g. Ubuntu & Debian current, same for the most used Fedora). Would you mind Andreas, Kelson? Any comments, confirmation of OK-ness, tests or last minute touches, this is the moment. TIA. -- Cheers Jorge.-

1 0

release candidate 1
by jcid＠dillo.org June 15, 2009

June 15, 2009

Johannes, Corvid, I'll be out of town this weekend, please attend any problem on the repo. as it's rc1. The idea is to make a release next week. On the pending things that you may like to advance: - write the dillo-2.1 about:splash page - check making packages - polish FAQ and help text -- Cheers Jorge.-

2 2

memory use
by jcid＠dillo.org June 13, 2009

June 13, 2009

On Sat, Jun 13, 2009 at 01:14:51AM +0000, corvid wrote: > Interesting that Firefox has a greater VSZ without images. > I don't think I noticed that earlier. Oh, I didn't notice too. Is this result reproducible in your computer? -- Cheers Jorge.-

3 2

[patch] Improving keyboard shortcuts for tab navigation
by tim.nieradzik＠gmx.de June 13, 2009

June 13, 2009

This patch adds keyboard shortcuts for the tabs 0 - 9. You can reach them by pressing Ctrl concurrently with the number number of the tab you would like to see. Thus, for the first tab this is Ctrl-0. Alternatively, you can press Ctrl-Home. To get to the last tab, simply press Ctrl-End. In addition the modifier for moving left and right between tabs has been changed from Shift to Ctrl because it makes sense to have an unified modifier for all tab movement operations. --Tim

7 51

[patch] change font size on the fly
by Johannes.Hofmann＠gmx.de June 12, 2009

June 12, 2009

Hi, attached patch allows to change the font size form the tools menu. I find this quite handy sometimes. What do you think? Cheers, Johannes

3 3

[PATCH] fix dillorc path
by paul＠claws-mail.org June 7, 2009

June 7, 2009

Hi, When opening dillo from the CLI I noticed the following message: paths: Cannot open file '/home/user/.dillo/dillorc' paths: Cannot open file '/etc/dillo/dillorc' paths: Using internal defaults... Checked my system and saw that dillorc had been installed in /usr/etc/dillorc Checking config.log showed sysconfdir='${prefix}/etc', so no surprises there. But Paths::getPrefsFP() in src/paths.cc uses PATHS_RC_SYS, which is "/etc/dillo/", so my patch uses DILLORC_SYS, as specified in src/Makefile.am, instead. Now all is fine! Seems the right thing to do. with regards Paul -- "Thanks to his green candle, the 'Pataphysician can lighten our darkness."

4 7

two image problems
by Johannes.Hofmann＠gmx.de June 7, 2009

June 7, 2009

Hi, I noticed two problems related to images: Some websites seems to send images with a charset set in the header, e.g: Content-Type: image/jpeg;charset=ISO-8859-1 In that case dillo seems to decode the data resulting in: Not a JPEG file: starts with 0xc3 0xbf Not sure whether we can/want to do anything about that. I don't understand the other problem: http://sz-magazin.sueddeutsche.de/upl/images/user/8059/17337.jpg loads and displays fine. from html however it doesn't work <img src="http://sz-magazin.sueddeutsche.de/upl/images/user/8059/17337.jpg" /> Cheers, Johannes

2 2

experimental patch: Re: dillo2 wrapping Chinese
by corvid＠lavabit.com June 6, 2009

June 6, 2009

dillo people: It turns out that dillo handles Chinese quite badly because the text doesn't contain whitespace (BUG 827). It looks like Japanese is that way as well. 1) People using East Asian languages: Does this do roughly the right thing? I've read that there are picky rules about certain individual characters, but I just mean in general here. 2) Thoughts on optimization are welcomed. littlebat: For some reason, I was thinking that whitespace was required between words in dillo, but I misremembered. littlebat wrote: > Hi, > Thanks for the work on Dillo. > > I think it isn't a good idea to resolve long chinese line wrapping > problem by inserting space after punctuation mark("?" or "?" or "?", > etc..). > > Here is a "Dillo2 Wrap Long Chinese Line Test" page: > http://www.learndiary.com/en/dillo2-wrap-long-chinese-line-test.htm . > Open it with Firefox or IE, then open it with Dillo2, You will see the > different of wrapping long chinese line between them. > > Note, unlike english, there is no space between Chinese characters. So, > it should be wrapped after any Chinese character which has reached the > end of a line. I don't know the technique detail about this, but, I > think, let the Chinese characters function as (english) commas or > periods is a good idea, so, a new line can be started after any Chinese > charcter which character has reached the end of a line. > > littlebat > > http://www.learndiary.com/en/ > > > corvid wrote: > >Would it make things better or worse than it is now if I inserted a > >space when encountering a "?" or "?" so that dillo would know > >that it can wrap there? If that doesn't show up properly, > >I mean the characters that function as commas and periods. > > > >(I'm trying to think of something simple that will at least help you > > somewhat without having to make big changes right now :) > > I realize that it _should_ be able to wrap after most CJK characters, > > but...) > > > >It could be worse because > >1) Those spaces would still be there if you were copying and pasting > > the text to something else. > >2) If you wanted to search a page for "text?text", you would have to > > enter "text? text" with a space in there.

3 10

patch: NOP keybinding
by corvid＠lavabit.com June 5, 2009

June 5, 2009

I think it would be nice to have a NOP binding to get rid of a default binding. complicated patch:

2 1

patch: smaller findbar buttons for narrow screen
by corvid＠lavabit.com June 4, 2009

June 4, 2009

I resized my window to make it narrower and discovered that the findbar buttons eat all of the space and make the input widget disappear. With button width reduced to 40 with a narrow window, if I resize down to match that ipaq screenshot that Andreas made, I can still see a reasonable amount of input field.

2 2

view source dpi question
by Johannes.Hofmann＠gmx.de June 2, 2009

June 2, 2009

Hi, annoyed by the slow source view widget I'd like to play with a source viewer dpi (Jorge, I think you suggested to use a dpi instead of coding it in dillo directly). However I want to avoid to reload the page from the dpi, as I want to be able to view the source exactly as it is currently displayed. How should I pass the html document from dillo to the dpi? Any ideas? Cheers, Johannes

3 6

https through a proxy
by corvid＠lavabit.com June 1, 2009

June 1, 2009

Yesterday I hacked up (it's still very hacky right now) some code that is intended to make https work through a proxy, but I haven't tested it on an actual proxy yet. Do you have any suggestions on what would be the most convenient way to go about this? (It'd be nicer just to test with something that's running out there than to do a bunch of building and configuring for something I'll just rm again afterward.)

2 4

[patch]: more doxygen fixes
by onepoint＠starurchin.org May 31, 2009

May 31, 2009

Doxygen is missing some methods because it's not smart enough about namespaces and sometimes thinks that the signatures of the declaration and the definition are different when they are not. You can see the warnings in its output (hint: redirect stdout to /dev/null to extract the warnings from the noise). Patch attached that adds some explicit namespace qualifiers. This removes doxygen's confusion, fixes the docs and silences the warnings. Regards, Jeremy Henty

4 4

Re: the *_freeall() patch
by jcid＠dillo.org May 31, 2009

May 31, 2009

On Sat, May 30, 2009 at 01:21:12AM +0000, corvid wrote: > In http://lists.auriga.wearlab.de/pipermail/dillo-dev/2009-March/006234.html > about key bindings, the matter of the *freeall()s came up. > My copy of the patch that I sent you has that same date on it, Mar 25, > and I think I sent it within maybe...two days. > > I've been running on it without incident since that time, other > than that since-fixed klist bug that it found. Committed. BTW, I also added a a_Prefs_freeall() function. Please test. -- Cheers Jorge.-

1 0

[patch]: spelling: instanciate
by onepoint＠starurchin.org May 31, 2009

May 31, 2009

The act of creating an instance is not "instanciation" but "instantiation". Patch attached. You have been reading "Great stupidities of English spelling", number 361 of a *very* long series. Regards, Jeremy Henty

3 3

patch: segfault in win_cb in uicmd.cc
by corvid＠lavabit.com May 30, 2009

May 30, 2009

I was closing dillo. So far as I can recall, I just hit ^Q a couple times (my usual method), but I don't think that would explain the win_cb(). (gdb) bt #0 0x08055a00 in UI::tabs (this=0x0) at ui.hh:105 #1 0x080555ef in win_cb (w=0x813fe48, cb_data=0x8190618) at uicmd.cc:351 #2 0x080df426 in fltk::handle () at ../fltk/Widget.h:127 #3 0x080e0995 in do_queued_events () at x11/run.cxx:399 #4 0x080e0c16 in fltk::wait (time_to_wait=1.00000002e+20) at x11/run.cxx:459 #5 0x080e0db0 in fltk::run () at run.cxx:399 #6 0x0804f737 in main (argc=1, argv=0xbffffb34) at dillo.cc:353 (gdb) p num_bws $13 = 2 (gdb) p bws[0] $14 = (BrowserWindow *) 0x81da720 (gdb) p bws[1] $15 = (BrowserWindow *) 0x81e4a10 (gdb) p bws[2] $16 = (BrowserWindow *) 0x81da720 Perhaps I had just closed the old bw[0], with a_List_remove moving the old bw[2] there. Here's a patch, since it's a good idea in any case to check what we've been given in a callback, but there is a bit of symptom-vs-disease in that I'm not positive how I got here in this situation.

1 0

--xid option
by Johannes.Hofmann＠gmx.de May 29, 2009

May 29, 2009

Hi, attached patch tries to implement the missing --xid option that is used to embed dillo into the claws mailer. The right way to do this would probabely be to implement the complete XEMBED protocol which is also mentioned in doc/fltk-problems.doc. However a simple XReparentWindow() seems to work more or less here. Please test. Cheers, Johannes

4 15

repo
by jcid＠dillo.org May 29, 2009

May 29, 2009

Johannes, In patch fd2b0634abe9 there's a long-word splitter, please add a comment on why it's there. IIRC it was discussed in dillo-dev recently but memory is fragile. BTW, there's a warning too: xembed.cc: In member function 'void Xembed::createInternal(uint32_t)': xembed.cc:126: warning: unused variable 'x' TIA. -- Cheers Jorge.-

2 1

x offset > 64k
by corvid＠lavabit.com May 28, 2009

May 28, 2009

I tried looking at a stylesheet, but nothing displayed. It was a one-line document about 7k chars long. It turned out that fltk::getwidth() calls XftTextExtents32() and the width comes back in a short, so it wrapped at 32k. I chopped the line into 1k-long "words", and now I saw text, but once I made a line somewhere over 8k chars long, drawing would wrap and start at the beginning again. Everything looked all right in fltk::drawtext_transformed() as it called XftDrawString32()... I have no idea whether this is something easy to deal with in some way, or a limitation in X...

2 3

patch: a binding for stop
by jcid＠dillo.org May 28, 2009

May 28, 2009

On Wed, May 27, 2009 at 02:22:57PM +0000, corvid wrote: > > Wondering whether we can come to a decision on this one.. > I don't like much Esc for stop, but if that's FFs default, it may be better than another (as the Ctrl-x clash you spot). Afterall this is what configurable keybindings is suppossed to solve so Esc is OK for me as default. > -------------------------------------- > > > Jeremy wrote: > > On Wed, May 06, 2009 at 11:32:31AM -0400, Jorge Arellano Cid wrote: > > > > > > The other one I can think of as mnemonic is <ctrl>x. What do > > > people think? > > > > That clashes with "Cut". Select some text in the URL window and > > <ctrl>+x will delete it. I think it's not a good idea to grab > > something that (a) already has a conventional action, and (b) already > > performs that conventional action in the GUI. > > > > And while we're talking about conventional use: Dillo is highly > > unconventional when it uses <ctrl>+q to close a tab. Everyone else[1] > > uses <ctrl>+w to close a tab (<ctrl>+<shift>+q to close a window) and > > <ctrl>+q to quit. Which can be annoying when switching between the > > two. > > > > Firefox uses Escape to stop, and arora[2] uses <ctrl>+. . > > > > Regards, > > > > Jeremy Henty > > > > [1] ie. Firefox :-) > > [2] http://code.google.com/p/arora/ - WebKit+Qt-based browser. It > > looks very nice and may supplant Firefox as "the browser to fall back > > to when Dillo can't handle the page". > > -- Cheers Jorge.-

1 0

(no subject)
by bogus＠does.not.exist.com May 27, 2009

May 27, 2009

From: Jorge Arellano Cid <jcid(a)dillo.org> To: Dillo mailing list <dillo-dev(a)dillo.org> Cc: Bcc: Subject: Re: [Dillo-dev] patch: a binding for stop Reply-To: In-Reply-To: <20090527142256.GH6001(a)local.gobigwest.com> On Wed, May 27, 2009 at 02:22:57PM +0000, corvid wrote: > > Wondering whether we can come to a decision on this one.. I don't like much Esc for stop, but if that's FFs default, it may be better than another (as the Ctrl-x clash you spot). Afterall this is what configurable keybindings is suppossed to solve so Esc is OK for me as default. > > Jeremy wrote: > > On Wed, May 06, 2009 at 11:32:31AM -0400, Jorge Arellano Cid wrote: > > > > > > The other one I can think of as mnemonic is <ctrl>x. What do > > > people think? > > > > That clashes with "Cut". Select some text in the URL window and > > <ctrl>+x will delete it. I think it's not a good idea to grab > > something that (a) already has a conventional action, and (b) already > > performs that conventional action in the GUI. > > > > And while we're talking about conventional use: Dillo is highly > > unconventional when it uses <ctrl>+q to close a tab. Everyone else[1] > > uses <ctrl>+w to close a tab (<ctrl>+<shift>+q to close a window) and > > <ctrl>+q to quit. Which can be annoying when switching between the > > two. > > > > Firefox uses Escape to stop, and arora[2] uses <ctrl>+. . > > > > Regards, > > > > Jeremy Henty > > > > [1] ie. Firefox :-) > > [2] http://code.google.com/p/arora/ - WebKit+Qt-based browser. It > > looks very nice and may supplant Firefox as "the browser to fall back > > to when Dillo can't handle the page". > > -- Cheers Jorge.-

1 0

include guard naming
by corvid＠lavabit.com May 27, 2009

May 27, 2009

I was just going through and changing the include guards that don't follow the dillo convention, and then I glanced at http://en.wikipedia.org/wiki/Include_guard and it reminds me that the C++ standard prohibits '_'s at the beginning. Does anyone care enough to have an opinion? If we want to rm the leading or surrounding underscores, I might as well do that too before committing. [pop/smtp isn't working at the moment, so I'm sending this through webmail. Let's hope they doesn't mangle line wrapping or anything.]

3 4

[patch]: [doc]: escape some hash in the text
by onepoint＠starurchin.org May 27, 2009

May 27, 2009

When you're sick of bugs, clean up the docs! This escapes some '#' characters so that "#define" and "#include" appear properly in the text. It also silences some Doxygen warnings (it was trying to interpret those words as links). Regards, Jeremy Henty

2 1

Spotted a tiny dillo bug with tabs
by freya128＠yahoo.com May 24, 2009

May 24, 2009

Tiny bug: When dillo changes tabs it doesn't update the main window header thing. Also the same if you close a tab. This can lead to weird effects such as the top header thing having info about a webpage that you don't have open at all. Hope that helps! :) love Freya

2 1

[bug]: uint32_t not declared
by onepoint＠starurchin.org May 24, 2009

May 24, 2009

The latest tip fails to build for me. I don't have uint32_t. Which header should include it? I'm using gcc 4.1.2. Regards, Jeremy Henty

3 5

Latest on invalid image callback references
by onepoint＠starurchin.org May 23, 2009

May 23, 2009

The valgrind logs are still showing invalid memory accesses that happen when image callbacks refer to DilloImages that have been freed by a_Web_free(). This is after the imgref patch was applied, since when I've been running an essentially unpatched Dillo. My image tracing still shows that the image pipeline sometimes continues to run after a_Web_free() frees the image it is attached to. As far as I can tell there has been no recent change in the behaviour of this problem. I have just applied a patch I was considering earlier that makes a_Cache_stop_client() close the client. I've attached it for comments, but I won't suggest applying it until we can see whether it's made any difference. That will take a few days, as I still haven't found a reliable way to reproduce the bug so I will just have to wait a few days and see if it stops appearing in the logs. Please comment on whether this patch is the right way to fix this. Regards, Jeremy Henty PS. You'll probably notice fewer valgrind reports in my log pages in future because I just returned to full-time work so I won't be surfing from home nearly as much.

2 2

[rutkowski.adrian1@gmail.com: Re: [Dillo-dev] Dillo failed to compile with GCC 4.4.0]
by jcid＠dillo.org May 23, 2009

May 23, 2009

On Fri, May 22, 2009 at 06:51:16PM +0000, corvid wrote: > No need. Adrian went through a few rounds of sending me the error > msgs, I committed some consts, and then he reported: > > > Success! :) > > Now it compiles and runs whitout glitches. Excellent! -- Cheers Jorge.-

1 0

Dillo failed to compile with GCC 4.4.0
by rutkowski.adrian1＠gmail.com May 22, 2009

May 22, 2009

Hi, I'm trying to compile latest dillo hg snapshot with GCC 4.4.0 and got this error: g++ -DHAVE_CONFIG_H -I. -I.. -I/usr/local/include -g -O2 -Wall -W -Wno-unused-parameter -fno-rtti -fno-exceptions -MT tablecell.o -MD -MP -MF .deps/tablecell.Tpo -c -o tablecell.o tablecell.cc tablecell.cc: In member function ?virtual void dw::TableCell::wordWrap(int)?: tablecell.cc:55: error: invalid conversion from ?const char*? to ?char*? make[2]: *** [tablecell.o] B??d 1 make[2]: Opuszczenie katalogu `/home/adrian/src/dillo/dw' make[1]: *** [all-recursive] B??d 1 make[1]: Opuszczenie katalogu `/home/adrian/src/dillo' make: *** [all] B??d 2 System is Arch Linux with latest gcc package (gcc-4.4.0-2). Adrian Rutkowski

4 4

Dillo failed to compile with GCC 4.4.0
by rutkowski.adrian1＠gmail.com May 22, 2009

May 22, 2009

On Fri, May 22, 2009 at 05:34:20PM +0000, corvid wrote: > Does this compile cleanly for you: > > diff -r 717e6ced084f src/uicmd.cc > --- a/src/uicmd.cc Fri May 22 17:16:18 2009 +0000 > +++ b/src/uicmd.cc Fri May 22 17:32:05 2009 +0000 > @@ -694,8 +694,8 @@ static char *UIcmd_make_save_filename(co > static char *UIcmd_make_save_filename(const char *pathstr) > { > size_t MaxLen = 64; > - char *FileName, *name, *newname, *o, *n; > - const char *dir = a_UIcmd_get_save_dir(); > + char *FileName, *newname, *o, *n; > + const char *name, *dir = a_UIcmd_get_save_dir(); > > if ((name = strrchr(pathstr, '/'))) { > if (strlen(++name) > MaxLen) { Yes that fixed it, now it stops here: g++ -DHAVE_CONFIG_H -I. -I.. -DDILLORC_SYS='"/usr/local/etc/dillorc"' -I/usr/local/include -I.. -I/usr/include/libpng12 -I/usr/local/include -Wno-non-virtual-dtor -g -O2 -Wall -W -Wno-unused-parameter -fno-rtti -fno-exceptions -MT html.o -MD -MP -MF .deps/html.Tpo -c -o html.o html.cc html.cc: In function ?void Html_tag_open_meta(DilloHtml*, const char*, int)?: html.cc:2812: error: invalid conversion from ?const char*? to ?char*? make[1]: *** [html.o] B??d 1 make[1]: Opuszczenie katalogu `/home/adrian/src/dillo/src' make: *** [all-recursive] B??d 1

1 0

image inputs, height and width attrs
by corvid＠lavabit.com May 19, 2009

May 19, 2009

I noticed that images in image inputs weren't scaling, but then I saw in the spec that INPUT isn't supposed to have height and width attrs anyway, so I don't know whether we should do anything about it.

2 2

patch: MSG* macros (Re: patch: default home dir)
by corvid＠lavabit.com May 18, 2009

May 18, 2009

Jorge wrote: > On Fri, May 15, 2009 at 11:06:33PM +0000, corvid wrote: > > BTW, is there a reason why MSG_HTTP doesn't have an "if (prefs.show_msg)"? > > Not much of a reason. > IIRC sometime ago I regarded those as "more important", or at least > they should be visible before searching bugs from the parser and up. > > > And is there a reason why only MSG gets a fflush? > > Most probably in a debugging session, important messages were > beong buffered and so the flush. > > I'd say yes, let's have the fflush and prefs.show_msg act on all of them. > Patch is welcomed! :) I'm sending this to the list in case I'm doing something clumsy or dangerous. It's working for me, anyway.

1 1

dillo sys config location
by corvid＠lavabit.com May 17, 2009

May 17, 2009

I think someone may have brought this up at some point, but what do we want to do about -DDILLORC_SYS='"/usr/local/etc/dillorc"' (as it happens to be in my case) and #define PATHS_RC_SYS "/etc/dillo/" ?

1 0

[patch] Do not change window title
by Christian.Kellermann＠nefkom.net May 16, 2009

May 16, 2009

Skipped content of type multipart/mixed

3 6

[patch] GCC v4.4
by tim.nieradzik＠gmx.de May 15, 2009

May 15, 2009

Dillo does not compile with GCC v4.4.0. A patch is attached. --Tim

4 6

JavaScript support experiment
by stevegigijoe＠yahoo.com.tw May 15, 2009

May 15, 2009

Hello all After I successful build and running dillo-2. I pay more attention on how JavaScript support on dillo. There are some discussion on mailing list and useful information. http://lists.auriga.wearlab.de/pipermail/dillo-dev/2008-October/005065.html http://lists.auriga.wearlab.de/pipermail/dillo-dev/2009-April/006280.html As dillo is fast and small footprint. Things like JavaScript should not be affect overall performence. What we need ? 1.JavaScript parse engine. I personally prefer SEE (simple ECMAScript Engine). Light weight and seems maintained 2.Document Object Model. We have to all objects and that's relate with engine and browser. An really huge task ... More important how to combine it into dillo ? hard code into or by DPI ? As I'm new to dillo, I don't know you guys interesting about this. So I decide to do an small experiment to implement JavaScript support with libSEE + Boehm-GC. Get patch here http://tw.myblog.yahoo.com/stevegigijoe/article?mid=145&prev=-1&next=138 ****** Modify configure.in to check libsee and gc DOM Supports document.write() document.getElementById() document.forms.length document.[form name].length document.[form name].method document.[form name].reset() document.[form name].[input name].value Input HTML scripts and *.js from url run onClick script before submit ****** Take it as an study case. Any discussion & suggestion is welcome Steve ______________________________________________________________________________________________________ ¥I¶O€~®e¶qµL€W¡HYahoo!©_Œ¯¹q€l«Hœc2.0§K¶Oµ¹§A¡A«H¥ó¥Ã»·€£¥²§R¡I http://tw.mg0.mail.yahoo.com/dc/landing

2 1

change FLTK scheme for in dillo2
by akhilman＠gmail.com May 14, 2009

May 14, 2009

Greetings. Can I change colours of dialogues and scrollbar in dillo2? I would like to gets something less bright. Even something like fluid2's -bg option will be ok. Thank you for great lightweight browser. -- ?????? ??????????? aka AkhIL ??? ??? 7 13:39:04 KRAST 2009 Thu May 7 05:39:04 UTC 2009 ---------------------------------- Visit my home page http://akhilman.blogspot.com/ jabber: akhil(a)jabber.ru ---------------------------------- ??? ????????? ????????? ????????? ??, ??? ?? ??????? ?????????, ?????? ?? ???? ????, ??? ?? ???????????? ??? ?????????. (?.?????????) ---------------------------------- Linux artstation 2.6.28-gentoo-r5 #1 SMP Sat May 2 18:40:24 KRAST 2009 x86_64 AMD Phenom(tm) 9550 Quad-Core Processor AuthenticAMD GNU/Linux up 4 days, 18:33, 4 users, load average: 0.48, 0.40, 0.37

4 13

Suggestion: update the Doxyfile
by onepoint＠starurchin.org May 13, 2009

May 13, 2009

Running doxygen is annoyingly noisy. One reason for that is that the Doxyfile is from an old version of doxygen and contains obsolete stuff that triggers warnings. I ran "doxygen -u" locally and submitted the patch to the list but it broke (just) the 40K limit and stuck in moderation. You can get it here: http://starurchin.org/dillo/patch/update-Doxyfile.txt Or a developer could run "doxygen -u" instead. Anyway, I think upgrading the Doxyfile is a good idea. Thoughts? Regards, Jeremy Henty

2 1

image memory leak
by Johannes.Hofmann＠gmx.de May 13, 2009

May 13, 2009

Hi, I did some more tests regarding that image related memory leak, Jeremy reported some time ago. It turns out that memory related to aborted images is not released. With CSS and the related repushs this seems to happen frequently. But I can also trigger it without CSS by hitting Ctrl-R while loading a large image. Attached diagnostic patch shows the issue. Load a page with a single large JPEG image and hit Ctrl-R while it is loading. In that case only the second instance of DilloJpeg is deleted. I tried to figure out, why the first cache client is not getting aborted properly but the cache related code is quite complex... Does anyone know how this should work? Cheers, Johannes

3 8

dillo man page
by corvid＠lavabit.com May 13, 2009

May 13, 2009

I tried making a little man page for dillo. 1) in order to be a proper, self-respecting unix program :) 2) good way of documenting the files, and I don't think environment variables were documented anywhere. I'd never made a proper man page before, but I think it's pretty close to how they're supposed to be... http://www.dillo.org/test/dillo.1 http://www.dillo.org/test/dillo.man.ascii http://www.dillo.org/test/dillo.man.html

2 1

[patch]: fix image logging messages
by onepoint＠starurchin.org May 11, 2009

May 11, 2009

This patch just tweaks the image debugging messages. It fixes the reported function names (currently a_*_new is reported as *_new) and enforces a consistent format. Regards, Jeremy Henty

2 1

Nav_repush() is hurting valgrind
by onepoint＠starurchin.org May 11, 2009

May 11, 2009

The valgrind logs[1] show a recent burst of errors from calls to Nav_repush(). Also, I'm currently trialling a patch to fix the image access errors from a_Cache_stop_client() [2]. I don't think it can be the cause of the Nav_repush() problems but I'd like a second opinion (I don't yet fully understand how this code should work). Regards, Jeremy Henty [1] http://starurchin.singularity/dillo/valgrind.html [2] http://starurchin.singularity/dillo/patch/make-a_Cache_stop_client-close-cl…

2 12