October 2008 - Dillo-dev - lists.mailman3.com

what crashes remain?

by jcid＠dillo.org

On Wed, Oct 01, 2008 at 02:48:10PM +0000, corvid wrote: > Jorge wrote: > > On Sat, Sep 27, 2008 at 05:38:03PM -0400, Jorge Arellano Cid wrote: > > > On Sat, Sep 27, 2008 at 05:29:28PM +0000, corvid wrote: > > > > It's still possible to crash dillo if you click on a remote link and > > > > quickly click on some images to load them. Doesn't crash every time, > > > > but typically within four tries or so. I had assumed it was the same > > > > thing as whatever was wrong with the menu crash, so I'd ignored it. > > > > > > > > If I stick > > > > a_Bw_stop_clients(bw, BW_Img); > > > > in the DilloHtml destructor, I can no longer get it to crash, but > > > > I randomly got a "Cache_process_queue Caught busy!!!" while browsing > > > > earlier today, so there's something wrong with just doing that. > > > > (Assuming it's related; I'm not in the habit of getting that MSG.) > > > > > > That message means big trouble: the event loop stepped over itself... > > > > > > I've tried several different ways but can't get to crash it. > > > > > > Does the attached patch help? > > > > > > -- > > > Cheers > > > Jorge.- > > > > > diff -pru dillo2/src/html.cc dillo2-cur/src/html.cc > > > --- dillo2/src/html.cc 2008-09-27 15:11:08.000000000 -0400 > > > +++ dillo2-cur/src/html.cc 2008-09-27 17:34:21.000000000 -0400 > > > @@ -711,6 +711,8 @@ bool_t DilloHtml::unloadedImages() > > > */ > > > void DilloHtml::loadImages (const DilloUrl *pattern) > > > { > > > + dReturn_if_fail (bw->nav_expecting == FALSE); > > > + > > > for (int i = 0; i < images->size(); i++) { > > > if (images->get(i)->image) { > > > if ((!pattern) || (!a_Url_cmp(images->get(i)->url, pattern))) { > > > > Can you still crash dillo this way, with the latest CVS? > > (i.e. without the above patch). > > Yes. > > Also, I found out why loading was sometimes randomly disabled (as mentioned > yesterday). You know how leaving the window sometimes causes an extra > Nav_open_url()? No, I don't know. How is it triggered? -- Cheers Jorge.-

Oct. 1, 2008

[patch] share button images between UIs

by jcid＠dillo.org

On Wed, Oct 01, 2008 at 02:58:44PM -0400, Jorge Arellano Cid wrote: > On Wed, Oct 01, 2008 at 07:36:54PM +0200, Johannes Hofmann wrote: > > Hi, > > > > this is independent from the recent crashes. > > This patch shares the images between multiple UIs. > > > > What do you think? > > Committed! > > Except that "img on/off" doesn't switch the icon. > Please send me a patch for this. Don't worry, it's fixed in CVS now. -- Cheers Jorge.-

Oct. 1, 2008

what crashes remain?

by jcid＠dillo.org

On Wed, Oct 01, 2008 at 06:14:47PM +0200, Johannes Hofmann wrote: > On Wed, Oct 01, 2008 at 11:45:53AM -0400, Jorge Arellano Cid wrote: > > On Wed, Oct 01, 2008 at 04:47:05PM +0200, Johannes Hofmann wrote: > > > Hi, > > > > > > On Wed, Oct 01, 2008 at 10:33:20AM -0400, Jorge Arellano Cid wrote: > > > > Hi, > > > > > > > > On Sat, Sep 27, 2008 at 05:38:03PM -0400, Jorge Arellano Cid wrote: > > > > > On Sat, Sep 27, 2008 at 05:29:28PM +0000, corvid wrote: > > > > > > It's still possible to crash dillo if you click on a remote link and > > > > > > quickly click on some images to load them. Doesn't crash every time, > > > > > > but typically within four tries or so. I had assumed it was the same > > > > > > thing as whatever was wrong with the menu crash, so I'd ignored it. > > > > > > > > > > > > If I stick > > > > > > a_Bw_stop_clients(bw, BW_Img); > > > > > > in the DilloHtml destructor, I can no longer get it to crash, but > > > > > > I randomly got a "Cache_process_queue Caught busy!!!" while browsing > > > > > > earlier today, so there's something wrong with just doing that. > > > > > > (Assuming it's related; I'm not in the habit of getting that MSG.) > > > > > > > > > > That message means big trouble: the event loop stepped over itself... > > > > > > > > > > I've tried several different ways but can't get to crash it. > > > > > > > > > > Does the attached patch help? > > > > > > > > > > -- > > > > > Cheers > > > > > Jorge.- > > > > > > > > > diff -pru dillo2/src/html.cc dillo2-cur/src/html.cc > > > > > --- dillo2/src/html.cc 2008-09-27 15:11:08.000000000 -0400 > > > > > +++ dillo2-cur/src/html.cc 2008-09-27 17:34:21.000000000 -0400 > > > > > @@ -711,6 +711,8 @@ bool_t DilloHtml::unloadedImages() > > > > > */ > > > > > void DilloHtml::loadImages (const DilloUrl *pattern) > > > > > { > > > > > + dReturn_if_fail (bw->nav_expecting == FALSE); > > > > > + > > > > > for (int i = 0; i < images->size(); i++) { > > > > > if (images->get(i)->image) { > > > > > if ((!pattern) || (!a_Url_cmp(images->get(i)->url, pattern))) { > > > > > > > > Can you still crash dillo this way, with the latest CVS? > > > > (i.e. without the above patch). > > > > > > I'm currently seeing all sorts of crashes and I suspect that they > > > are related to xpmImage from fltk. > > > Could someone with a linux machine please check current dillo with > > > valgrind especially when the new mini_bug_xpm is shown? > > > > > > Also changing mini_bug_xpm to 15x15 seems to fix things for me. > > > I don't understand all that completely, but other crashes might be > > > related to this. > > > > OK, I'll check with valgrind... > > Does changing mini_ok_xpm to 16x16 fix it too? > > No it doesn't, that's what I thought first too :-) > > But it really only happens when fltk was compiled with > --disable-xft I did that and FLTK's tests don't use xft, but my dillo-fltk still uses them! BTW, does a: - new_w = strlen(str)*8 + 20; + new_w = strlen(str)*8 + 40; in ui.cc solve the problem for you? If true, I can make a patch using measure(). -- Cheers Jorge.-

Oct. 1, 2008

[patch] share button images between UIs

by jcid＠dillo.org

On Wed, Oct 01, 2008 at 07:36:54PM +0200, Johannes Hofmann wrote: > Hi, > > this is independent from the recent crashes. > This patch shares the images between multiple UIs. > > What do you think? Committed! Except that "img on/off" doesn't switch the icon. Please send me a patch for this. -- Cheers Jorge.-

Oct. 1, 2008

[patch] share button images between UIs

by corvid＠lavabit.com

Johannes wrote: > this is independent from the recent crashes. > This patch shares the images between multiple UIs. > > What do you think? Good!

Oct. 1, 2008

[patch] share button images between UIs

by Johannes.Hofmann＠gmx.de

Hi, this is independent from the recent crashes. This patch shares the images between multiple UIs. What do you think? Cheers, Johannes

Oct. 1, 2008

Basic authentication

by Christian.Kellermann＠nefkom.net

Dear List, since I have not found any traces of discussion on this list about this topic I dare to ask where to place code handling basic authentication in dillo's source? As I see it the cache module is responsible for handling server responses. So I would at a first try add the code there. How can I connect to a UI dialog from here to get the necessary credentials? Maybe a possible solution looks like this: - react to 403 header and look for Authorization request - if there is one ask the user for username/pass + resend the request with the Authorisation header added + or display the 403 if the user aborts What do you think? I will come up with some example code to illustrate the idea in case this is not very clear. Kind regards, Christian -- You may use my gpg key for replies: pub 1024D/47F79788 2005/02/02 Christian Kellermann (C-Keen)

Oct. 1, 2008

what crashes remain?

by Johannes.Hofmann＠gmx.de

On Wed, Oct 01, 2008 at 11:45:53AM -0400, Jorge Arellano Cid wrote: > On Wed, Oct 01, 2008 at 04:47:05PM +0200, Johannes Hofmann wrote: > > Hi, > > > > On Wed, Oct 01, 2008 at 10:33:20AM -0400, Jorge Arellano Cid wrote: > > > Hi, > > > > > > On Sat, Sep 27, 2008 at 05:38:03PM -0400, Jorge Arellano Cid wrote: > > > > On Sat, Sep 27, 2008 at 05:29:28PM +0000, corvid wrote: > > > > > It's still possible to crash dillo if you click on a remote link and > > > > > quickly click on some images to load them. Doesn't crash every time, > > > > > but typically within four tries or so. I had assumed it was the same > > > > > thing as whatever was wrong with the menu crash, so I'd ignored it. > > > > > > > > > > If I stick > > > > > a_Bw_stop_clients(bw, BW_Img); > > > > > in the DilloHtml destructor, I can no longer get it to crash, but > > > > > I randomly got a "Cache_process_queue Caught busy!!!" while browsing > > > > > earlier today, so there's something wrong with just doing that. > > > > > (Assuming it's related; I'm not in the habit of getting that MSG.) > > > > > > > > That message means big trouble: the event loop stepped over itself... > > > > > > > > I've tried several different ways but can't get to crash it. > > > > > > > > Does the attached patch help? > > > > > > > > -- > > > > Cheers > > > > Jorge.- > > > > > > > diff -pru dillo2/src/html.cc dillo2-cur/src/html.cc > > > > --- dillo2/src/html.cc 2008-09-27 15:11:08.000000000 -0400 > > > > +++ dillo2-cur/src/html.cc 2008-09-27 17:34:21.000000000 -0400 > > > > @@ -711,6 +711,8 @@ bool_t DilloHtml::unloadedImages() > > > > */ > > > > void DilloHtml::loadImages (const DilloUrl *pattern) > > > > { > > > > + dReturn_if_fail (bw->nav_expecting == FALSE); > > > > + > > > > for (int i = 0; i < images->size(); i++) { > > > > if (images->get(i)->image) { > > > > if ((!pattern) || (!a_Url_cmp(images->get(i)->url, pattern))) { > > > > > > Can you still crash dillo this way, with the latest CVS? > > > (i.e. without the above patch). > > > > I'm currently seeing all sorts of crashes and I suspect that they > > are related to xpmImage from fltk. > > Could someone with a linux machine please check current dillo with > > valgrind especially when the new mini_bug_xpm is shown? > > > > Also changing mini_bug_xpm to 15x15 seems to fix things for me. > > I don't understand all that completely, but other crashes might be > > related to this. > > OK, I'll check with valgrind... > Does changing mini_ok_xpm to 16x16 fix it too? No it doesn't, that's what I thought first too :-) But it really only happens when fltk was compiled with --disable-xft Cheers, Johannes

Oct. 1, 2008

what crashes remain?

by jcid＠dillo.org

On Wed, Oct 01, 2008 at 04:47:05PM +0200, Johannes Hofmann wrote: > Hi, > > On Wed, Oct 01, 2008 at 10:33:20AM -0400, Jorge Arellano Cid wrote: > > Hi, > > > > On Sat, Sep 27, 2008 at 05:38:03PM -0400, Jorge Arellano Cid wrote: > > > On Sat, Sep 27, 2008 at 05:29:28PM +0000, corvid wrote: > > > > It's still possible to crash dillo if you click on a remote link and > > > > quickly click on some images to load them. Doesn't crash every time, > > > > but typically within four tries or so. I had assumed it was the same > > > > thing as whatever was wrong with the menu crash, so I'd ignored it. > > > > > > > > If I stick > > > > a_Bw_stop_clients(bw, BW_Img); > > > > in the DilloHtml destructor, I can no longer get it to crash, but > > > > I randomly got a "Cache_process_queue Caught busy!!!" while browsing > > > > earlier today, so there's something wrong with just doing that. > > > > (Assuming it's related; I'm not in the habit of getting that MSG.) > > > > > > That message means big trouble: the event loop stepped over itself... > > > > > > I've tried several different ways but can't get to crash it. > > > > > > Does the attached patch help? > > > > > > -- > > > Cheers > > > Jorge.- > > > > > diff -pru dillo2/src/html.cc dillo2-cur/src/html.cc > > > --- dillo2/src/html.cc 2008-09-27 15:11:08.000000000 -0400 > > > +++ dillo2-cur/src/html.cc 2008-09-27 17:34:21.000000000 -0400 > > > @@ -711,6 +711,8 @@ bool_t DilloHtml::unloadedImages() > > > */ > > > void DilloHtml::loadImages (const DilloUrl *pattern) > > > { > > > + dReturn_if_fail (bw->nav_expecting == FALSE); > > > + > > > for (int i = 0; i < images->size(); i++) { > > > if (images->get(i)->image) { > > > if ((!pattern) || (!a_Url_cmp(images->get(i)->url, pattern))) { > > > > Can you still crash dillo this way, with the latest CVS? > > (i.e. without the above patch). > > I'm currently seeing all sorts of crashes and I suspect that they > are related to xpmImage from fltk. > Could someone with a linux machine please check current dillo with > valgrind especially when the new mini_bug_xpm is shown? > > Also changing mini_bug_xpm to 15x15 seems to fix things for me. > I don't understand all that completely, but other crashes might be > related to this. OK, I'll check with valgrind... Does changing mini_ok_xpm to 16x16 fix it too? -- Cheers Jorge.-

Oct. 1, 2008

what crashes remain?

by Johannes.Hofmann＠gmx.de

On Wed, Oct 01, 2008 at 04:47:05PM +0200, Johannes Hofmann wrote: > Hi, > > On Wed, Oct 01, 2008 at 10:33:20AM -0400, Jorge Arellano Cid wrote: > > Hi, > > > > On Sat, Sep 27, 2008 at 05:38:03PM -0400, Jorge Arellano Cid wrote: > > > On Sat, Sep 27, 2008 at 05:29:28PM +0000, corvid wrote: > > > > It's still possible to crash dillo if you click on a remote link and > > > > quickly click on some images to load them. Doesn't crash every time, > > > > but typically within four tries or so. I had assumed it was the same > > > > thing as whatever was wrong with the menu crash, so I'd ignored it. > > > > > > > > If I stick > > > > a_Bw_stop_clients(bw, BW_Img); > > > > in the DilloHtml destructor, I can no longer get it to crash, but > > > > I randomly got a "Cache_process_queue Caught busy!!!" while browsing > > > > earlier today, so there's something wrong with just doing that. > > > > (Assuming it's related; I'm not in the habit of getting that MSG.) > > > > > > That message means big trouble: the event loop stepped over itself... > > > > > > I've tried several different ways but can't get to crash it. > > > > > > Does the attached patch help? > > > > > > -- > > > Cheers > > > Jorge.- > > > > > diff -pru dillo2/src/html.cc dillo2-cur/src/html.cc > > > --- dillo2/src/html.cc 2008-09-27 15:11:08.000000000 -0400 > > > +++ dillo2-cur/src/html.cc 2008-09-27 17:34:21.000000000 -0400 > > > @@ -711,6 +711,8 @@ bool_t DilloHtml::unloadedImages() > > > */ > > > void DilloHtml::loadImages (const DilloUrl *pattern) > > > { > > > + dReturn_if_fail (bw->nav_expecting == FALSE); > > > + > > > for (int i = 0; i < images->size(); i++) { > > > if (images->get(i)->image) { > > > if ((!pattern) || (!a_Url_cmp(images->get(i)->url, pattern))) { > > > > Can you still crash dillo this way, with the latest CVS? > > (i.e. without the above patch). > > I'm currently seeing all sorts of crashes and I suspect that they > are related to xpmImage from fltk. > Could someone with a linux machine please check current dillo with > valgrind especially when the new mini_bug_xpm is shown? Update: This only seems to happen when fltk was compiled with the --disable-xft switch. Cheers, Johannes

Oct. 1, 2008