- Dillo-dev - lists.mailman3.com

how about if we make a 3.0.5

by eocene＠gmx.com

I wrote: > noname wrote: > > SSL3 and compression are not the main issue. > > HTTPS in dillo is completely broken because it does not check for domain > > name in the certificate. > > hg tip has checking code copied from wget and current dillo release has > > no code for it at all. It means that Dillo accepts any valid certificate > > as a certificate for, let's say, gmail. You can get one from StartSSL > > for free and test, it works. > > Right, I hadn't wanted to do any real New Work for 3.0.5 that would require > a somewhat higher level of scrutiny and testing, but all right, I'll take > a look at gluing that stuff into the https dpi. Adapted the name checking to fit into the dpi, plus server name indication while I was at it. Looks like it may be working. I'll push the code to the server later if you can promise to help give it heavy testing in coming days. When you're browsing, if you just change http to https, so many sites that don't expect TLS connections will offer up certificates that are broken in some way. And then try the various cancel/continue combinations...

June 12, 2015

how about if we make a 3.0.5

by eocene＠gmx.com

noname wrote: > SSL3 and compression are not the main issue. > HTTPS in dillo is completely broken because it does not check for domain > name in the certificate. > hg tip has checking code copied from wget and current dillo release has > no code for it at all. It means that Dillo accepts any valid certificate > as a certificate for, let's say, gmail. You can get one from StartSSL > for free and test, it works. Right, I hadn't wanted to do any real New Work for 3.0.5 that would require a somewhat higher level of scrutiny and testing, but all right, I'll take a look at gluing that stuff into the https dpi.

June 12, 2015

how about if we make a 3.0.5

by noname＠inventati.org

On Tue, Jun 09, 2015 at 02:50:30PM +0000, eocene wrote: > > It would be nice to put out some new stuff. > > Primary motivations: > > 1. placeholder attributes for textarea/input are everywhere, and often > they have no other explanation of their purpose. Having to guess > "this is probably for search, this is probably for username, this > is probably for password" is no fun. > 2. I know there are users just trusting dillo's https, so at the > least it would be good for their https dpis to have a better > cipher list and to disable SSL3 and compression. SSL3 and compression are not the main issue. HTTPS in dillo is completely broken because it does not check for domain name in the certificate. hg tip has checking code copied from wget and current dillo release has no code for it at all. It means that Dillo accepts any valid certificate as a certificate for, let's say, gmail. You can get one from StartSSL for free and test, it works.

June 12, 2015

[jcid@dillo.org: Re: how about if we make a 3.0.5]

by eocene＠gmx.com

Jorge wrote: > Good! This list looks quite safe to me. > > Feel free to ask if you find more patches thay may be included. Just now I was adding in: 1. Image buffer/cache improvements. 2. that which I found myself missing during use yesterday: a. Better notification when user's domainrc settings block page redirection. b. a bunch of assorted BUG_MSG/MSG changes made over the course of the past year. Looking at page bugs now for pages, I find myself sort of wanting my character reference changes, but...I have to draw the line somewhere. Although we have been running on it for a year... I don't know. Thoughts: - I guess it's a good sign that we're making some sort of progress if I miss having various changes. - My perception while using this prospective 3.0.5 yesterday was that it feels good, it feels speedy. Oh, one more thing: Does "Add break option after <img>" have something to do with floats, or can/should it go in as well?

June 12, 2015

[jcid@dillo.org: Re: how about if we make a 3.0.5]

by jcid＠dillo.org

On Thu, Jun 11, 2015 at 10:02:37PM +0000, eocene wrote: > Jorge wrote: > > On Thu, Jun 11, 2015 at 03:46:57PM +0000, eocene wrote: > > > I wrote: > > > > > > > > It would be nice to put out some new stuff. > > > > > > [snip] > > > > > > Been experimenting with a 3.0.5 branch (hg graft). > > > > > > Jorge: Do you think it would be a good idea to pull in your image changes > > > as well? We've been running on them for a long time and I imagine they > > > wouldn't be entangled to any great degree with anything else. > > > > Certainly. > > > > FWIW, I'm not comfortable with pushing a release yet because lots of sites > > render "better" (more readable) with the old version (be it because of float > > positioning or CSS colors). > > > > It's also true that a lot of time has passed since our last release, so > > if you can get some win-win features of the new branch and merge them into > > a better-than-the-former release, that we can confidently handle to our > > users, it may be worth the effort. > > Yeah, I'm staying away from float, grows, http, and tls for this. > I started with the last changeset in the 3.0.4.1 branch and added the > following: > > user: corvid > date: Thu Jun 11 15:20:02 2015 +0000 > summary: now give it a version number > > user: Sebastian Geerken > date: Sat Jul 12 15:27:37 2014 +0200 > summary: Tests are now linked to libX11 when needed. > > user: corvid > date: Wed Jul 23 19:02:56 2014 +0000 > summary: clarify situation with MENU element > > user: corvid > date: Wed Oct 08 15:27:10 2014 +0000 > summary: more compact MSG > > user: corvid > date: Wed Oct 22 01:29:46 2014 +0000 > summary: trim the publicsuffix TLDs yet again > > user: corvid > date: Fri Jul 18 19:34:35 2014 +0000 > summary: TODO > > user: corvid > date: Thu Jun 11 02:20:55 2015 +0000 > summary: 3.0.5 splash > > user: corvid > date: Wed Jun 10 22:22:11 2015 +0000 > summary: ChangeLog > > user: corvid > date: Wed Jun 10 22:08:12 2015 +0000 > summary: https rm RC4 from cipher list > > user: corvid > date: Sun May 31 02:17:35 2015 +0000 > summary: rm MSG > > user: corvid > date: Thu May 28 21:15:21 2015 +0000 > summary: libpng 1.6 series works for me > > user: Jorge Arellano Cid > date: Wed May 27 11:07:04 2015 -0300 > summary: Made view-source dpi use CSS formatting (it's shorter and cleaner) > > user: Jorge Arellano Cid > date: Tue May 26 11:29:21 2015 -0300 > summary: Fix view-source dpi to handle null characters correctly > > user: corvid > date: Sat May 16 17:28:24 2015 +0000 > summary: dillorc: search urls duckduckgo and startpage always redirect to https, so just specify https > > user: corvid > date: Sat May 09 15:24:53 2015 +0000 > summary: splash: urge users to read help. mention domainrc. > > user: corvid > date: Sat May 09 07:10:21 2015 +0000 > summary: keys: add Menu key > > user: corvid > date: Sat May 09 02:54:15 2015 +0000 > summary: fix font-size:(larger|smaller) > > user: corvid > date: Sun Apr 12 17:35:43 2015 +0000 > summary: more html5 doctype strings > > user: corvid > date: Fri Apr 03 19:24:10 2015 +0000 > summary: html5 (and css3) permit gr[ae]y in color names > > user: Johannes Hofmann > date: Sun Mar 08 20:05:21 2015 +0100 > summary: fix leak introduced in bbd25bf5 > > user: Johannes Hofmann > date: Fri Jul 25 22:48:32 2014 +0200 > summary: avoid requesting background images if an ancestor has display:none > > user: Johannes Hofmann > date: Tue Jan 06 21:29:27 2015 +0100 > summary: disable drag and drop > > user: Johannes Hofmann > date: Sun Feb 15 19:27:04 2015 +0100 > summary: use new count field to override default search url > > user: Johannes Hofmann > date: Sun Feb 15 19:20:21 2015 +0100 > summary: add count field in PrefsParser SymNode_t > > user: Johannes Hofmann > date: Sun Feb 15 19:17:51 2015 +0100 > summary: use single symbols table in PrefsParser::parse() > > user: corvid > date: Tue Feb 03 01:06:26 2015 +0000 > summary: for input placeholder, set position to beginning in case it's too long to show all at once > > user: corvid > date: Mon Feb 02 22:23:45 2015 +0000 > summary: TEXTAREA placeholder attribute > > user: corvid > date: Sat Jan 24 00:21:18 2015 +0000 > summary: input init usual_color > > user: Jorge Arellano Cid > date: Thu Jan 22 12:59:41 2015 -0300 > summary: Fix for segfault when there's no dpid and view source is requested. > > user: corvid > date: Thu Jan 22 08:51:02 2015 +0000 > summary: INPUT placeholder attribute > > user: corvid > date: Fri Jan 16 02:59:01 2015 +0000 > summary: don't inheritBackgroundColor for html5 sectioning stuff > > user: corvid > date: Fri Oct 17 01:36:00 2014 +0000 > summary: everyone's finally disabling SSL3; let's do so too > > user: corvid > date: Wed Sep 24 23:11:07 2014 +0000 > summary: handle irix's version of vsnprintf() > > user: corvid > date: Tue Sep 16 00:21:49 2014 +0000 > summary: url positioning in location bar > > user: corvid > date: Wed Sep 03 22:17:07 2014 +0000 > summary: a couple of quick https improvements > > user: Johannes Hofmann > date: Thu Aug 14 19:45:53 2014 +0200 > summary: improve lang and xml:lang attribute parsing > > user: corvid > date: Tue Aug 05 20:50:31 2014 +0000 > summary: html5 doctype, whitespace permitted between "html" and '>' > > user: corvid > date: Sat Aug 02 05:14:24 2014 +0000 > summary: fix the icon name fix > > user: corvid > date: Tue Jul 29 21:54:09 2014 +0000 > summary: fix icon name > > user: corvid > date: Fri Jul 18 18:53:28 2014 +0000 > summary: dillorc consistency > > user: corvid > date: Sun Jul 13 15:54:20 2014 +0000 > summary: let's not mention ipaq anymore > > user: Johannes Hofmann > date: Tue Jun 03 21:15:10 2014 +0200 > summary: change fltk-config test to be more crosscompile/buildroot friendly > > user: corvid > date: Tue Apr 15 21:49:30 2014 +0000 > summary: startpage in search urls > > parent: 3997:04d67ca748ba > user: corvid > date: Thu Apr 10 00:23:00 2014 +0000 > summary: user_help fix character reference Good! This list looks quite safe to me. Feel free to ask if you find more patches thay may be included. -- Cheers Jorge.-

June 12, 2015

[jcid@dillo.org: Re: how about if we make a 3.0.5]

by eocene＠gmx.com

Jorge wrote: > On Thu, Jun 11, 2015 at 03:46:57PM +0000, eocene wrote: > > I wrote: > > > > > > It would be nice to put out some new stuff. > > > > [snip] > > > > Been experimenting with a 3.0.5 branch (hg graft). > > > > Jorge: Do you think it would be a good idea to pull in your image changes > > as well? We've been running on them for a long time and I imagine they > > wouldn't be entangled to any great degree with anything else. > > Certainly. > > FWIW, I'm not comfortable with pushing a release yet because lots of sites > render "better" (more readable) with the old version (be it because of float > positioning or CSS colors). > > It's also true that a lot of time has passed since our last release, so > if you can get some win-win features of the new branch and merge them into > a better-than-the-former release, that we can confidently handle to our > users, it may be worth the effort. Yeah, I'm staying away from float, grows, http, and tls for this. I started with the last changeset in the 3.0.4.1 branch and added the following: user: corvid date: Thu Jun 11 15:20:02 2015 +0000 summary: now give it a version number user: Sebastian Geerken date: Sat Jul 12 15:27:37 2014 +0200 summary: Tests are now linked to libX11 when needed. user: corvid date: Wed Jul 23 19:02:56 2014 +0000 summary: clarify situation with MENU element user: corvid date: Wed Oct 08 15:27:10 2014 +0000 summary: more compact MSG user: corvid date: Wed Oct 22 01:29:46 2014 +0000 summary: trim the publicsuffix TLDs yet again user: corvid date: Fri Jul 18 19:34:35 2014 +0000 summary: TODO user: corvid date: Thu Jun 11 02:20:55 2015 +0000 summary: 3.0.5 splash user: corvid date: Wed Jun 10 22:22:11 2015 +0000 summary: ChangeLog user: corvid date: Wed Jun 10 22:08:12 2015 +0000 summary: https rm RC4 from cipher list user: corvid date: Sun May 31 02:17:35 2015 +0000 summary: rm MSG user: corvid date: Thu May 28 21:15:21 2015 +0000 summary: libpng 1.6 series works for me user: Jorge Arellano Cid date: Wed May 27 11:07:04 2015 -0300 summary: Made view-source dpi use CSS formatting (it's shorter and cleaner) user: Jorge Arellano Cid date: Tue May 26 11:29:21 2015 -0300 summary: Fix view-source dpi to handle null characters correctly user: corvid date: Sat May 16 17:28:24 2015 +0000 summary: dillorc: search urls duckduckgo and startpage always redirect to https, so just specify https user: corvid date: Sat May 09 15:24:53 2015 +0000 summary: splash: urge users to read help. mention domainrc. user: corvid date: Sat May 09 07:10:21 2015 +0000 summary: keys: add Menu key user: corvid date: Sat May 09 02:54:15 2015 +0000 summary: fix font-size:(larger|smaller) user: corvid date: Sun Apr 12 17:35:43 2015 +0000 summary: more html5 doctype strings user: corvid date: Fri Apr 03 19:24:10 2015 +0000 summary: html5 (and css3) permit gr[ae]y in color names user: Johannes Hofmann date: Sun Mar 08 20:05:21 2015 +0100 summary: fix leak introduced in bbd25bf5 user: Johannes Hofmann date: Fri Jul 25 22:48:32 2014 +0200 summary: avoid requesting background images if an ancestor has display:none user: Johannes Hofmann date: Tue Jan 06 21:29:27 2015 +0100 summary: disable drag and drop user: Johannes Hofmann date: Sun Feb 15 19:27:04 2015 +0100 summary: use new count field to override default search url user: Johannes Hofmann date: Sun Feb 15 19:20:21 2015 +0100 summary: add count field in PrefsParser SymNode_t user: Johannes Hofmann date: Sun Feb 15 19:17:51 2015 +0100 summary: use single symbols table in PrefsParser::parse() user: corvid date: Tue Feb 03 01:06:26 2015 +0000 summary: for input placeholder, set position to beginning in case it's too long to show all at once user: corvid date: Mon Feb 02 22:23:45 2015 +0000 summary: TEXTAREA placeholder attribute user: corvid date: Sat Jan 24 00:21:18 2015 +0000 summary: input init usual_color user: Jorge Arellano Cid date: Thu Jan 22 12:59:41 2015 -0300 summary: Fix for segfault when there's no dpid and view source is requested. user: corvid date: Thu Jan 22 08:51:02 2015 +0000 summary: INPUT placeholder attribute user: corvid date: Fri Jan 16 02:59:01 2015 +0000 summary: don't inheritBackgroundColor for html5 sectioning stuff user: corvid date: Fri Oct 17 01:36:00 2014 +0000 summary: everyone's finally disabling SSL3; let's do so too user: corvid date: Wed Sep 24 23:11:07 2014 +0000 summary: handle irix's version of vsnprintf() user: corvid date: Tue Sep 16 00:21:49 2014 +0000 summary: url positioning in location bar user: corvid date: Wed Sep 03 22:17:07 2014 +0000 summary: a couple of quick https improvements user: Johannes Hofmann date: Thu Aug 14 19:45:53 2014 +0200 summary: improve lang and xml:lang attribute parsing user: corvid date: Tue Aug 05 20:50:31 2014 +0000 summary: html5 doctype, whitespace permitted between "html" and '>' user: corvid date: Sat Aug 02 05:14:24 2014 +0000 summary: fix the icon name fix user: corvid date: Tue Jul 29 21:54:09 2014 +0000 summary: fix icon name user: corvid date: Fri Jul 18 18:53:28 2014 +0000 summary: dillorc consistency user: corvid date: Sun Jul 13 15:54:20 2014 +0000 summary: let's not mention ipaq anymore user: Johannes Hofmann date: Tue Jun 03 21:15:10 2014 +0200 summary: change fltk-config test to be more crosscompile/buildroot friendly user: corvid date: Tue Apr 15 21:49:30 2014 +0000 summary: startpage in search urls parent: 3997:04d67ca748ba user: corvid date: Thu Apr 10 00:23:00 2014 +0000 summary: user_help fix character reference

June 11, 2015

[jcid@dillo.org: Re: how about if we make a 3.0.5]

by jcid＠dillo.org

Ooops, I mistakenly replied private... Here's the FWD. -- Cheers Jorge.-

June 11, 2015

how about if we make a 3.0.5

by eocene＠gmx.com

I wrote: > > It would be nice to put out some new stuff. [snip] Been experimenting with a 3.0.5 branch (hg graft). Jorge: Do you think it would be a good idea to pull in your image changes as well? We've been running on them for a long time and I imagine they wouldn't be entangled to any great degree with anything else.

June 11, 2015

how about if we make a 3.0.5

by eocene＠gmx.com

It would be nice to put out some new stuff. Primary motivations: 1. placeholder attributes for textarea/input are everywhere, and often they have no other explanation of their purpose. Having to guess "this is probably for search, this is probably for username, this is probably for password" is no fun. 2. I know there are users just trusting dillo's https, so at the least it would be good for their https dpis to have a better cipher list and to disable SSL3 and compression. And then pick out some small, safe fixes. Just to look at ChangeLog, some candidates might be: - Fix crash that's possible searching for text while page still being built. - Fix for segfault when there's no dpid and view source is requested. - Fix view-source dpi to handle null characters correctly. - Made view-source dpi use CSS formatting (it's shorter and cleaner). - Crosscompile/buildroot-friendly fltk-config test. - Fix X11 icon name. - In location bar, tend toward showing beginning of URL instead of end. - Handle irix's version of vsnprintf(). - Fix bug with font_factor preference and CSS font-size:(larger|smaller). - Recognize Menu key in keysrc. - Avoid requesting background images if an ancestor has display:none. - Ignore built-in search url if any are specified in dillorc.

June 9, 2015

Google broke recently with 3.0.4 causing repush loop

by james.from.wellington＠gmail.com

Some pages which show lots of 'redirected 0 times', show less when I block *.gravatar.com in domainrc. On 09/05/2015, James C <james.from.wellington at gmail.com> wrote: > All of this is with the cookiesrc below, which lets google have > cookies on two domains. > > If I: > - log out of google > -- it stalls on a meta refresh tag > - select home, which for me is https://mail.google.com/?ui=html > - provide a username and password > -- I am eventually shown my email messages > then the code below redirects: > - once during the logout > - twice before asking for a password > - eight (8) times before showing me my mail > > Adding back the hack to share cookies probably makes no difference to > these figures, but I'm not sure, because I think that I remember > inconsistently getting seven and nine redirects, so I think that > google is A/B testing me. > > This code cannot be put into production, because it produces messages > about redirecting 0 times reasonably often, for reasons that I haven't > tracked down yet. > > -------------------------------- > #host action > DEFAULT DENY > .google.com ACCEPT_SESSION > .google.co.nz ACCEPT_SESSION > -------------------------------- > > render/dillo-jfw01$ hg diff > diff -r 07c0c28f8960 src/cache.c > --- a/src/cache.c Fri Mar 06 13:12:11 2015 +1300 > +++ b/src/cache.c Sat May 09 22:03:31 2015 +1200 > @@ -1007,11 +1007,24 @@ > return 0; > } > > - /* if there's a redirect loop, stop now */ > - if (bw->redirect_level >= 5) > - entry->Flags |= CA_RedirectLoop; > +// if (0 == bw->redirect_level % 5) { > + if (0 == bw->redirect_level % 1) { > + char *redirMessage; > + int choice; > + > + asprintf(&redirMessage, "Dillo: redirected %d times", > bw->redirect_level); > + > + choice = a_Dialog_choice(redirMessage, URL_STR(entry->Url), > + "Stop", "Continue", NULL); > + > + if (1 == choice) > + entry->Flags |= CA_RedirectLoop; > + > + free(redirMessage); > + } > > if (entry->Flags & CA_RedirectLoop) { > + MSG("Redirect loop for: %s\n", URL_STR_(entry->Url)); > a_UIcmd_set_msg(bw, "ERROR: redirect loop for: %s", > URL_STR_(entry->Url)); > bw->redirect_level = 0; > return 0; > diff -r 07c0c28f8960 src/cookies.c > --- a/src/cookies.c Fri Mar 06 13:12:11 2015 +1300 > +++ b/src/cookies.c Sat May 09 22:03:31 2015 +1200 > @@ -196,6 +196,15 @@ > > if (requester == NULL) { > /* request made by user */ > +#if 0 > + } else if (!strcmp("accounts.google.co.nz", URL_HOST(requester)) && > + !strcmp("mail.google.com", URL_HOST(query_url))) { > + MSG("Cookies: sent for google NZ special case 1.\n"); > + } else if (!strcmp("accounts.google.com", URL_HOST(requester)) && > + !strcmp("accounts.google.co.nz", URL_HOST(query_url))) { > + MSG("Cookies: sent for google NZ special case 2.\n"); > + > +#endif > } else if (!a_Url_same_organization(query_url, requester)) { > MSG("Cookies: not sent for request by '%s' for '%s'\n", > URL_HOST(requester), URL_HOST(query_url)); > > > On 09/05/2015, eocene <eocene at gmx.com> wrote: >> James wrote: >>> Here's a guess without justification: it's trying to get confirmation >>> that you accepted a cookie, and failing. >> >> I did make a quick attempt with accept_session in cookiesrc, and it >> didn't appear to make a difference... >> >>> The basis for my guess is that login to gmail is possible, but >>> requires more than five redirects, so I run a hacked up dillo 3.1, >>> with a dialog about whether to keep redirecting. >> >> How many redirects does it need? The only site that I tend to run into >> redirect problems with is nytimes.com (Not logging into any account; >> just looking at articles that people post links to...), and the last >> time I checked it was something like 12 redirects before it will give >> up and provide the page. And I wonder whether raising the redirect >> limit might be good, overall. >> >> Does gmail still need all of those redirects if you let it have a >> cookie? >> >> _______________________________________________ >> Dillo-dev mailing list >> Dillo-dev at dillo.org >> http://lists.dillo.org/cgi-bin/mailman/listinfo/dillo-dev >> >

June 6, 2015

Patch: fix the flowchart diagram in the dw::Table class docs

by sgeerken＠dillo.org

On So, Mai 31, 2015, Jeremy Henty wrote: > The main body of the dw::Table class documentation starts with a > flowchart diagram of some dw::Table method calls. But this diagram > refers to a method dw::Table::calcColumnExtremes() that is no longer > used and has been renamed to dw::Table::_unused_calcColumnExtremes(). > (There is a warning that GROWS made some of this documentation > obsolete.) So I dug into the code and came up with a patch that I > think fixes the diagram. I have attached before and after images of > the diagram so that you can compare. I would love some feedback! > > http://starurchin.org/dillo/patch/doxygen-dw-table-hh-dependency-diagram.ht… > > And this patch removes dw::Table::_unused_calcColumnExtremes(). > > http://starurchin.org/dillo/patch/remove-dw-table-_unused_calcColumnExtreme… > > Comments are welcome. I've committed both, the first (diagram) with some changes. The table code is now stable enough so that it makes sense to update the documentation. (The rest should be reviewed, too.) BTW, there are also some more changes in the repository at <http://flpsed.org/hgweb/dillo_grows>, see the general documentation about widget sizes there. Sebastian

June 2, 2015

Documentation in the tarball?

by sgeerken＠dillo.org

Hi! The documentation is now split: - "doc" contains user documentation, and is included in the tarball. - "devdoc" contains developer documentation, and is only contained in the Hg repository, not in the tarball. Running "doxygen" will still create documentation in the "html" directory. Sebastian

June 1, 2015

Oh, list moderator...

by eocene＠gmx.com

Jorge wrote: > I don't remember who the moderator is these days, AFAIR > sometimes pending messages got to me, and other people have > taken this duty too (along the years). I wouldn't be shocked if it turned out that no one is at present. Sometimes I wonder whether the webmaster mail goes to anyone now.

June 1, 2015

Oh, list moderator...

by jcid＠dillo.org

Hi Jeremy, On Sun, May 31, 2015 at 06:14:12PM +0100, Jeremy Henty wrote: > > (I hope you will take this in the light-hearted tone that I wrote it. > I don't often write poetry.) > > Oh, list moderator, help me please. > I sent the list an email with attached PNGs. > I saw there was a problem, had an itch to scratch. > And that's why I was hoping you would take my patch. > But I knew that you would need some actual evidence > To show you what I'm saying really makes some sense. > > I saw a flowchart image in the HTML page > And noticed with regret that it had rotted with age. > It linked to a method that wasn't there no more > And that's when I decided I should settle this score. > > I made a better version and then sent it to you > And I chose to enclose the original too > But sadly those two images were too big for the list > So now now your mailer's told me I should cease and desist. > > Please, list moderator, let my work through. > Whatever you may think I only did it for you. > I have helped you before, I want to help you once more. > But you've got to tell your mail software to open the door. Hey, I liked your text! ;) I don't remember who the moderator is these days, AFAIR sometimes pending messages got to me, and other people have taken this duty too (along the years). Anyway, I went into the admin interface and allowed your message, but don't remember receiving an authorization request from the system. -- Cheers Jorge.-

June 1, 2015

Oh, list moderator...

by onepoint＠starurchin.org

(I hope you will take this in the light-hearted tone that I wrote it. I don't often write poetry.) Oh, list moderator, help me please. I sent the list an email with attached PNGs. I saw there was a problem, had an itch to scratch. And that's why I was hoping you would take my patch. But I knew that you would need some actual evidence To show you what I'm saying really makes some sense. I saw a flowchart image in the HTML page And noticed with regret that it had rotted with age. It linked to a method that wasn't there no more And that's when I decided I should settle this score. I made a better version and then sent it to you And I chose to enclose the original too But sadly those two images were too big for the list So now now your mailer's told me I should cease and desist. Please, list moderator, let my work through. Whatever you may think I only did it for you. I have helped you before, I want to help you once more. But you've got to tell your mail software to open the door. Regards, Jeremy Henty

May 31, 2015

Patch: fix the flowchart diagram in the dw::Table class docs

by onepoint＠starurchin.org

The main body of the dw::Table class documentation starts with a flowchart diagram of some dw::Table method calls. But this diagram refers to a method dw::Table::calcColumnExtremes() that is no longer used and has been renamed to dw::Table::_unused_calcColumnExtremes(). (There is a warning that GROWS made some of this documentation obsolete.) So I dug into the code and came up with a patch that I think fixes the diagram. I have attached before and after images of the diagram so that you can compare. I would love some feedback! http://starurchin.org/dillo/patch/doxygen-dw-table-hh-dependency-diagram.ht… And this patch removes dw::Table::_unused_calcColumnExtremes(). http://starurchin.org/dillo/patch/remove-dw-table-_unused_calcColumnExtreme… Comments are welcome. Regards, Jeremy Henty

May 31, 2015

Alternative styles support?

by higuita7＠yahoo.co.uk

Hi! On Sun, 24 May 2015 20:39:55 +0000, eocene <eocene at gmx.com> wrote: > Sebastian wrote: > > Iceweasel (Firefox): > > View -> Page Style -> (No Style | Fresh, Revolution) > Only now do I find that it's possible to get the menu bar back. I hehehe, most people don't know about the firefox menu bar, but is easy to get to it, just press "alt" for a temporary access or right click on the existent bar and select that you want to see the menu permanently > believe mine disappeared when firefox decided to move everything into > that stupid three-horizontal-bars icon because they're trend-followers, Not just trend followers, netbooks have a lot less screen size than desktops and the existence of the status bar and menu bar removes precious screen size, so that decision was to increase usable space. Of course i have the opinion that FF should check the screen size and change the default based on that... but i understand that would break consistency. Back to the on-topic: i would also agree that alternative styles support would be good thing. right now i have to switch off embebed or external CSS to be able to read some sites, a way to use other styles or even force a "personal" style would help in many sites. Best regards, higuita -- Naturally the common people don't want war... but after all it is the leaders of a country who determine the policy, and it is always a simple matter to drag the people along, whether it is a democracy, or a fascist dictatorship, or a parliament, or a communist dictatorship. Voice or no voice, the people can always be brought to the bidding of the leaders. That is easy. All you have to do is tell them they are being attacked, and denounce the pacifists for lack of patriotism and exposing the country to danger. It works the same in every country. -- Hermann Goering, Nazi and war criminal, 1883-1946

May 26, 2015

long lines and slowness (common when viewing style sheets)

by sgeerken＠dillo.org

On Di, Mai 26, 2015, eocene wrote: > > Now that everyone likes to send their style sheets as one big long line, > like 250k long, viewing them in dillo is slow. > > gprof says > % cumulative self self total > time seconds seconds calls s/call s/call name > 32.47 1.88 1.88 189407242 0.00 0.00 lout::misc::NotSoSimpleVector<dw::Textblock::Word>::getRef(int) const > 29.36 3.58 1.70 187437456 0.00 0.00 dw::Textblock::BadnessAndPenalty::lineCanBeBroken(int) > 28.15 5.21 1.63 29345 0.00 0.00 dw::Textblock::wrapWordInFlow(int, bool) > > and I wonder whether any optimizations are possible. Yes, I know this bug (you sent it me some time ago), and it is on my todo list. Since this occured first with hyphenation (or rather per-line optimization), there is certainly a way to optimize it. A vague guess is that it has to do with the iteration over the line to find the best break point. (BTW: the functionality used for plain text could also be provided by a new widget which is much simpler than Textblock.) Sebastian

May 26, 2015

long lines and slowness (common when viewing style sheets)

by eocene＠gmx.com

Now that everyone likes to send their style sheets as one big long line, like 250k long, viewing them in dillo is slow. gprof says % cumulative self self total time seconds seconds calls s/call s/call name 32.47 1.88 1.88 189407242 0.00 0.00 lout::misc::NotSoSimpleVector<dw::Textblock::Word>::getRef(int) const 29.36 3.58 1.70 187437456 0.00 0.00 dw::Textblock::BadnessAndPenalty::lineCanBeBroken(int) 28.15 5.21 1.63 29345 0.00 0.00 dw::Textblock::wrapWordInFlow(int, bool) and I wonder whether any optimizations are possible.

May 26, 2015

view source hang

by jcid＠dillo.org

On Thu, May 21, 2015 at 07:05:45PM +0000, eocene wrote: > > If a page has null padding, viewing source doesn't complete. > Example attached. I'm working on it... -- Cheers Jorge.-

May 25, 2015

Alternative styles support?

by sgeerken＠dillo.org

On Mo, Mai 25, 2015, eocene wrote: > Sebastian wrote: > > In dillo, it should be rather simple to integrate the styles into the > > tools menu. Perhaps some configuration per site would be nice, too. > > Do you suppose it would be more suitable to keep the user's setting for > a page in nav or in cache? Objectively, cache seems right, but in some > way feels like an unusual intrusion. It should resemble the current behaviour for "Use remote CSS" etc. Which I find rather confusing, admittedly: Switching it in one browser window switches the check boxes in all windows, but re-renders the page only in one window. And, what I meant previously: a simple text file in which pages and sites can be configured. Sebastian

May 25, 2015

Alternative styles support?

by eocene＠gmx.com

Sebastian wrote: > In dillo, it should be rather simple to integrate the styles into the > tools menu. Perhaps some configuration per site would be nice, too. Do you suppose it would be more suitable to keep the user's setting for a page in nav or in cache? Objectively, cache seems right, but in some way feels like an unusual intrusion.

May 25, 2015

Documentation in the tarball?

by sgeerken＠dillo.org

On So, Mai 24, 2015, Jorge Arellano Cid wrote: > On Sun, May 24, 2015 at 02:25:49PM +0200, Sebastian Geerken wrote: > > Hi! > > > > Does it make sense to include all developer documentation into the > > tarball? As an alternative, only user documentation, plus a notice > > where to find the developer docs, could be included, while the rest > > can only be found in the repository at hg.dillo.org. I assume that > > someone who is interested in the internals of dillo will anyway refer > > to the lastest version from the Hg repository. > > > > Keeping it out of the tarball would have advantages: > > > > - It makes the tarball smaller. > > > > - We developers would not have to think that much about file sizes of > > the documentation: these files do not go into the tarball; and for > > those using the lastest version from the Hg repository, it should be > > reasonable to spend some more disc space. > > > > What do you think? > > One disadvantage is no longer having both synced in the same > tarball. > > Now, if docs live in the same hg repo as sources, but just don't > make it into the tarball, I see no problems. They are kept in sync for > devs, and the tarball goes smaller, in which case I do agree. I want to keep them in the same repository, especially since Doxygen works also on the sources. I thought of something simple like removing most files from EXTRA_DIST in doc/Makefile.am. (BTW, EXTRA_DIST has been poorly maintained in the past, so the tarball did have an incomplete documentation sometimes.) Sebastian

May 24, 2015

Documentation in the tarball?

by jcid＠dillo.org

On Sun, May 24, 2015 at 02:25:49PM +0200, Sebastian Geerken wrote: > Hi! > > Does it make sense to include all developer documentation into the > tarball? As an alternative, only user documentation, plus a notice > where to find the developer docs, could be included, while the rest > can only be found in the repository at hg.dillo.org. I assume that > someone who is interested in the internals of dillo will anyway refer > to the lastest version from the Hg repository. > > Keeping it out of the tarball would have advantages: > > - It makes the tarball smaller. > > - We developers would not have to think that much about file sizes of > the documentation: these files do not go into the tarball; and for > those using the lastest version from the Hg repository, it should be > reasonable to spend some more disc space. > > What do you think? One disadvantage is no longer having both synced in the same tarball. Now, if docs live in the same hg repo as sources, but just don't make it into the tarball, I see no problems. They are kept in sync for devs, and the tarball goes smaller, in which case I do agree. -- Cheers Jorge.-

May 24, 2015

Alternative styles support?

by johnfound＠asm32.info

On Sun, 24 May 2015 20:39:55 +0000 eocene <eocene at gmx.com> wrote: > Sebastian wrote: > > Iceweasel (Firefox): > > View -> Page Style -> (No Style | Fresh, Revolution) > > Only now do I find that it's possible to get the menu bar back. I believe > mine disappeared when firefox decided to move everything into that stupid > three-horizontal-bars icon because they're trend-followers, and anyway I > thought that was just the new terrible way of things to not have access > to anything anymore. LOL! That is why I use PaleMoon browser. The same FireFox, but with classic interface. :D -- http://fresh.flatassembler.net http://asm32.info John Found <johnfound at asm32.info>

May 24, 2015

Alternative styles support?

by eocene＠gmx.com

Sebastian wrote: > Iceweasel (Firefox): > View -> Page Style -> (No Style | Fresh, Revolution) Only now do I find that it's possible to get the menu bar back. I believe mine disappeared when firefox decided to move everything into that stupid three-horizontal-bars icon because they're trend-followers, and anyway I thought that was just the new terrible way of things to not have access to anything anymore.

May 24, 2015

Alternative styles support?

by johnfound＠asm32.info

On Sun, 24 May 2015 19:26:17 +0000 eocene <eocene at gmx.com> wrote: > > How do other browsers inform the user that alternative stylesheets are > available? > In firefox, all styles can be switched from "View|Page Style" menu. Also, there is a plugin that displays the available styles in the status bar, allows fast switching and can remember the preferred style for the particular web sites. -- http://fresh.flatassembler.net http://asm32.info John Found <johnfound at asm32.info>

May 24, 2015

Alternative styles support?

by sgeerken＠dillo.org

On So, Mai 24, 2015, eocene wrote: > John wrote: > > On Sun, 24 May 2015 15:39:14 +0000 > > eocene <eocene at gmx.com> wrote: > > > > > John wrote: > > > > Are there plans about supporting alternative page styles? > > > > > > > > I mean, those defined by <link rel="alternate stylesheet" />. > > > > > > I'd sort of idly considered it. How is the existence of these > > > alternatives presented in other browsers? How common are > > > alternative stylesheets on the contemporary web? > > > > This is feature, supported by all major browsers. I don't know about the > > whole www, but I have a site, that uses such approach: > > > > http://fresh.flatassembler.net > > > > The supplementary style (named Revolution) was made because of user > > request - he/she wanted to use more screen area and without hidden > > menus. > > How do other browsers inform the user that alternative stylesheets are > available? Iceweasel (Firefox): View -> Page Style -> (No Style | Fresh, Revolution) Haven't yet found out how it works with Chromium. In dillo, it should be rather simple to integrate the styles into the tools menu. Perhaps some configuration per site would be nice, too. Sebastian

May 24, 2015

Alternative styles support?

by eocene＠gmx.com

John wrote: > On Sun, 24 May 2015 15:39:14 +0000 > eocene <eocene at gmx.com> wrote: > > > John wrote: > > > Are there plans about supporting alternative page styles? > > > > > > I mean, those defined by <link rel="alternate stylesheet" />. > > > > I'd sort of idly considered it. How is the existence of these > > alternatives presented in other browsers? How common are > > alternative stylesheets on the contemporary web? > > This is feature, supported by all major browsers. I don't know about the > whole www, but I have a site, that uses such approach: > > http://fresh.flatassembler.net > > The supplementary style (named Revolution) was made because of user > request - he/she wanted to use more screen area and without hidden > menus. How do other browsers inform the user that alternative stylesheets are available?

May 24, 2015

Alternative styles support?

by johnfound＠asm32.info

On Sun, 24 May 2015 15:39:14 +0000 eocene <eocene at gmx.com> wrote: > John wrote: > > Are there plans about supporting alternative page styles? > > > > I mean, those defined by <link rel="alternate stylesheet" />. > > I'd sort of idly considered it. How is the existence of these > alternatives presented in other browsers? How common are > alternative stylesheets on the contemporary web? This is feature, supported by all major browsers. I don't know about the whole www, but I have a site, that uses such approach: http://fresh.flatassembler.net The supplementary style (named Revolution) was made because of user request - he/she wanted to use more screen area and without hidden menus. -- http://fresh.flatassembler.net http://asm32.info John Found <johnfound at asm32.info>

May 24, 2015

Documentation in the tarball?

by eocene＠gmx.com

Sebastian wrote: > Keeping it out of the tarball would have advantages: > > - It makes the tarball smaller. > > - We developers would not have to think that much about file sizes of > the documentation: these files do not go into the tarball; and for > those using the lastest version from the Hg repository, it should be > reasonable to spend some more disc space. > > What do you think? *agrees*

May 24, 2015

Alternative styles support?

by eocene＠gmx.com

John wrote: > Are there plans about supporting alternative page styles? > > I mean, those defined by <link rel="alternate stylesheet" />. I'd sort of idly considered it. How is the existence of these alternatives presented in other browsers? How common are alternative stylesheets on the contemporary web?

May 24, 2015

Documentation in the tarball?

by sgeerken＠dillo.org

Hi! Does it make sense to include all developer documentation into the tarball? As an alternative, only user documentation, plus a notice where to find the developer docs, could be included, while the rest can only be found in the repository at hg.dillo.org. I assume that someone who is interested in the internals of dillo will anyway refer to the lastest version from the Hg repository. Keeping it out of the tarball would have advantages: - It makes the tarball smaller. - We developers would not have to think that much about file sizes of the documentation: these files do not go into the tarball; and for those using the lastest version from the Hg repository, it should be reasonable to spend some more disc space. What do you think? Sebastian

May 24, 2015

Alternative styles support?

by johnfound＠asm32.info

Hi, Dillo team. Are there plans about supporting alternative page styles? I mean, those defined by <link rel="alternate stylesheet" />. This should be really easy to be implemented. In addition, if there is defined style with name="Dillo", it can be switched by default and this way to provide better support for Dillo from the web masters. Of course this behaviour can be made optional. Best Regards. -- http://fresh.flatassembler.net http://asm32.info John Found <johnfound at asm32.info>

May 24, 2015

valgrind

by onepoint＠starurchin.org

eocene wrote: > Jeremy wrote: > > [...] > > Something has broken my hg access. I just noticed that my hg update > > job has been failing for a while, but I haven't made any changes to it > > ages. > > > > $ hg incoming > > abort: error: [SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed (_ssl.c:581) > > > > Can anyone help? > > I see that I get the same thing if I have cacerts in .hgrc. My .hgrc does not mention cacerts. All it does is set my name and enable the mq and graphlog extensions. > (It would seem that I normally use hostfingerprints instead. It's > been long enough that I don't remember whether I had any good reason > for that...) Maybe I should use host fingerprints too. How do you set that up? Regards, Jeremy Henty

May 23, 2015

valgrind

by eocene＠gmx.com

Jeremy wrote: > eocene wrote: > > I note the logs still say > > > > date: Tue Apr 28 21:42:28 2015 +0000 > > summary: BUG_MSG > > > > Are the https changes not playing well with your setup? > > Something has broken my hg access. I just noticed that my hg update > job has been failing for a while, but I haven't made any changes to it > ages. > > $ hg incoming > abort: error: [SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed (_ssl.c:581) > > Can anyone help? I see that I get the same thing if I have cacerts in .hgrc. (It would seem that I normally use hostfingerprints instead. It's been long enough that I don't remember whether I had any good reason for that...)

May 22, 2015

valgrind

by onepoint＠starurchin.org

eocene wrote: > I note the logs still say > > date: Tue Apr 28 21:42:28 2015 +0000 > summary: BUG_MSG > > Are the https changes not playing well with your setup? Something has broken my hg access. I just noticed that my hg update job has been failing for a while, but I haven't made any changes to it ages. $ hg incoming abort: error: [SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed (_ssl.c:581) Can anyone help? Jeremy

May 22, 2015

view source hang

by eocene＠gmx.com

If a page has null padding, viewing source doesn't complete. Example attached.

May 21, 2015

not following 301 Moved Permanently

by dennisn＠dennisn.mooo.com

When accessing some sites, say for example http://thepiratebay.org , wget seems to successfully follow the "301 Moved Permanently" redirects a few times, but Dillo only seems to do one redirect and then stops. The last output is: HTTP warning: Message with a body lacked Content-Type header.

May 20, 2015

No edit/delete/report Buttons in Some phpBB Forums

by miro.rovis＠croatiafidelis.hr

I've been using dillo for my regular browsing for a few weeks longer than from when I started the thread: Secure: yes, Though Features Missing; and Bugs http://lists.dillo.org/pipermail/dillo-dev/2015-April/010495.html on dillo mailing list, and I'd like to help at least occasionally with the sole kind of contribution that I can offer: reporting like those in that link, as I'm not a developer, and can program very little only. I have a new one report. on issue that I only mentioned (and wouldn't know how to solve back then) over at: Grsecurity/Pax installation on Debian GNU/Linux http://forums.debian.net/viewtopic.php?f=16&t=108616&start=60#p577608 But I can try and tell more about it as I saw the same missing edit and delete buttons in: https://forum.sleuthkit.org/ where, to verify it, one would need to register, I believe. And while that registering on a specialized tool forum can take a few hours or more time to get the maintainer's approval, true, I believe it is the same kind of missing functionality and in the Debian Forums, so I guess one can verify this behavior by, if they are not, registering into Debian Forums. I have noticed the same edit and delete buttons missing in Debian Forums too, and I choose the Sleuthkit Forum, because I'm currently posting on it, and can accomplish this report with much more ease then on Debian where I am done with my posts for some time. I have a slow and lengthy (and tiring: too advanced for me) topic on Sleuthkit Forum: An Avi Video Recovery with SleuthKit http://forum.sleuthkit.org/viewtopic.php?f=6&t=2441 ) and I have actually managed to edit my posts, and I managed to delete one (only the post not replied to can be deleted on Sleutkit Forum). But not by clicking on the respective icons, as they don't show in dillo. But by issuing Ctrl-U to open the source of the page while logged in and having my topic opened in my dillo browser and searching for the right (relative) link under the icon 'edit-icon' or 'delete-icon' and reconstructing the right page, because that was the only way to do it. I did search thoroughly, and no link, even on empty surfaces, did I manage to detect under the mouse (as I didn't in Debian Forums either). This is how I found it, after saving the source for the page, which is still opened for me since I posted on it, and where I deleted the last misposted post: The page was (and will not remain available under that exact address): Re: An Avi Video Recovery with SleuthKit https://forum.sleuthkit.org/viewtopic.php?f=6&t=2441&p=2630&sid=4e30a8bd740… (which is Page 2 of 2 of the topic; for those who will try elsewhere, or who just want to get the jist reading this without trying, this is the topic page 2: https://forum.sleuthkit.org/viewtopic.php?f=6&t=2441&p=2630#p2630 but that one has, surely no edit or delete -icon(s)) I saved it as: SK_150514-10_pg2_src.php ('10' because it's post 10 of the topic, 150514 is for 2015-05-14) and, for deleting, which is quicker to show, since it can be only one instance found, only the last post: cat /Cmn/dLo/SK_150514-10_pg2_src.php | grep delete-icon | \ sed 's/\&amp;/\&/g' and that gives, in this case (surely anyone would need their own current address given by the forum phpBB, but this is the principle, and even I as the same miroRR user will still get a different one variant of the string 'sid=...' for the same page the next time), [and that gives, in this case:] </pre><tr><td bgcolor='#DD7F32'>238<td><pre> <li class="edit-icon"><a href="./posting.php?mode=edit&f=6&p=2634&sid=4e30a8bd7401b94368a85571eaebe8f2" title="Edit post"><span>Edit post</span></a></li><li class="delete-icon"><a href="./posting.php?mode=delete&f=6&p=2634&sid=4e30a8bd7401b94368a85571eaebe8f2" title="Delete post"><span>Delete post</span></a></li><li class="report-icon"><a href="./report.php?f=6&p=2634&sid=4e30a8bd7401b94368a85571eaebe8f2" title="Report this post"><span>Report this post</span></a></li><li class="quote-icon"><a href="./posting.php?mode=quote&f=6&p=2634&sid=4e30a8bd7401b94368a85571eaebe8f2" title="Reply with quote"><span>Reply with quote</span></a></li> (When grep'ing 'edit-icon', 'report-icon' or 'quote-icon' there would be two founds for each, as there are two posts.) in one line, however, but surely I prefer my Vim setting it to the usual 79 colums width, so readers in the mailing list would not have trouble viewing it as one line (I have mercy on my poor brothers in *nix ;-) ). Surely, if I will have to post this text on Gentoo Forums [*] and [**] and [***], case shown to be that this e-mail just does not send by the absolute joyous great provider of mine, urgh!, I will rather set it to around 110-120 colums width... But from that HTML code, I figured out from this segment: href="./posting.php?mode=delete&f=6&p=2634&sid=4e30a8bd7401b94368a85571eaebe8f2" that if I wanted to delete that post, I would change the address from what it looked like into: https://forum.sleuthkit.org/posting.php?mode=delete&f=6&p=2634&sid=4e30a8bd… And indeed, after hitting Enter on that address I was asked for confirmation and was able to delete the post. There are also strings to use for changing the same address into the one that would accomplish editing that post: href="./posting.php?mode=edit&f=6&p=2634&sid=4e30a8bd7401b94368a85571eaebe8f2" reporting that post: href="./report.php?f=6&p=2634&sid=4e30a8bd7401b94368a85571eaebe8f2" and quoting that post: href="./posting.php?mode=quote&f=6&p=2634&sid=4e30a8bd7401b94368a85571eaebe8f2" None of those icons, however, show in dillo for me. This is about the whole report, as I can't even dedicate time to follow the suggestion be kind unixer eocene: Secure: yes, Though Features Missing; and Bugs http://lists.dillo.org/pipermail/dillo-dev/2015-April/010500.html and get my fonts sorted in dillo... since even the regular maintainance of my system during a data recovery work is already schedule completely exhausted, let alone DIY compilations. [*] I'll post this, in case it is not sent to dillo ML by my provider, on Gentoo Forums where I have anyway been planning for some time the idea to open a topic on browsers and really recommend dillo, as the its being so sea-calm secure when compared with the big surveillor browsers outweighs the missing features such as the lack of javascript and poor support for CSS, so this would fit on Gentoo Forums. [**] And I'll open a special section for suspected censored e-mails of mine, at http://www.CroatiaFidelis.hr/foss/cenz/ . However, uncenzorizing is a lot of work, it's in the order of a tenfold work than non-censored free e-mail correspondence, so I'll do it, I hope, some time in the future. [***] START Speaking of big surveillor browsers, I usually have issues with dillo right after I use one of those: the SchmoogleMoz Fox (but freely replace it with any of the other big ones and the potential for harm like below is there!). --- I had to use it, because the freaking Croatian corporate robber Zagreba?ka banka at www.zaba.hr require the use of javascript for internet banking... --- And that certainly is an issue, the requirement of javascript and privacy/intrusion/surveillance issues, which requirement and these issues is what I want to open the topic on Gentoo on for, and which would be great that dillo managed, with some also, hopefully, speaking as an incompetent but wishful observer --incompetent because not really a programmer-- with some also new honest young hackers/developers' blood to join in among the current herd of dillo developers [[ the work needed is huuuge ]] -- I'm dreaming, brothers, dreaming of a free unsurveilled surfing available for many; dillo looks really really closest to that accomplishment in the field of today's computing which is the sole island in all of computing with potential to be unsurveilled and free users of it therefore not controlled, and which [the sole island] is the FOSS *nix, from BSD variants to Linux variants... --- And so I had to use the SchmoogleFox, and lo and behold, certain stuff in dillo would not work after the bout of only some ten (10) minutes of the Schfox use. Such as I wasn't able to accomplish anything with either Ctrl-U nor right mouse click > View page source. Dillo would crash immediately. Updated my Gentoo from my local mirror (it's kernel 4 now in Gentoo ;-) !!) ( see my topic: Air-Gapped Gentoo Install, Tentative https://forums.gentoo.org/viewtopic-t-987268.html and links therefrom, such as to cloning system, a part-topic in: Postfix smtp/TLS, Bkp/Cloning Mthd, Censorship/Intrusion https://forums.gentoo.org/viewtopic-t-999436.html for how to always be able to get renewable clean system even after intrusions ) and all now works. END [***] --- All links in this message checked and all I have found to be live before sending. -- Miroslav Rovis Zagreb, Croatia http://www.CroatiaFidelis.hr

May 14, 2015

Google broke recently with 3.0.4 causing repush loop

by eocene＠gmx.com

Dennis wrote: > On Sat, 9 May 2015 15:59:56 +0000, eocene wrote: > > James wrote: > > > This code cannot be put into production, because it produces > > > messages about redirecting 0 times reasonably often, for reasons > > > that I haven't tracked down yet. > > > > Having a dialog when we're looping does sound like a good idea to > > me... > > Why automatically follow the meta redirect at all? I notice that when > the time period is non-zero, it doesn't automatically follow, but > instead outputs something like: > > "The author wanted you to go here after 1 second" > > That seems to make more sense to me -- it avoids loops, and is more > informative for users? Jorge put it in. I don't remember why now. http://lists.dillo.org/pipermail/dillo-dev/2009-April/006294.html In any case, if we don't automatically redirect based on the meta tags, there are still the redirections from Location headers in HTTP.

May 9, 2015

Google broke recently with 3.0.4 causing repush loop

by dennisn＠dennisn.mooo.com

On Sat, 9 May 2015 15:59:56 +0000, eocene wrote: > James wrote: > > This code cannot be put into production, because it produces > > messages about redirecting 0 times reasonably often, for reasons > > that I haven't tracked down yet. > > Having a dialog when we're looping does sound like a good idea to > me... Why automatically follow the meta redirect at all? I notice that when the time period is non-zero, it doesn't automatically follow, but instead outputs something like: "The author wanted you to go here after 1 second" That seems to make more sense to me -- it avoids loops, and is more informative for users?

May 9, 2015

Google broke recently with 3.0.4 causing repush loop

by eocene＠gmx.com

James wrote: > This code cannot be put into production, because it produces messages > about redirecting 0 times reasonably often, for reasons that I haven't > tracked down yet. Having a dialog when we're looping does sound like a good idea to me...

May 9, 2015

Google broke recently with 3.0.4 causing repush loop

by andreas＠kemnade.info

Hi, On Sat, 9 May 2015 01:28:40 +0000 eocene <eocene at gmx.com> wrote: > If I use a firefox user agent string, though, I am redirected from http to > an https page which works and doesn't loop because...it puts the meta refresh > down in the body instead of the head, so Dillo disallows it. > trying with lynx gives no refresh at all! http request resulting from a search with lynx: GET /search?ie=ISO-8859-1&hl=de&source=hp&q=dillo&btnG=Google-Suche&gbv=1 HTTP/1.0 Host: www.google.de Accept: text/html, text/plain, text/css, text/sgml, */*;q=0.01 Accept-Encoding: gzip, compress, bzip2 Accept-Language: en User-Agent: Lynx/2.8.8dev.12 libwww-FM/2.14 SSL-MM/1.4.1 GNUTLS/2.12.18 Referer: http://www.google.de/ It gives javascript in the output but no meta refresh. Lynx works perfectly with google. Greetings Andreas

May 9, 2015

Google broke recently with 3.0.4 causing repush loop

by james.from.wellington＠gmail.com

All of this is with the cookiesrc below, which lets google have cookies on two domains. If I: - log out of google -- it stalls on a meta refresh tag - select home, which for me is https://mail.google.com/?ui=html - provide a username and password -- I am eventually shown my email messages then the code below redirects: - once during the logout - twice before asking for a password - eight (8) times before showing me my mail Adding back the hack to share cookies probably makes no difference to these figures, but I'm not sure, because I think that I remember inconsistently getting seven and nine redirects, so I think that google is A/B testing me. This code cannot be put into production, because it produces messages about redirecting 0 times reasonably often, for reasons that I haven't tracked down yet. -------------------------------- #host action DEFAULT DENY .google.com ACCEPT_SESSION .google.co.nz ACCEPT_SESSION -------------------------------- render/dillo-jfw01$ hg diff diff -r 07c0c28f8960 src/cache.c --- a/src/cache.c Fri Mar 06 13:12:11 2015 +1300 +++ b/src/cache.c Sat May 09 22:03:31 2015 +1200 @@ -1007,11 +1007,24 @@ return 0; } - /* if there's a redirect loop, stop now */ - if (bw->redirect_level >= 5) - entry->Flags |= CA_RedirectLoop; +// if (0 == bw->redirect_level % 5) { + if (0 == bw->redirect_level % 1) { + char *redirMessage; + int choice; + + asprintf(&redirMessage, "Dillo: redirected %d times", bw->redirect_level); + + choice = a_Dialog_choice(redirMessage, URL_STR(entry->Url), + "Stop", "Continue", NULL); + + if (1 == choice) + entry->Flags |= CA_RedirectLoop; + + free(redirMessage); + } if (entry->Flags & CA_RedirectLoop) { + MSG("Redirect loop for: %s\n", URL_STR_(entry->Url)); a_UIcmd_set_msg(bw, "ERROR: redirect loop for: %s", URL_STR_(entry->Url)); bw->redirect_level = 0; return 0; diff -r 07c0c28f8960 src/cookies.c --- a/src/cookies.c Fri Mar 06 13:12:11 2015 +1300 +++ b/src/cookies.c Sat May 09 22:03:31 2015 +1200 @@ -196,6 +196,15 @@ if (requester == NULL) { /* request made by user */ +#if 0 + } else if (!strcmp("accounts.google.co.nz", URL_HOST(requester)) && + !strcmp("mail.google.com", URL_HOST(query_url))) { + MSG("Cookies: sent for google NZ special case 1.\n"); + } else if (!strcmp("accounts.google.com", URL_HOST(requester)) && + !strcmp("accounts.google.co.nz", URL_HOST(query_url))) { + MSG("Cookies: sent for google NZ special case 2.\n"); + +#endif } else if (!a_Url_same_organization(query_url, requester)) { MSG("Cookies: not sent for request by '%s' for '%s'\n", URL_HOST(requester), URL_HOST(query_url)); On 09/05/2015, eocene <eocene at gmx.com> wrote: > James wrote: >> Here's a guess without justification: it's trying to get confirmation >> that you accepted a cookie, and failing. > > I did make a quick attempt with accept_session in cookiesrc, and it > didn't appear to make a difference... > >> The basis for my guess is that login to gmail is possible, but >> requires more than five redirects, so I run a hacked up dillo 3.1, >> with a dialog about whether to keep redirecting. > > How many redirects does it need? The only site that I tend to run into > redirect problems with is nytimes.com (Not logging into any account; > just looking at articles that people post links to...), and the last > time I checked it was something like 12 redirects before it will give > up and provide the page. And I wonder whether raising the redirect > limit might be good, overall. > > Does gmail still need all of those redirects if you let it have a > cookie? > > _______________________________________________ > Dillo-dev mailing list > Dillo-dev at dillo.org > http://lists.dillo.org/cgi-bin/mailman/listinfo/dillo-dev >

May 9, 2015

SSL in the browser

by eocene＠gmx.com

I wrote: > I had planned to wait until after 3.1 to integrate the ssl-in-browser > code, but since 1) dillo hasn't been moving toward release lately and > 2) Johannes tells me that the patch has been working well for him, I > now plan to put it into dillo when I get a chance in coming days. > > Having it in dillo means that dillo can be a better internet citizen > (http_max_conns, http_persistent_conns), there is certificate hostname > checking borrowed from wget, server name indication (fewer certificate > warnings), we can remember for the session that the user accepted the > use of a questionable certificate instead of continuing to give warnings, > we can check non-root-url certificates without endless warning popups... (Also the Tor + dpi issue that Johannes mentioned last year) Committed. I'm looking forward to seeing how it works for you...

May 9, 2015

Google broke recently with 3.0.4 causing repush loop

by eocene＠gmx.com

James wrote: > Here's a guess without justification: it's trying to get confirmation > that you accepted a cookie, and failing. I did make a quick attempt with accept_session in cookiesrc, and it didn't appear to make a difference... > The basis for my guess is that login to gmail is possible, but > requires more than five redirects, so I run a hacked up dillo 3.1, > with a dialog about whether to keep redirecting. How many redirects does it need? The only site that I tend to run into redirect problems with is nytimes.com (Not logging into any account; just looking at articles that people post links to...), and the last time I checked it was something like 12 redirects before it will give up and provide the page. And I wonder whether raising the redirect limit might be good, overall. Does gmail still need all of those redirects if you let it have a cookie?

May 9, 2015

Google broke recently with 3.0.4 causing repush loop

by eocene＠gmx.com

Dennis wrote: > This past week or so Google (main search site, image.google) broke for > me -- Dillo (3.0.4) goes into an endless "repush" loop: > > Nav_open_url: new url='http://images.google.com/images?q=bra' > Dns_server [0]: images.google.com is 96.127.250.44 96.127.250.42 > 96.127.250.38 96.127.250.34 96.127.250.30 96.127.250.29 96.127.250.27 > 96.127.250.23 96.127.250.19 96.127.250.15 96.127.250.59 96.127.250.57 > 96.127.250.53 96.127.250.49 96.127.250.45 2607:f8b0:4006:80f::200e > > Connecting to 96.127.250.44 > Nav_open_url: new url='http://images.google.com/search?q=bra&tbm=isch' > Connecting to 96.127.250.44 > > >>>> a_Nav_repush <<<< > Nav_open_url: new url='http://images.google.com/search?q=bra&tbm=isch' > a_Nav_expect_done: repush! > Nav_open_url: new > url='http://images.google.com/search?q=bra&ie=UTF-8&tbm=isch&gbv=1&sei=zkNNVd6sI…' > Connecting to 96.127.250.44 > > >>>> a_Nav_repush <<<< > Nav_open_url: new > url='http://images.google.com/search?q=bra&ie=UTF-8&tbm=isch&gbv=1&sei=zkNNVd6sI…' > a_Nav_expect_done: repush! Nav_open_url: new > url='http://images.google.com/search?q=bra&ie=UTF-8&tbm=isch&gbv=1&sei=zkNNVYTpK…' > Connecting to 96.127.250.44 > >>>> a_Nav_repush <<<< > > ... and so on, extermely quickly, which then results in Google > temporarily blocking me since Dillo is behaving like a DDOS attacker :s. > > Any idea what's causing these repush'es? I'm pretty sure nothing > changed on my end. Other browsers work for me (eg. Links, Uzbl). > Changing the User Agent had no effect. It looks like google is misbehaving. http://www.google.com/search?ie=UTF-8&oe=UTF-8&q=whatever contains <noscript><meta content="0;url=/search?q=whatever&ie=UTF-8&oe=UTF-8&gbv=1&sei=sV5NVfbJMtfsoASOo4GQBw" http-equiv="refresh"><style>table,div,span,p{display:none}</style><div style="display:block">Please click <a href="/search?q=whatever&ie=UTF-8&oe=UTF-8&gbv=1&sei=sV5NVfbJMtfsoASOo4GQBw">here</a> if you are not redirected within a few seconds.</div></noscript> so we follow the refresh, and the new page has <noscript><meta content="0;url=/search?q=whatever&ie=UTF-8&gbv=1&sei=PV9NVej-Noj9oQSd-ICQDA" http-equiv="refresh"><style>table,div,span,p{display:none}</style><div style="display:block">Please click <a href="/search?q=whatever&ie=UTF-8&gbv=1&sei=PV9NVej-Noj9oQSd-ICQDA">here</a> if you are not redirected within a few seconds.</div></noscript> ...etc.... If I use a firefox user agent string, though, I am redirected from http to an https page which works and doesn't loop because...it puts the meta refresh down in the body instead of the head, so Dillo disallows it.

May 9, 2015

Google broke recently with 3.0.4 causing repush loop

by james.from.wellington＠gmail.com

Here's a guess without justification: it's trying to get confirmation that you accepted a cookie, and failing. The basis for my guess is that login to gmail is possible, but requires more than five redirects, so I run a hacked up dillo 3.1, with a dialog about whether to keep redirecting. On 09/05/2015, Dennis New <dennisn at dennisn.mooo.com> wrote: > This past week or so Google (main search site, image.google) broke for > me -- Dillo (3.0.4) goes into an endless "repush" loop: > > Nav_open_url: new url='http://images.google.com/images?q=bra' > Dns_server [0]: images.google.com is 96.127.250.44 96.127.250.42 > 96.127.250.38 96.127.250.34 96.127.250.30 96.127.250.29 96.127.250.27 > 96.127.250.23 96.127.250.19 96.127.250.15 96.127.250.59 96.127.250.57 > 96.127.250.53 96.127.250.49 96.127.250.45 2607:f8b0:4006:80f::200e > > Connecting to 96.127.250.44 > Nav_open_url: new url='http://images.google.com/search?q=bra&tbm=isch' > Connecting to 96.127.250.44 > >>>>> a_Nav_repush <<<< > Nav_open_url: new url='http://images.google.com/search?q=bra&tbm=isch' > a_Nav_expect_done: repush! > Nav_open_url: new > url='http://images.google.com/search?q=bra&ie=UTF-8&tbm=isch&gbv=1&sei=zkNNVd6sI…' > Connecting to 96.127.250.44 > >>>>> a_Nav_repush <<<< > Nav_open_url: new > url='http://images.google.com/search?q=bra&ie=UTF-8&tbm=isch&gbv=1&sei=zkNNVd6sI…' > a_Nav_expect_done: repush! Nav_open_url: new > url='http://images.google.com/search?q=bra&ie=UTF-8&tbm=isch&gbv=1&sei=zkNNVYTpK…' > Connecting to 96.127.250.44 >>>>> a_Nav_repush <<<< > > ... and so on, extermely quickly, which then results in Google > temporarily blocking me since Dillo is behaving like a DDOS attacker :s. > > Any idea what's causing these repush'es? I'm pretty sure nothing > changed on my end. Other browsers work for me (eg. Links, Uzbl). > Changing the User Agent had no effect. > > _______________________________________________ > Dillo-dev mailing list > Dillo-dev at dillo.org > http://lists.dillo.org/cgi-bin/mailman/listinfo/dillo-dev >

May 9, 2015

Google broke recently with 3.0.4 causing repush loop

by dennisn＠dennisn.mooo.com

This past week or so Google (main search site, image.google) broke for me -- Dillo (3.0.4) goes into an endless "repush" loop: Nav_open_url: new url='http://images.google.com/images?q=bra' Dns_server [0]: images.google.com is 96.127.250.44 96.127.250.42 96.127.250.38 96.127.250.34 96.127.250.30 96.127.250.29 96.127.250.27 96.127.250.23 96.127.250.19 96.127.250.15 96.127.250.59 96.127.250.57 96.127.250.53 96.127.250.49 96.127.250.45 2607:f8b0:4006:80f::200e Connecting to 96.127.250.44 Nav_open_url: new url='http://images.google.com/search?q=bra&tbm=isch' Connecting to 96.127.250.44 >>>> a_Nav_repush <<<< Nav_open_url: new url='http://images.google.com/search?q=bra&tbm=isch' a_Nav_expect_done: repush! Nav_open_url: new url='http://images.google.com/search?q=bra&ie=UTF-8&tbm=isch&gbv=1&sei=zkNNVd6sI…' Connecting to 96.127.250.44 >>>> a_Nav_repush <<<< Nav_open_url: new url='http://images.google.com/search?q=bra&ie=UTF-8&tbm=isch&gbv=1&sei=zkNNVd6sI…' a_Nav_expect_done: repush! Nav_open_url: new url='http://images.google.com/search?q=bra&ie=UTF-8&tbm=isch&gbv=1&sei=zkNNVYTpK…' Connecting to 96.127.250.44 >>>> a_Nav_repush <<<< ... and so on, extermely quickly, which then results in Google temporarily blocking me since Dillo is behaving like a DDOS attacker :s. Any idea what's causing these repush'es? I'm pretty sure nothing changed on my end. Other browsers work for me (eg. Links, Uzbl). Changing the User Agent had no effect.

May 9, 2015