Dear Dillo developers, since some days I get a segfault when starting the web search via <CTRL>s (using the dillo 3.0 trunk). Here is the backtrace: $ gdb ./dillo GNU gdb (GDB) 7.0.1-debian Copyright (C) 2009 Free Software Foundation, Inc. License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html> This is free software: you are free to change and redistribute it. There is NO WARRANTY, to the extent permitted by law. Type "show copying" and "show warranty" for details. This GDB was configured as "x86_64-linux-gnu". For bug reporting instructions, please see: <http://www.gnu.org/software/gdb/bugs/>... Reading symbols from /PATH/TO/DILLO/dillo...done. (gdb) r Starting program: /PATH/TO/DILLO/dillo [Thread debugging using libthread_db enabled] dillo_dns_init: Here we go! (threaded) Disabling cookies. Nav_open_url: new url='http://www.google.de' [New Thread 0x7ffff5111700 (LWP 21693)] Dns_server [0]: www.google.de is 74.125.43.106 74.125.43.104 74.125.43.99 74.125.43.147 74.125.43.105 74.125.43.103 [Thread 0x7ffff5111700 (LWP 21693) exited] Connecting to 74.125.43.106 <snip>Now I press <CTRL> s</snip> Program received signal SIGSEGV, Segmentation fault. __strrchr_sse2 () at ../sysdeps/x86_64/multiarch/../strrchr.S:33 33 ../sysdeps/x86_64/multiarch/../strrchr.S: Can't find file or directory. in ../sysdeps/x86_64/multiarch/../strrchr.S Current language: auto The current source language is "auto; currently asm". (gdb) l 28 in ../sysdeps/x86_64/multiarch/../strrchr.S (gdb) bt #0 __strrchr_sse2 () at ../sysdeps/x86_64/multiarch/../strrchr.S:33 #1 0x0000000000410640 in a_Misc_parse_search_url (source=0x0, label=0x7fffffffdb58, urlstr=0x7fffffffdb50) at misc.c:395 #2 0x0000000000414ddf in a_Dialog_input (msg=<value optimized out>) at dialog.cc:167 #3 0x000000000040d137 in a_UIcmd_search_dialog (vbw=0x774180) at uicmd.cc:820 #4 0x000000000040a9fd in UI::handle (this=0x7951a0, event=12) at ui.cc:821 #5 0x000000000046451b in send(int, Fl_Widget*, Fl_Window*) () #6 0x0000000000465415 in Fl::handle_(int, Fl_Window*) () #7 0x000000000046533c in Fl::handle_(int, Fl_Window*) () #8 0x0000000000485530 in fl_handle(_XEvent const&) () #9 0x000000000048558d in do_queued_events() () #10 0x00000000004857ac in fl_wait(double) () #11 0x0000000000465c23 in Fl::wait(double) () #12 0x0000000000465c93 in Fl::run() () #13 0x0000000000407f14 in main (argc=1, argv=0x7fffffffe3c8) at dillo.cc:429 Kind regards, Alexander
On Sat, Jul 30, 2011 at 12:14:30AM +0200, Alexander Voigt wrote:
Dear Dillo developers,
since some days I get a segfault when starting the web search via <CTRL>s (using the dillo 3.0 trunk). Here is the backtrace:
$ gdb ./dillo GNU gdb (GDB) 7.0.1-debian Copyright (C) 2009 Free Software Foundation, Inc. License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html> This is free software: you are free to change and redistribute it. There is NO WARRANTY, to the extent permitted by law. Type "show copying" and "show warranty" for details. This GDB was configured as "x86_64-linux-gnu". For bug reporting instructions, please see: <http://www.gnu.org/software/gdb/bugs/>... Reading symbols from /PATH/TO/DILLO/dillo...done. (gdb) r Starting program: /PATH/TO/DILLO/dillo [Thread debugging using libthread_db enabled] dillo_dns_init: Here we go! (threaded) Disabling cookies. Nav_open_url: new url='http://www.google.de' [New Thread 0x7ffff5111700 (LWP 21693)] Dns_server [0]: www.google.de is 74.125.43.106 74.125.43.104 74.125.43.99 74.125.43.147 74.125.43.105 74.125.43.103 [Thread 0x7ffff5111700 (LWP 21693) exited] Connecting to 74.125.43.106
<snip>Now I press <CTRL> s</snip>
Program received signal SIGSEGV, Segmentation fault. __strrchr_sse2 () at ../sysdeps/x86_64/multiarch/../strrchr.S:33 33 ../sysdeps/x86_64/multiarch/../strrchr.S: Can't find file or directory. in ../sysdeps/x86_64/multiarch/../strrchr.S Current language: auto The current source language is "auto; currently asm". (gdb) l 28 in ../sysdeps/x86_64/multiarch/../strrchr.S (gdb) bt #0 __strrchr_sse2 () at ../sysdeps/x86_64/multiarch/../strrchr.S:33 #1 0x0000000000410640 in a_Misc_parse_search_url (source=0x0, label=0x7fffffffdb58, urlstr=0x7fffffffdb50) at misc.c:395 #2 0x0000000000414ddf in a_Dialog_input (msg=<value optimized out>) at dialog.cc:167 #3 0x000000000040d137 in a_UIcmd_search_dialog (vbw=0x774180) at uicmd.cc:820 #4 0x000000000040a9fd in UI::handle (this=0x7951a0, event=12) at ui.cc:821 #5 0x000000000046451b in send(int, Fl_Widget*, Fl_Window*) () #6 0x0000000000465415 in Fl::handle_(int, Fl_Window*) () #7 0x000000000046533c in Fl::handle_(int, Fl_Window*) () #8 0x0000000000485530 in fl_handle(_XEvent const&) () #9 0x000000000048558d in do_queued_events() () #10 0x00000000004857ac in fl_wait(double) () #11 0x0000000000465c23 in Fl::wait(double) () #12 0x0000000000465c93 in Fl::run() () #13 0x0000000000407f14 in main (argc=1, argv=0x7fffffffe3c8) at dillo.cc:429
Does this fix it? dialog.cc:167 - if (a_Misc_parse_search_url(source, &label, &url) < 0) + if (!source || a_Misc_parse_search_url(source, &label, &url) < 0) -- Cheers Jorge.-
On Fri, Jul 29, 2011 at 07:02:17PM -0400, Jorge Arellano Cid wrote:
On Sat, Jul 30, 2011 at 12:14:30AM +0200, Alexander Voigt wrote:
Dear Dillo developers,
since some days I get a segfault when starting the web search via <CTRL>s (using the dillo 3.0 trunk). Here is the backtrace: [...]
Does this fix it?
dialog.cc:167
- if (a_Misc_parse_search_url(source, &label, &url) < 0) + if (!source || a_Misc_parse_search_url(source, &label, &url) < 0)
Please try the current repo (it has a better patch). -- Cheers Jorge.-
Hi Jorge and Corvid. On Fri, Jul 29, 2011 at 08:24:30PM -0400, Jorge Arellano Cid wrote:
Please try the current repo (it has a better patch).
Thank you for the fast reply. The current repository fixes the segfault for me. (By the way, the idea of switchable search engines is great!) Kind regards, Alexander
On Sat, Jul 30, 2011 at 09:40:29AM +0200, Alexander Voigt wrote:
Hi Jorge and Corvid.
On Fri, Jul 29, 2011 at 08:24:30PM -0400, Jorge Arellano Cid wrote:
Please try the current repo (it has a better patch).
Thank you for the fast reply. The current repository fixes the segfault for me.
Good.
(By the way, the idea of switchable search engines is great!)
Thanks. This is the first positive feedback I get so far on it. I'm a bit surprised on the lack of them given the flexibility of the feature. BTW, from some time I'm using duckduckgo (see dillorc) as my default search engine. It has served me very well, and recently discovered it also has plenty of handy macros: http://duckduckgo.com/bang.html (just in case you feel like experimenting). -- Cheers Jorge.-
Jorge wrote:
BTW, from some time I'm using duckduckgo (see dillorc) as my default search engine. It has served me very well, and recently discovered it also has plenty of handy macros:
Do you think we should add to its search string to disable safe search? At least in English, it likes to take out anything that can be construed as a rude word, which irritates me -- or sometimes you look up a quotation that does have strong language in it, and it should just give the results instead of acting like you're doing something immoral.
On Sat, Jul 30, 2011 at 06:44:11PM +0000, corvid wrote:
Jorge wrote:
BTW, from some time I'm using duckduckgo (see dillorc) as my default search engine. It has served me very well, and recently discovered it also has plenty of handy macros:
Do you think we should add to its search string to disable safe search? At least in English, it likes to take out anything that can be construed as a rude word, which irritates me -- or sometimes you look up a quotation that does have strong language in it, and it should just give the results instead of acting like you're doing something immoral.
Yes, after trying both, I strongly tend to a non-filtered search. I was about to ask the same question. Personally, after all these months, I'm all for making non-filtered duckduckgo our default: search_url="http://duckduckgo.com/html?kp=-1&q=%s" (we can move google somewhere after it). -- Cheers Jorge.-
On Sat, 30 Jul 2011 15:57:20 -0400, Jorge Arellano Cid <jcid@dillo.org> wrote:
Yes, after trying both, I strongly tend to a non-filtered search.
I was about to ask the same question. Personally, after all these months, I'm all for making non-filtered duckduckgo our default:
search_url="http://duckduckgo.com/html?kp=-1&q=%s"
(we can move google somewhere after it).
Of course, then you risk frustrating users who want/expect Google as the default. I always set the defaults in my software to things the average user would want or expect, even if it differs from my own preferences -- if I want to change it, I can set it in my personal configuration. The way I see it, I'm just a software developer; I'm here to write useful code, not to inflict my way of doing things on the world. Just a thought, ~Benjamin
Benjamin wrote:
On Sat, 30 Jul 2011 15:57:20 -0400, Jorge Arellano Cid <jcid@dillo.org> wrote:
Yes, after trying both, I strongly tend to a non-filtered search.
I was about to ask the same question. Personally, after all these months, I'm all for making non-filtered duckduckgo our default:
search_url="http://duckduckgo.com/html?kp=-1&q=%s"
(we can move google somewhere after it).
Of course, then you risk frustrating users who want/expect Google as the default.
I always set the defaults in my software to things the average user would want or expect, even if it differs from my own preferences -- if I want to change it, I can set it in my personal configuration. The way I see it, I'm just a software developer; I'm here to write useful code, not to inflict my way of doing things on the world.
Staying away from the google monster when reasonably possible seems most consistent with the Dillo project's objectives. (To be fair, on one hand, the duckduckgo guy talks, in the context of that project, a lot about privacy and such, but when I've read about what he's been up to in general, he sounds kind of rah-rah-hacker-news-let's-sell- a-start-up-to-a-huge-company-and-be-rich, like he knows the talk to talk.)
On Sat, 30 Jul 2011 17:18:53 -0400, corvid <corvid@lavabit.com> wrote:
Staying away from the google monster when reasonably possible seems most consistent with the Dillo project's objectives.
(To be fair, on one hand, the duckduckgo guy talks, in the context of that project, a lot about privacy and such, but when I've read about what he's been up to in general, he sounds kind of rah-rah-hacker-news-let's-sell- a-start-up-to-a-huge-company-and-be-rich, like he knows the talk to talk.)
You might consider Scroogle (http://www.scroogle.org), which doesn't filter results and isn't a buzzword-savvy startup. As an added bonus, it uses simple HTML code that renders quite nicely in Dillo. :-) Frankly, though, I think a lot of it is just paranoia. Google does push the boundaries of what's acceptable from time to time, and I maintain a healthy skepticism, but I don't outright distrust them. Anyway, at least they're established, unlike duckduckgo: as the saying goes, you don't trade the devil you know for the devil you don't know. ~Benjamin
participants (4)
-
corvid@lavabit.com
-
Hole.destructor@gmx.de
-
jcid@dillo.org
-
obeythepenguin@gmail.com