October 2007 - Dillo-dev - lists.mailman3.com

Re: [Dillo-dev] Dillo FLTK2 alpha release for developers

by Andreas Kemnade

Hi Jorge, That are really great news. I had some compiling problems. in dillo source code src/IO/Makefile.am and dpi/Makefile.am needs a AM_CXXFLAGS = @LIBFLTK_CXXFLAGS@ Same for dw-testbed/test/Makefile.am The other problem is that I need to comment out the tablet support in fltk due to missing libXi. Ok, then I got it to compile finally for my ipaq with familar 0.8.4 on it. I runs with X-forwarding so that I get the window on my notebook display. But it segfaults without x forwarding: dillo_dns_init: Here we go! (threaded) Disabling cookies. bad font: variable Segmentation fault I think I have to check my font setup, Bitstream Vera seems to be available... Greetings Andreas Kemnade

Oct. 3, 2007

Re: [Dillo-dev] Re: Dillo FLTK2 alpha release for developers

by Alexander Becher

* Johannes Hofmann: >On Wed, Oct 03, 2007 at 02:11:56PM +0200, Alexander Becher wrote: >> I compiled dillo-fltk2 on NetBSD-3.1, implementing strndup was the only >> change needed. However, strndup needs to null-terminate its return >> value, and strncpy does not do this if the source string is longer than >> the destination, so: > >As calloc() initializes the allocated buffer with 0, this is not >needed, I think. Hm, right. I had used malloc() in my replacement function, but with calloc() it is indeed unnecessary. I have wrapped this in autoconf, see attached patch. I also made autoconf in dillo and dw-testbed check for the path to the perl binary and substitute it in dpidc and Doxyfile, respectively, see other patch. Actually, these patches can't simply be applied: first, dillo-f15/dpid/dpidc and dw-testbed-0.0.43j/Doxyfile have to be moved to *.in. Regards, Alexander -- PGP key available citator(a)jabber.ccc.de

Oct. 3, 2007

Re: [Dillo-dev] Re: Dillo FLTK2 alpha release for developers

by Johannes Hofmann

Hi Alexander, On Wed, Oct 03, 2007 at 02:11:56PM +0200, Alexander Becher wrote: > Hi, > > I compiled dillo-fltk2 on NetBSD-3.1, implementing strndup was the only > change needed. However, strndup needs to null-terminate its return > value, and strncpy does not do this if the source string is longer than > the destination, so: As calloc() initializes the allocated buffer with 0, this is not needed, I think. Johannes > > * Johannes Hofmann: > > > >* I added a own version of strndup() as it does not exist for > > DragonFly: > > > >char* strndup(const char *s, size_t size) { > > char *r = (char*) calloc(size + 1, 1); > > > > if (r) { > > strncpy(r, s, size); > r[size] = '\0'; /* null-terminate return value */ > > } > > > > return r; > >} > > Also, strndup is found in <string.h> only if _GNU_SOURCE was #define'd. > This should probably be done in configure, with something like > HAVE_STRNDUP, but I have no experience with autoconf. > > Regards, > Alexander > -- > PGP key available > > _______________________________________________ > Dillo-dev mailing list > Dillo-dev(a)dillo.org > http://lists.auriga.wearlab.de/cgi-bin/mailman/listinfo/dillo-dev

Oct. 3, 2007

Re: Dillo FLTK2 alpha release for developers

by Alexander Becher

Hi, I compiled dillo-fltk2 on NetBSD-3.1, implementing strndup was the only change needed. However, strndup needs to null-terminate its return value, and strncpy does not do this if the source string is longer than the destination, so: * Johannes Hofmann: > >* I added a own version of strndup() as it does not exist for > DragonFly: > >char* strndup(const char *s, size_t size) { > char *r = (char*) calloc(size + 1, 1); > > if (r) { > strncpy(r, s, size); r[size] = '\0'; /* null-terminate return value */ > } > > return r; >} Also, strndup is found in <string.h> only if _GNU_SOURCE was #define'd. This should probably be done in configure, with something like HAVE_STRNDUP, but I have no experience with autoconf. Regards, Alexander -- PGP key available

Oct. 3, 2007

Re: [Dillo-dev] Dillo FLTK2 alpha release for developers

by None

On Tue, Oct 02, 2007 at 10:23:19PM +0000, higuita wrote: > On Tue, 2 Oct 2007 16:42:16 -0400, wrote: > > -Wno-unused-parameter -MT ui.o -MD -MP -MF .deps/ui.Tpo -c -o ui.o ui.cc > > > ui.cc: In function 'void b1_cb(fltk::Widget*, void*)': > > > ui.cc:235: error: cast from 'void*' to 'int' loses precision > > > make[3]: *** [ui.o] Error 1 > > > > Try: > > - int bn = (int)cb_data; > > + int bn = (long)cb_data; > > > > If this works OK, the same may be used for C++ in Dw (i.e. default to > > long). > > Thanks, it works!! it allowed the build to continue > it gave several errors like this and i fix then the same way. > > I can extract a patch and send it if you want. Yes, please send your patch to check suspicious parts. -- Cheers Jorge.-

Oct. 2, 2007

https and OpenBSD

by None

Hi, > Dear Jorge & dillo fans, > > I can confirm this will run and build on OpenBSD-4.1 with minor > modifications: > > * remove the GL & glut targets from the fltk Makefile if you don't > need it Did you just "./configure --disable-gl"? Or: - DIRS = src $(LOCALIMAGES) images OpenGL fluid glut test + DIRS = src $(LOCALIMAGES) images fluid test else? > * implement a strndup() function in dw/selection.cc in the testbed > as OpenBSD doesn't have that in string.h Yes. > > I still have one question: > > I wanted to try out https (yes I know the README says it doesn't > work). And of course it did not work: It works bery basically. If you can see "https://www.rapidssl.com/test/rapidssl.htm", that's it. > > prefs: Using internal defaults. > dillo_dns_init: Here we go! (threaded) > Enabling cookies as from cookiesrc... > Nav_open_url: Url=>https://www.ccc.de< > [dpid]: dpi.c:70: a_Dpi_rd_dpi_socket_dir: fopen > Dpi_blocking_start_dpid: try 1 > [dpid]: No such file or directory > [dpid]: - /home/ckeen/.dillo/dpi_socket_dir > [dpid]: init_sockdir: The dpi_socket_dir file (null) does not exist > dpid started > a_Dpiapi_dialog: > dpip_tag: <cmd='dialog' msg='The remote certificate cannot be > verified (code 20)' alt1='Continue' alt2='Cancel' '> > mouse button 1 was pressed > a_UIcmd_stop() > > > Is the backward channel form dpis supposed to work? Yes. > If yes I will > look into it. If there is too much missing atm, I would be gladful > for a hint on where to start. The point here is that https is a "filter" dpi (i.e. one instance per request is spawned). The solution is to implement it as a "server" (e.g. like the file server). That way, when a https request has many embedded resources (i.e. many https requests to the same site), the server can make only one query and not flood the backward channel for dpis. With regard to the actual certificate-validating code, AFAIR much of it is already there, it needs some testing and enabling. It would be great if you can work on it. > Thank you very _very_ much for this nice programme and for bearing > with us! Thank you for offering your help. -- Cheers Jorge.-

Oct. 2, 2007

Re: [Dillo-dev] Dillo FLTK2 alpha release for developers

by higuita

On Tue, 2 Oct 2007 16:42:16 -0400, wrote: > -Wno-unused-parameter -MT ui.o -MD -MP -MF .deps/ui.Tpo -c -o ui.o ui.cc > > ui.cc: In function 'void b1_cb(fltk::Widget*, void*)': > > ui.cc:235: error: cast from 'void*' to 'int' loses precision > > make[3]: *** [ui.o] Error 1 > > Try: > - int bn = (int)cb_data; > + int bn = (long)cb_data; > > If this works OK, the same may be used for C++ in Dw (i.e. default to > long). Thanks, it works!! it allowed the build to continue it gave several errors like this and i fix then the same way. I can extract a patch and send it if you want. Now it starts, but when i try to load a page it crashes: (gdb) run Starting program: /usr/local/bin/dillo-fltk [Thread debugging using libthread_db enabled] [New Thread 47988767561568 (LWP 17385)] dillo_dns_init: Here we go! (threaded) Disabling cookies. Nav_open_url: Url=>http://localhost< [New Thread 1082132800 (LWP 17390)] Dns_server [0]: localhost is 0x77aef0 [Thread 1082132800 (LWP 17390) exited] Connecting to 127.0.0.1 Program received signal SIGSEGV, Segmentation fault. [Switching to Thread 47988767561568 (LWP 17385)] 0x00002ba54013bba0 in strlen () from /lib64/libc.so.6 (gdb) bt full #0 0x00002ba54013bba0 in strlen () from /lib64/libc.so.6 No symbol table info available. #1 0x00002ba54010bb9b in vfprintf () from /lib64/libc.so.6 No symbol table info available. #2 0x00002ba54012cd7a in vsnprintf () from /lib64/libc.so.6 No symbol table info available. #3 0x0000000000423b9a in dStr_vsprintfa (ds=0x77bf40, format=0x4a1798 "GET %s HTTP/1.0\r\n%sAccept-Charset: utf-8, iso-8859-1\r\nHost: %s%s\r\n%sUser-Agent: Dillo/%s\r\nCookie2: $Version=\"1\"\r\n%s\r\n", argp=0x7fff6c9976e0) at dlib.c:349 n = <value optimized out> n_sz = <value optimized out> #4 0x0000000000423c56 in dStr_sprintfa (ds=0x10, format=0x4a179c "%s HTTP/1.0\r\n%sAccept-Charset: utf-8, iso-8859-1\r\nHost: %s%s\r\n%sUser-Agent: Dillo/%s\r\nCookie2: $Version=\"1\"\r\n%s\r\n") at dlib.c:397 argp = {{gp_offset = 48, fp_offset = 48, overflow_arg_area = 0x7fff6c9977e0, reg_save_area = 0x7fff6c997700}} #5 0x0000000000424f43 in a_Http_make_query_str (url=0x77afe0, use_proxy=<value optimized out>) at http.c:205 str = <value optimized out> cookies = 0x77f940 "" s_port = (Dstr *) 0x77bf20 query = (Dstr *) 0x77bf40 full_path = (Dstr *) 0x77bf80 proxy_auth = (Dstr *) 0x77f900 #6 0x000000000042511d in a_Http_dns_cb (Status=<value optimized out>, addr_list=<value optimized out>, data=<value optimized out>) at http.c:244 status = <value optimized out> name = {sin_family = 2, sin_port = 20480, sin_addr = {s_addr = 16777343}, sin_zero = "\000\000\000\000\000\000\000"} dh = <value optimized out> socket_len = 1097963232 SKey = 1 #7 0x0000000000420482 in Dns_timeout_client (data=<value optimized out>) at dns.c:457 channel = 0 srv = (DnsServer *) 0x6db680 #8 0x000000000047349d in fltk::wait () No symbol table info available. #9 0x00000000004735e3 in fltk::run () No symbol table info available. #10 0x00002ba5400e38c4 in __libc_start_main () from /lib64/libc.so.6 No symbol table info available. #11 0x0000000000408c99 in _start () No symbol table info available. it crashes with any page -- 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

Oct. 2, 2007

[Johannes.Hofmann@gmx.de: [Johannes.Hofmann@gmx.de: Re: [Dillo-dev] Dillo FLTK2 alpha release for developers]]

by Jorge Arellano Cid

Hi, -- Cheers Jorge.-

Oct. 2, 2007

Re: [Dillo-dev] Dillo FLTK2 alpha release for developers

by None

On Tue, Oct 02, 2007 at 02:27:06AM +0000, higuita wrote: > Hi jorge and Sebastian > > Thanks for releasing this version!! > > On Sun, 30 Sep 2007 20:09:25 -0400, wrote: > > ----- > > FLTK2 > > ----- > > builds fine, its really a great lib!! > > > ---------- > > dw-0.0.43j > > ---------- > > err... failed to build: > > make[1]: Entering directory `/home/higuita/dillo/dw-testbed-0.0.43j/lout' > g++ -DPACKAGE_NAME=\"\" -DPACKAGE_TARNAME=\"\" -DPACKAGE_VERSION=\"\" -DPACKAGE_STRING=\"\" -DPACKAGE_BUGREPORT=\"\" -DPACKAGE=\"dw-testbed\" -DVERSION=\"0.0.43j\" -DSTDC_HEADERS=1 -DHAVE_SYS_TYPES_H=1 -DHAVE_SYS_STAT_H=1 -DHAVE_STDLIB_H=1 -DHAVE_STRING_H=1 -DHAVE_MEMORY_H=1 -DHAVE_STRINGS_H=1 -DHAVE_INTTYPES_H=1 -DHAVE_STDINT_H=1 -DHAVE_UNISTD_H=1 -DHAVE_FCNTL_H=1 -DHAVE_UNISTD_H=1 -DHAVE_SYS_UIO_H=1 -I. -g -O2 -D_REENTRANT -D_THREAD_SAFE -Wall -MT object.o -MD -MP -MF .deps/object.Tpo -c -o object.o object.cc > object.cc: In member function 'virtual int object::Pointer::hashValue()': > object.cc:118: error: cast from 'void*' to 'int' loses precision > object.cc: In member function 'virtual void object::Pointer::intoStringBuffer(misc::StringBuffer*)': > object.cc:132: error: cast from 'void*' to 'int' loses precision > make[1]: *** [object.o] Error 1 > make[1]: Leaving directory `/home/higuita/dillo/dw-testbed-0.0.43j/lout' > make: *** [all-recursive] Error 1 > > i'm using slamd64 (slackware clone in 64bit)... gcc (GCC) 4.1.2 > > > ---------- > > dillo f15j > > ---------- > > also fails, looks like its the same reason: > > make[3]: Entering directory `/home/higuita/dillo/dillo-f15/src' > g++ -DHAVE_CONFIG_H -I. -I.. -DDILLORC_SYS='"/usr/local/etc/dillorc"' -I/usr/local/include -I../../dw-testbed -I/usr/include/libpng12 -DPNG_NO_MMX_CODE -I/usr/local/include -I/usr/include/freetype2 -Wno-non-virtual-dtor -g -O2 -Wall -W -Wno-unused-parameter -MT ui.o -MD -MP -MF .deps/ui.Tpo -c -o ui.o ui.cc > ui.cc: In function 'void b1_cb(fltk::Widget*, void*)': > ui.cc:235: error: cast from 'void*' to 'int' loses precision > make[3]: *** [ui.o] Error 1 > > so my test stuck :) > > as i'm no programmer, any help is welcome... > > thanks for dillo Try: - int bn = (int)cb_data; + int bn = (long)cb_data; If this works OK, the same may be used for C++ in Dw (i.e. default to long). -- Cheers Jorge.-

Oct. 2, 2007

Re: [Dillo-dev] new unfinished code

by Diego .

2007/9/30, <>: > > Hi Diego, > > On Sat, Sep 29, 2007 at 08:20:21PM +0200, Diego . wrote: > > This start to seem like spam. Sorry. > > No need to apologize. > > This is quite on topic and a contribution to the project. > > The fact of having no active developers for some time is a > separate issue (as I presume not much answers had come on the > patch). By the web stats y see that people go to the page, but only see the index page. Next time i will put links to the sections. > I have upload to my page unfinished code for a prefs dpi and my first > > screenshot :). The patch do not change dillo code, but need dillo code > to > > compile. (dpi lib). > > > > The page is: > > http://personales.ya.com/darkspirit/ > > > > > > If i get time i will try to post some docs about mime code for > > dillo(unfinished, unpublished, break, etc) and questions about internals > of > > dillo. > > I my experience it is best if you have some code that works > cleanly and needs to be extended, rather than a big cloud that > needs more work. I agree, but i have that code since a lot of time and prefer to make public than forget it. Another pro is that the next time i work with the code only need to use the patch instead of search the diferents code trees. -- > Cheers > Jorge.- > > _______________________________________________ > Dillo-dev mailing list > Dillo-dev(a)dillo.org > http://lists.auriga.wearlab.de/cgi-bin/mailman/listinfo/dillo-dev >

Oct. 2, 2007