- Dillo-dev - lists.mailman3.com

Re: [Dillo-dev]why use "ccc chain" instead of calling the function directly?

by Jorge Arellano Cid

On Wed, 7 Apr 2004, [gb2312] Lei Jin wrote: > Thank you for your detail explanation. ACK. First of all, there's no need to CC me, I read dillo-dev! > I am still confused. Would you provide an example such as the > situation of the two threads doing network job(what happens > when using threads and mutexes and when using CCC chain)? Maybe > it will help to understand. When using threads and mutexes (pthreads), thread scheduling is done by a third thread or by the kernel (depends upon the implementation), and it is preemptive. Now, the address space may be shared, so the need for mutexes arise, and because of both reasons, performance is degraded. Using a pseudo-thread (CCC thread), doesn't require scheduling by an external process or by the kernel. All is done in the same process, but there's no preemption. (If you ever have heard that unthreaded servers are faster than threaded counterparts, you'll find the same reasons behind.) Now, since CCC-threads are not preemtive, Dillo also makes use of pthreads when it needs this ability (for instance for DNS resolving). As all the CCC-threads run in the same process, it's easy to sync with pthreads with a polling wait and an event. Cheers Jorge.-

April 7, 2004

[Dillo-dev]Re: New dillo user (contributor?)

by January Weiner

> least a rudimentary man page should be in each programs distribution. Is > there any document available on Dillos shortcuts and configuration, or > should I download the sources right away? Ok, ok, my fault, I got it. j. -- ------------ January Weiner 3 ---------------------+--------------- Division of Bioinformatics, University of Muenster | Schlo�platz 4 (+49)(251)21634 | D48149 M�nster http://www.uni-muenster.de/Biologie.Botanik/ebb/ | Germany

April 7, 2004

[Dillo-dev]New dillo user (contributor?)

by January Weiner

Hello, first of all, let me thank you for this great piece of software. It's fast, it's simple, it keeps up good Unix traditions -- small, efficient programs which do one job. I have used dillo a couple of years ago, but it was a little too brittle and not mature enough for me, especially due to the lack of tables (or was it CSS?) -- anyway, it is very nice now[*]. A thing that's really lacking (at least for me) -- is a.. man page. I'm an unbelievably traditional when it comes to man pages, mostly because of the apropos database[#]. Further, an introductory readme page for users (with shortcuts etc., and not necessarily compilator options) would also be a nice thing to have, I think. How about that -- if you do not have anything principally against man pages / readme's for newbie's, I'll write it, and promise to keep up with Dillo's releases for the next 12 months. I do not have much time, but at least a rudimentary man page should be in each programs distribution. Is there any document available on Dillos shortcuts and configuration, or should I download the sources right away? Regards, January P.S. A nice detail: when running dillo for the first time, make it copy the dillorc into ~/.dillo/. This is the first place a literate Unix user will look for a configuration when he or she notices that Dillo created ~/.dillo/. Tested on three individuals :-) [*] the only thing I really miss is the support for ISO-8859-2. [#] besides, it took me a couple of minutes to figure out that "s <keyword>" is not a thing you should type in the address widget, but an abbreviation for "ctrl-s keyword", esp. because ctrl-s is such a widespread shortcut for "save". -- ------------ Dr January Weiner 3 ------------------+--------------- Division of Bioinformatics, University of Muenster | Schlo�platz 4 (+49)(251)21634 | D48149 M�nster http://www.uni-muenster.de/Biologie.Botanik/ebb/ | Germany

April 7, 2004

Re: [Dillo-dev]Disable toolbars

by Jorge Arellano Cid

On Sun, 4 Apr 2004, Florian Angehrn wrote: > if i don't want both toolbars, what do i have to change in the sourcecode? > > many thanks for your great product (and an answer :)! Hmmm, you should be able to find this by yourself, but... Go to Interface_browser_window_customize() and set Interface_toggle_panel() to do what you want. Cheers Jorge.-

April 7, 2004

Re: [Dillo-dev]Can't start dpid

by Jorge Arellano Cid

On Sat, 3 Apr 2004, Barry Kauler wrote: > Hi, I'm not on the list, so hope this can be submitted! > Kindly also reply a copy to bk(a)goosee.com > > I have been following Dillo for sometime, as a user, and recently > upgraded to v0.8.0. > > Problem is, when I click on a ftp url, the message: > > "Error: can't start dpid daemon" > > So, I tried to start dpid directly on the commandline: > > # dpid > # dpid.c:352: register_all: fatal error: > -can't find the directory for dpis > > So, is there something missing here? -- what is dpis? > Appreciate any advice on this! <q> ------------ Dpi programs ------------ These are installed by "make install". If you don't have root access, copy "dillo" and "dpid" to some directory in your path and install the dpis by running "./install-dpi-local" from the top directory (they will be installed under ~/.dillo). </q> (Extracted from the README file! ;-) Cheers Jorge.- PS: Now that you've tried manually, it'd be wise to assert there's no running dpid process when you retry: ps aux | grep dpi [kill dpid processes if present]

April 7, 2004

Re: [Dillo-dev]why use "ccc chain" instead of calling the function directly?

by Lei Jin

Thank you for your detail explanation. I am still confused. Would you provide an example such as the situation of the two threads doing network job(what happens when using threads and mutexes and when using CCC chain)? Maybe it will help to understand. Jorge Arellano Cid <jcid(a)dillo.org> wrote: On Sat, 3 Apr 2004, [gb2312] Lei Jin wrote: > Use "ccc chain" mechanism makes the code seeming uniform. But > it cause the source code difficult to understand because lots > of indirect function call and variety of parameters. The other way is to use threads and mutexes (much more complex and a bit slower too). > I wonder whether we can call the target function directly > instead of using "ccc chain", but I do not know if there is > difference between these two methods ??? The CCC is not just a way of making function calls, is a theory that allows for implementing non-preemptive parallelism without threads or forks. All the networking and error control in Dillo is driven by these pseudo-threads each shaped as a dynamic CCC. There can be even hundreds of them running at the same time (depending on the page, and on how many pages are loading in the background). Each thread doing network jobs can linger for a long time (generally measured in seconds, depending on latency, network load, size of the transfer, bandwidth etc) --that explains the need for a parallel approach--, now, each one of those threads needs independent error control and recovery, and it may have several steps to perform, so it also has to keep track of its state. That's the job of the CCC. It looks like contrived complexity, for sure, but that's mainly because people don't even start to realize that it handles parallelism in a single address space. This is a bit of my fault, I created the theory while working on Dillo, but I've never had the time to write it down and present it in abstract. Priorities always stood in the way. To my relief, it has worked rock solid, flexible enough to even account for plugins, and has needed low maintenance. Beleieve it or not, when things arrive to this state, the Jargon file regards them as "Deep magic". Cheers Jorge.- _______________________________________________ Dillo-dev mailing list Dillo-dev(a)lists.auriga.wearlab.de http://lists.auriga.wearlab.de/cgi-bin/mailman/listinfo/dillo-dev --------------------------------- Do You Yahoo!? 惠普TT游戏剧，玩游戏，中大奖！

April 7, 2004

Re: [Dillo-dev]progress on AIX? (fwd)

by Jorge Arellano Cid

---------- Forwarded message ---------- Date: Tue, 06 Apr 2004 22:29:57 +0200 From: Riccardo Mottola <rollei(a)tiscalinet.it> To: Jorge Arellano Cid <jcid(a)dillo.org> Subject: Re: [Dillo-dev]progress on AIX? Jorge Arellano Cid wrote: > > On Thu, 4 Mar 2004, Riccardo wrote: > > > > [...] > > > Please try changing these to int. This is: > > > > > > #define socklen_t int > > I did that right now and recompiled. Only a bunch of files recompiled. > > > > I still get the same SIGILL, illegal instruction. > > Does current CVS produce SIGILL too? This time, boostrapping and configuring was like a charm. but I get a SIGILL anyway... And no usable bt, as always. Program received signal SIGILL, Illegal instruction. 0x0 in ?? () from /usr/local/build/dillo-cvs/dillo/src/dillo (gdb) bt #0 0x0 in ?? () from /usr/local/build/dillo-cvs/dillo/src/dillo clueless. I remember there was a german guy on the list who used AIX. -R

April 7, 2004

Re: [Dillo-dev]progress on AIX?

by Jorge Arellano Cid

On Thu, 4 Mar 2004, Riccardo wrote: > > [...] > > Please try changing these to int. This is: > > > > #define socklen_t int > I did that right now and recompiled. Only a bunch of files recompiled. > > I still get the same SIGILL, illegal instruction. Does current CVS produce SIGILL too? > > > > The other warnings worth of considering at this point are: > > > > <q> > > ld: 0711-224 WARNING: Duplicate symbol: .strcmp > > ld: 0711-224 WARNING: Duplicate symbol: strcmp > > ld: 0711-224 WARNING: Duplicate symbol: .pow > > ld: 0711-345 Use the -bloadmap or -bnoquiet option to obtain more > > information. > > </q> > I stil have no clue about that. Wasn't here on the list someone with > more AIX experience? I don't have any AIX experience, but google searching showed that sometimes the LANG variable has incidence: http://ferret.wrc.noaa.gov/Ferret/FAQ/system/aix_setenv_LANG.html Other than that, I don't know... :( Cheers Jorge.-

April 6, 2004

Re: [Dillo-dev]About proxy authorization && password storing

by Jorge Arellano Cid

On Fri, 2 Apr 2004, Ivan Daniluk wrote: > Hi Jorge. Hi! > > > Maybe, we should open popup dialogue and ask user for > > > login/password? > > > > The login may be specified in dillorc, and the password asked > > in a popup dialog (whenever there's a login in dillorc :). > > > > This adds the burden of entering the password, but keeps things > > a bit more private. > I've implement this burden :), but stuck in a problem and need advice > and/or suggestion from you. > First, look to the attached current patch for stable dillo 0.8.0 > release. Oh, I haven't yet. Anyway, here's the answer. > Main problem for me is the way to stopping the query, waiting > until password will be entered or canceled, and resuming it. > When we call a_Interface_proxy_passwd_dialog(), the dialog window pops > up, and query goes forward, without being formed with proxy password. > So, I see two ways: > 1) to stop the query forming(in Http_get() function) and wait, while > proxy_passwd_dialog_window will be destroyed. > 2) to abort the query, and recall it when password will be entered. > > I'm a newbie in gtk programming, and can't figure out how to graceful > open dialog window and wait until it'll be closed. If it's possible, > than imho it's the best way. > About second choice.. There are complexity for me in a_Http_ccc stuff. > Sure, I'll win this problem, but it will take more time. And really > want to hear your thoughts about it. Use the third alternative! :-) AFAIS the simplest way to to it is to ask for the proxy password when dillo starts, that is, before any query is made, probably after the splash screen is displayed. That way, things would reach the same simplicity as in the first patch! > One more question around it. > Password dialog window is activating in case if there is option > "http_proxyuser" in config, but not when http query gets "Cache Access > Denied" error. > Is this way ok? AFAIU, yes. The proxy login/password pair is one thing, and HTTP's simple authentication mechanism (section 11) is another. It may happen that you reach a realm/password protected page through a login/password authenticated proxy. Now, if after this patch, you want to keep on working on basic authentication, be sure to contact Madis (AFAIR he once worked on it), and I can make the CCC bindings to stop/resume/forward the querying process. > And, the last note on patch. It's about following lines in IO/http.c: > + if ( strchr(prefs.http_proxyuser, ':') ) > + { > + DEBUG_MSG(5, "Storing proxy password in dillorc is insecure!\n"); > + ask_proxy_passwd = FALSE; > + HTTP_Proxy_Auth_base64 = a_Misc_encode_base64(prefs.http_proxyuser); > + } > It allows user to store in config file not only proxy user name, but > pair "user:password" as well. I've done it's mainly for convenience, > but of cource we can remove it from patch. Hmmm, this is one I'm not yet sure of. On one hand, it is very handy not to have to enter the password every time, and on the other, stealing the proxy's password becomes quite simple (for instance, a bathroom break and it's gone! ;). [minutes thinking...] It's not a thing I'd recommend. Now, as Dillo project is explicitly concerned about security, it can't do otherwise. Perhaps you may leave that portion of code commented out, so the responsibility is left to the one that decides to enable it. I'd like to hear comments about this. Specially from sysadmins. > Your suggestion for me are welcome. Thanks. Cheers Jorge.-

April 6, 2004

Re: [Dillo-dev]why use "ccc chain" instead of calling the function directly?

by Jorge Arellano Cid

On Sat, 3 Apr 2004, [gb2312] Lei Jin wrote: > Use "ccc chain" mechanism makes the code seeming uniform. But > it cause the source code difficult to understand because lots > of indirect function call and variety of parameters. The other way is to use threads and mutexes (much more complex and a bit slower too). > I wonder whether we can call the target function directly > instead of using "ccc chain", but I do not know if there is > difference between these two methods ??? The CCC is not just a way of making function calls, is a theory that allows for implementing non-preemptive parallelism without threads or forks. All the networking and error control in Dillo is driven by these pseudo-threads each shaped as a dynamic CCC. There can be even hundreds of them running at the same time (depending on the page, and on how many pages are loading in the background). Each thread doing network jobs can linger for a long time (generally measured in seconds, depending on latency, network load, size of the transfer, bandwidth etc) --that explains the need for a parallel approach--, now, each one of those threads needs independent error control and recovery, and it may have several steps to perform, so it also has to keep track of its state. That's the job of the CCC. It looks like contrived complexity, for sure, but that's mainly because people don't even start to realize that it handles parallelism in a single address space. This is a bit of my fault, I created the theory while working on Dillo, but I've never had the time to write it down and present it in abstract. Priorities always stood in the way. To my relief, it has worked rock solid, flexible enough to even account for plugins, and has needed low maintenance. Beleieve it or not, when things arrive to this state, the Jargon file regards them as "Deep magic". Cheers Jorge.-

April 6, 2004