November 2003 - Dillo-dev - lists.mailman3.com

patched dillo via Arch
by Paul Pelzl 26 Nov '03

26 Nov '03

If anyone on the list is using GNU Arch, you can obtain: Dillo CVS + Frank's tab/frame/kbnav/etc. patch v16 + Indan's HTTPS patch v2 + my toggle-force-my-colors patch integrated into branch dillo--tabframe-pelzl--16.0 . The archive is: pelzlpj(a)eecs.umich.edu--2003 ftp://pelzlpj.dyndns.org/pub/paul/tla/2003 If you find this useful, please send me a private email. If no one cares, I won't bother to announce this archive in the future. Paul

2 1

Re: [Dillo-dev] patched dillo via Arch
by Frank de Lange 26 Nov '03

26 Nov '03

Paul Pelzl wrote: >I wouldn't mind setting up a patches list, but only if there is demand >for it. Is patch discussion creating too much traffic on the wearlab >mailing list? If not, then it seems like keeping patch discussion on >the list would give the patches more exposure. > >Anyone else have an opinion? > >Paul > > > I think that patch traffic belongs on this list, as it is quite often relevant for all of Dillo (patched or unpatched). It also keeps the patches in sight for users/developers of unpatched Dillo, which should generate exposure. If/when the patches really take on a life of theirselves it is time to move, but that time has not arrived as far as I'm concerned. Good initiative btw, the Archive. Cheers//Frank -- WWWWW ________________________ ## o o\ / Frank de Lange \ }# \| / +46-734352015 \ \ `--| _/ <Hacker for Hire> \ `---' \ +31-640037120 / \ frank(a)unternet.org / `------------------------' [ "Omnis enim res, quae dando non deficit, dum habetur et non datur, nondum habetur, quomodo habenda est." ]

1 0

kbnav bugs anyone?
by Frank de Lange 25 Nov '03

25 Nov '03

Hi'all, I just noticed that kbnav does not do the right thing when tabbing out of a mouse-selected form field. Instead of going to the next field, it starts at the first form field/button on the page. I'll resolve this for the next patch. If anyone has any other problems with kbnav (or tabs or frames, but that is another patch...) let me know. I'd like to resolve problems in the current feature set before adding anything new... Cheers//Frank -- WWWWW ________________________ ## o o\ / Frank de Lange \ }# \| / +46-734352015 \ \ `--| _/ <Hacker for Hire> \ `---' \ +31-640037120 / \ frank(a)unternet.org / `------------------------' [ "Omnis enim res, quae dando non deficit, dum habetur et non datur, nondum habetur, quomodo habenda est." ]

1 0

[PATCH] HTTPS version 2.
by Indan Zupancic 25 Nov '03

25 Nov '03

Hi, I updated my https patch, can get it from: http://www.xs4all.nl/~dorinek/dillo/dillo-https.patch.gz I will update the installer tomorrow. Not much time, so here the readme text, which can be found at http://www.xs4all.nl/~dorinek/dillo/readme.txt: ------------------------------------------------------------------------ Last updated: 24 November 2003 This patch adds https support to the Dillo CVS version (Pre 0.80). FILES: * Latest https patch is called dillo-https.patch.gz. BUGS: * Dillo will crash if you reload a https webpage very fast multiple times. But you can't do that because I was so mean to edit the refresh function so that you can't reload faster than once per second, which makes no sense anyway. With fast reloads I mean clicking absurdly fast: I can't do it while holding my mouse in a normal way. Keeping Ctrl+R pressed is too though fast though: It aborts the previous reload before it can make a connection. TODO: * Proper proxy support. Tell me if you need it. Now proxy may work only if it uses https with the proxy port and if https is just http over ssl with proxies too (which is the case with the only proxy I know). * Fancy stuff like auto ACCEPT_SESSION for cookies for https connections, auto refresh every 15 minutes or so to keep the ssl connection alive, etc. (things to make Dillo work better with webmail) * Realy fix the weird bug. I tried it, and I gave up for now. STATS: With CFLAGS="-O2 -s" (Os is much slower than O2), gcc 3.2.2 347K, 355368 bytes: unpatched 344K, 351816 bytes: patched, configure --disable-https 346K, 353928 bytes: patched Yes, Dillo becomes smaller after my patch. I cleaned up IO.c a lot. CHANGELOG: - Synchronized to the current CVS. - Added better proxy handling: no idea if proxy works or not though. - Made it harder to crash: needs more fast reloads now (~10 instead of 3). - Prevented fast reloads so that Dillo can't crash anymore. :p - Cleanup of IO.c (smaller, faster, better). - Cleanup of all the https code (bigger, faster, better). - Proper autoconfiguring thanks to Paul Pelzl's patch. Thanks Paul. :) Enjoy, Indan P.S. If you reply, please don't include the whole email you reply to, only the parts you're replying to, or nothing at all. If everyone quotes everything every time you'll get huge emails very fast...

2 1

Proposal: United Dillo
by Indan Zupancic 24 Nov '03

24 Nov '03

Hello, With "united Dillo" I mean that there is always just one Dillo running per user, so when you start Dillo while it is already running, then the current one will open a new window. I made a patch as proof of concept, see: http://www.xs4all.nl/~dorinek/dillo This approach has two big advantages: 1) Always one Dillo has less overhead than running multiple instances at the same time. 2) Automatic synchronization/no need for synchronization anymore. First advantage is in the spirit of Dillo, wanting to be small, light and fast. Second advantage means that doing certain things will be much simpler, like having always cookies, no need to synchronize plugins, meaning that dpid is not really needed anymore, because most of the points from the problem list are solved (see Dpid.txt). Problems that aren't solved: * Forks can be expensive (Dillo had to fork its dpis). But dpid does much more than simply forking: it forks and then starts up an external program (which often also forks and starts up wget). Beside, if Dillo uses library plugins then there is no need to fork at all. * It increments dillo's core size. The patch adds about 70 lines of code to dillo.c and increases dillo with about 1 K. It is true that if you get rid of dpid then more code is needed in Dillo's core. But that's only moving the code to some other place, no real increase if you ask me. Having a seperate dpid has much more overhead at runtime (seperate process versus bigger executable). You can always make a plugin handling library, then it isn't part of Dillo's core. Of course a united Dillo has problems/disadvantages too: It is sometimes still needed to synchronize multiple windows. If it's not implemented carefully, you may lose functionality (aka command line options). A user can't run more than one Dillo instance. Sometimes it's wanted to run multiple Dillo's, for instance two different versions. But that isn't possible at all, or if version awareness is added, without losing synchronization. What do you think about it? Greetings, Indan

6 13

[PATCH] Version 5 of kbnav patch is ready
by Frank de Lange 24 Nov '03

24 Nov '03

Hi'all, The next version (5) of my keyboard navigation patch is ready. See the end of this message for download instructions. Do note that the tab/frame patch already contains this functionality, so there is no need (and no way) to apply this patch on top of that or the other way around. This release does not add any new features, it is mainly a bug fix and CVS resync release. For more info on what this patch does, see the doc/Keyboard_Navigation.txt document (after applying the patch, naturally...) Fixed bugs ---------- - erroneously highlighted link after leaving a form element, was caused by wrong function call order in focus_next_link. Fixed. CVS date/time ============= The patch is made against CVS as it was on 20031124@13:18 GMT. It should apply against CVS as it was on 20031121(a)21.33 GMT and later. For those who can not or do not want to use CVS, a snapshot can be found here: http://www.geocities.com/ikbenfrank/dillo-200311241318.tar.gz Download instructions ===================== There is currently only one version of the patch, made against 2003124@13:18GMT CVS. As always, have a look at the patch website (http://www.geocities.com/ikbenfrank/) for the latest version of the patch. This patch does NOT apply on top of the tab/frame patch, and the tab/frame patch does not apply on top of this one! The tab/frame patch already contains this functionality, so it is not needed. http://www.geocities.com/ikbenfrank/dillo-200311241458-kbnav.patch.gz size: 30633 bytes (gzipped, as downloaded) 152984 bytes (uncompressed) md5sum: 0ed60695cab750669f3d50c201bdf217 (gzipped, as downloaded) 20aeb6b96e551cd8f4fa65ac5d74ef9e (uncompressed) Cheers//Frank -- WWWWW ________________________ ## o o\ / Frank de Lange \ }# \| / +46-734352015 \ \ `--| _/ <Hacker for Hire> \ `---' \ +31-640037120 / \ frank(a)unternet.org / `------------------------' [ "Omnis enim res, quae dando non deficit, dum habetur et non datur, nondum habetur, quomodo habenda est." ]

1 0

RPM Latest Frame tabs patch with basic authentication:
by Ronald W. Henderson 24 Nov '03

24 Nov '03

Folks: Could someone direct me to a location that has a RPM built with the latest Frame and Tabs patch but also basic authenitcation code. I find dillo as an excellent browser but the lack of basic authentication as a standard built in feature limits my use. Today, many networking Vendor's support a web interface for administation of their product and typically require basic authentication for login. I would like to use dillo as my browser for administration purposes. I appreciate your support on this excellent browser! ---Ron Henderson

1 0

Re: [Dillo-dev] [PATCH] Version 15 of tab/frame/kbnav/etc patch, now with keywords and more...
by Frank de Lange 24 Nov '03

24 Nov '03

Paul Pelzl wrote: > Now that I look at the problem again, I appear to have misdiagnosed it. > >It looks like the problem is with multiple spaces between keywords, not >with newlines. I've attached an example that causes a segfault after >entering any URL in the location bar. > >keywords = "s:http://www.google.com/search?q=%s ~:file:~" > > OK, solved this one as well. g_strsplit creates an empty string when two or more consecutive split-characters are encountered (quite reasonable behaviour, but not what I want for keywords) and the keywords processing code was not prepared for empty keywords (this in constrast to keywords with empty URLs which were handled OK). Is fixed in my code now, I'll release a new version of the patch with all recent fixes, made against the most recent version of CVS today or tomorrow. Cheers//Frank -- WWWWW ________________________ ## o o\ / Frank de Lange \ }# \| / +46-734352015 \ \ `--| _/ <Hacker for Hire> \ `---' \ +31-640037120 / \ frank(a)unternet.org / `------------------------' [ "Omnis enim res, quae dando non deficit, dum habetur et non datur, nondum habetur, quomodo habenda est." ]

1 0

Sigh... CVS out of sync 5 minutes after I upload my patches
by Frank de Lange 23 Nov '03

23 Nov '03

It is as if the devil plays with it... 5 minutes after I upload my latest patches, CVS gets updated and the patch does not apply anymore. Well, this time I was prepared for that. Anyone who wants to try tab/frame patch no. 15 can download the CVS source on which it is based from the patch site: http://www.geocities.com/ikbenfrank/dillo-200311191409.tar.gz This is the UNPATCHED CVS code that was used to create the patch. To get a working tabbed/frames/etc browser, download this tarball and the patch, then apply the patch. From now on I will always include the CVS code on which the patch was based to get around these irritating synchronisation problems. Indan, if you want to update your HTTPS patch, may I suggest you use this source code tarball instead of current Dillo CVS? This way we are less dependent on the state of CVS... Cheers//Frank -- WWWWW ________________________ ## o o\ / Frank de Lange \ }# \| / +46-734352015 \ \ `--| _/ <Hacker for Hire> \ `---' \ +31-640037120 / \ frank(a)unternet.org / `------------------------' [ "Omnis enim res, quae dando non deficit, dum habetur et non datur, nondum habetur, quomodo habenda est." ]

3 4

latest commit: cleanup on glib ids
by Jorge Arellano Cid 21 Nov '03

21 Nov '03

Hi Sebastian! While making the latest patch (now in CVS), I finally had the time to track gtk_timeout_add, gtk_idle_add and g_io_add_watch inside gtk/gdk/glib sources. The interesting part is that after the related wrappers, they all end using g_source_add; this function uses hook_id (ghook.h) which gets the id from seq_id. :-) In short: seq_id starts from 1 and hook_id is largely used with zero as a non-used/non-active/non-valid value! I updated dillo accordingly and eliminated the uneasy feel of initializing a guint with -1 as an invalid value. Regards Jorge.-

1 0