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

Re: [Dillo-dev] [PATCH] Version 15 of tab/frame/kbnav/etc patch, now with keywords and more...

by Frank de Lange

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." ]

Nov. 24, 2003

Re: [Dillo-dev] Proposal: United Dillo

by Stephan Goetter

Am Thu, Nov 20, 2003 at 10:59:47PM +0200 schrieb Jyri Jokinen und sprach: > On Thu, Nov 20, 2003 at 08:48:56PM +0100, Indan Zupancic wrote: > > > > I think it's the stupidest idea since browser tabs. (*) > > Thanks for your (honest and to the point) oppinion, but really, 50 windows > > open, and you hate tabs? > > I said "browser tabs". Window manager level window grouping is ok, > and renders browser tabs useless. That was/is the point. > > I don't like seeing code duplication, and certainly not somewhere it > doesn't belong. Web browsers should help the user with web browsing, > window managers with window managing. > I'm using fluxbox as well and still do use dillo-tabs heavily. Why? If you want to read 30 news messages in a seperate window (fast open- and closeable) I would have 30 additional window-manager tabs. With Frank's patch I do just open one window per news-site and every message in a new sub-tab. My open window-tabs still get clear i.e. <20 instead of having 50 mixed. So the can-be-done-with-window-manager-as-well argument doesn't really exist for me. Stephan

Nov. 24, 2003

gcc 2.95 doesn't know about -Wno-unused-parameter

by David McKee

Jorge, I suppose you're getting tired of hearing about my ancient Redhat system, but my gcc doesn't believe in -Wno-unused-parameters, preferning "-Wno-unused". Easy enough to fix by hand. Perhaps I should look nto configure in my copious spare time. Thanks, -- -- David McKee -- dmckee(a)jlab.org -- (757) 269-7492 (Office)

Nov. 24, 2003

Re: [Dillo-dev] [PATCH] Version 15 of tab/frame/kbnav/etc patch, now with keywords and more...

by Frank de Lange

Paul Pelzl wrote: >On Fri, Nov 21, 2003 at 04:46:31PM +0100, Frank de Lange wrote: > > >>The next version (15) of my tab/frame/keyboard navigation patch is >>ready (see bottom of this message for download instructions). >> >> > >I've got a couple of early bugs to report: > >1) when using kbnav to move from a link to an entry box in a form, the > most recently selected link is not properly deselected. Example: > google.com. > > OK, solved this one. It was not related to the reported rendering problem with Dw (which still exists), but to the new focus_in/focus_out handling code which unhighlights a focused link when the viewport loses focus, and highlights it when it regains focus. Fix will be included in next patch... Thanks for the bug report! 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." ]

Nov. 23, 2003

Re: [Dillo-dev] Sigh... CVS out of sync 5 minutes after I upload my patches

by Indan Zupancic

> 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... I use the -D option in my install script, and I rewrote the thing so I could just give the date at the top and it will download and install the correct CVS and patches, but now that won't work... To make matters worse, I also updated my https code to the current CVS (which didn't change for two days..). I would appreciate it a lot if you make a patch against the current state of CVS and give the file the usual name. I use FILE1="dillo-${YEAR}${MONTH}${DAY}-tabs.patch.gz" in my script, so it's very easy to update when there's a new version. I could add hours and minutes, but I'd rather leave those out, and when you make a updated version, I also don't have to "downdate" my code/patch. Another possibility is that you make a link named dillo-$DATE-tabs.patch.gz where DATE is the date of the Dillo CVS version that works with your patch, and points to the dillo-$RELEASE_DATE-tabs.patch.gz file. You don't have to host the Dillo source you used, not for me anyway, I won't use it, all I need to know is the date of the CVS you used, and sometimes the time too ;-). I think the install script causes more exposure, because it makes it very easy to get a patched Dillo; people don't even have to know what CVS is. The new https patch will be out soon, maybe even today, so you may expect a new installer soon. Greetings, Indan

Nov. 23, 2003

Re: [Dillo-dev] Sigh... CVS out of sync 5 minutes after I upload my patches

by Frank de Lange

Indan Zupancic wrote: >>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. >> >> > >Oh so funny. :) > >I saw you updated your patch today, and the first thing I thought was: Why >didn't Frank say that he used the CVS version of yesterday, because CVS >didn't change for two days, and _if_ CVS would be updated today, there >will be problems. > > That is what the website is for. The name on the patch is the date it is released, the CVS date is the last date CVS is changed. The patch is always based on CVS for the release date (right before I release a patch I update CVS and reconcile any differences, if the differences are big I delay the patch to test the changes), but CVS is not always updated on that day (as you noticed). FYI, the CVS date/time I give is the date/time stamp for the src/ directory. Using CVS features is nice, but have you noticed how many (or few...) people even know how to get source from CVS in the first place? I want the patch to get exposure to as many people as possible, and including the source on which it is based is one way of doing that. Another way would be to release a patched source tree, but I prefer the patch + clean approach. >Could you reply to the CVS offer from Trent Jarvi and say what you think >about a CVS branch? See: > >http://lists.auriga.wearlab.de/pipermail/dillo-dev/2003-November/001470.html > > I noticed that posting, have been busy lately hence did not post reply. As there are only two active patch developers at this moment I think we do not need a CVS branch. It should be possible to synchronise releases for two patches. Of course, having a branch on the main Dillo CVS or elsewhere would not hurt, and I do not oppose it. I just think we can do without it for the moment. 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." ]

Nov. 23, 2003

Re: [Dillo-dev] [PATCH] Version 15 of tab/frame/kbnav/etc patch, now with keywords and more...

by Frank de Lange

Paul Pelzl wrote: >On Fri, Nov 21, 2003 at 04:46:31PM +0100, Frank de Lange wrote: > > >>The next version (15) of my tab/frame/keyboard navigation patch is >>ready (see bottom of this message for download instructions). >> >> > >I've got a couple of early bugs to report: > >1) when using kbnav to move from a link to an entry box in a form, the > most recently selected link is not properly deselected. Example: > google.com. > > If this is what I think it is, it actually seems to be a widget bug. I've discussed this with Sebastian before. Relevant part of discussion plus link to bug description: >Another corner case (to be tried with the new patch): >> >> - highlight a link which is followed by an embedded GTK+ widget, >> so that the embedded widget is just outside the visible part of the >> document. >> - move focus to the embedded widget. The widget scrolls into view and >> is focused, >> as it should. >> - but... the link still appeats highlighted. This even though the >> unhighlight >> function is called. >> - cover the 'highlighted' link with another window. When it is >> uncovered, the >> highlight is gone - as it should. >> >> It appears the page is not redrawn correctly when focus is handed over >> to an embedded GTK+ widget. > > I'm not sure, but I assume that this is the another symptom of a known bug (#202). More on this bug at http://www.geocrawler.com/archives/3/702/2002/3/50/8045756/ It might be something else though (I've noticed this many times, and just thought of it as the described widget bug) so I'll have a look at it. >2) the "keywords" user pref is apparently not read in a robust manner. > If I separate keywords by a newline instead of a space, I can > generate a segfault upon typing any URL in the location bar. > Hmmm, I'll have a look at that... 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." ]

Nov. 23, 2003

Re: [Dillo-dev] Confused

by Indan Zupancic

Hi, > Can someone tell me where to download a source with the https, tab patches > without having to track down what and or where to patch? Or send me a > copy:) thanks Currently I have a sort of installer that downloads Dillo, the tabs and https patches, and applies them, Can get it from: http://www.xs4all.nl/~dorinek/dillo/install-dillo.sh But it isn't finished yet, nor up to date (Frank has a newer tabs+frames patch). I have a better installer, but I need to get my https patch working first. It stopped working after I changed it a bit :(. There is a weird bug that I can't seem to find, will hopefully fix it today and then update my https patch and the installer. Greetings, Indan

Nov. 23, 2003

Confused

by Donald McKnight

Can someone tell me where to download a source with the https, tab patches without having to track down what and or where to patch? Or send me a copy:) thanks Thanks Donald ----- Original Message ----- From: "Indan Zupancic" <indan(a)nul.nu> To: <dillo-dev(a)dillo.org> Sent: Saturday, November 22, 2003 10:18 AM Subject: [Dillo-dev] [patch] Cleanup of IO.c > Hi, > > Just cleaned IO.c up, so nothing fundamental changed. Besides some > cleanups in multiple functions this are the two major changes: > > Removal of a_IO_write_chunks, mainly because it isn't used anywhere and it > doesn't add any real functionality, it does the same only in a different > way. Thanks to CVS it is easy to add it back when you really needed, if > you want. > > Removal of the ValidIOs list. This is the logical next step after > g_source_remove was discovered, because the extra indirection isn't needed > anymore. > > All in all the result is a much smaller IO.c (400 lines instead of 700), > smaller Dillo (about 3K with -O2), and O(1) behaviour instead of O(n) in > the IO engine. > > The changes have really no negative points as far as I can tell. > > Link to the patch: > > http://www.xs4all.nl/~dorinek/dillo/IO-cleanup.patch.gz > > Greetings, > > Indan > > > _______________________________________________ > Dillo-dev mailing list > Dillo-dev(a)lists.auriga.wearlab.de > http://lists.auriga.wearlab.de/cgi-bin/mailman/listinfo/dillo-dev

Nov. 23, 2003

[patch] Cleanup of IO.c

by Indan Zupancic

Hi, Just cleaned IO.c up, so nothing fundamental changed. Besides some cleanups in multiple functions this are the two major changes: Removal of a_IO_write_chunks, mainly because it isn't used anywhere and it doesn't add any real functionality, it does the same only in a different way. Thanks to CVS it is easy to add it back when you really needed, if you want. Removal of the ValidIOs list. This is the logical next step after g_source_remove was discovered, because the extra indirection isn't needed anymore. All in all the result is a much smaller IO.c (400 lines instead of 700), smaller Dillo (about 3K with -O2), and O(1) behaviour instead of O(n) in the IO engine. The changes have really no negative points as far as I can tell. Link to the patch: http://www.xs4all.nl/~dorinek/dillo/IO-cleanup.patch.gz Greetings, Indan

Nov. 22, 2003