| 00:29.36 | *** join/#gpe HET2 (~diman@chello080110105196.509.15.vie.surfer.at) |
| 00:42.44 | *** join/#gpe HET2 (~diman@chello080110105196.509.15.vie.surfer.at) |
| 02:16.11 | *** join/#gpe jg (~jg@dialup-63.214.108.180.Dial1.Boston1.Level3.net) |
| 02:30.50 | mallum | hey jg |
| 02:31.08 | jg | mallum: hi. |
| 02:31.57 | mallum | jg: any joy with a new X Server ? |
| 02:32.20 | jg | yeah, works fine. Just haven't had time to do the packaging work. |
| 02:32.43 | mallum | jg: oh cool :-) |
| 02:33.00 | mallum | jg: do the packaging scripts need work ? |
| 02:33.54 | jg | mallum: the title bar problem is caused by the fact that I didn't have all the locales packaged up: dynamically loaded locale support got added last summer to XFree86. So I have to build scripts to package them all up independently. |
| 02:35.34 | mallum | jg: aha I didn't realise that needed extra packaging work |
| 02:35.56 | jg | yeah, there are a bunch of Xlib locales defined, and there ought to be a package per locale. |
| 02:36.21 | jg | In theory, all the data can be extracted from a configuration file, but building the scripts is a bit of work. |
| 02:36.28 | mallum | jg: I have a little time in the week and I may put my arm box to work building prelinked enabled X librarys |
| 02:36.52 | mallum | jg: According to pb there isn't a cross compiler yet that can do it |
| 02:36.53 | jg | what do you want/need prelinking for? |
| 02:37.31 | mallum | jg: hopefully speed up app loading |
| 02:38.49 | mallum | jg: apparently opie has seen large increases though Im not sure how much of that was due to them being pretty slow anyway due to c++ ness |
| 02:39.39 | jg | sounds like C++... |
| 02:40.25 | mallum | yeah, itd be nice if it helped a little with gtk2 apps |
| 02:44.28 | mallum | jg: oh I got my extended abstract in for freeneix |
| 02:53.55 | jg | mallum: good. Keep Carl happy... |
| 02:54.34 | mallum | hehe |
| 02:56.33 | jg | well, it is beddie bye time for bonzo bear.... |
| 02:56.48 | jg | mallum: good night... |
| 03:05.43 | *** join/#gpe gdsx (xsdg@xsdg.staff.freenode) |
| 09:07.37 | *** join/#gpe musb_out (~musb@200.163.16.38) |
| 09:19.06 | *** join/#gpe pierre (~pierre@AClermont-Ferrand-201-1-4-249.abo.wanadoo.fr) |
| 09:53.28 | *** join/#gpe pb_ (~pb_@dsl-62-3-66-201.zen.co.uk) |
| 11:03.57 | *** join/#gpe mallum (~mallum@pc-80-193-218-32-hw.blueyonder.co.uk) |
| 11:07.25 | *** join/#gpe HET2 (~diman@chello080110105196.509.15.vie.surfer.at) |
| 11:31.46 | *** join/#gpe PaxAnima (~petter@v058b.studby.ntnu.no) |
| 11:44.59 | mallum | hey pb_ |
| 11:45.52 | pb_ | hi mallum |
| 11:46.40 | mallum | pb_: hmm, Im just in the middle of a 150mb apt-get upgrade |
| 11:47.46 | pb_ | :-} |
| 12:10.52 | PaxAnima | mallum: is it possible to embed xkbd and still have it send key strokes to other windows? |
| 12:18.33 | mallum | PaxAnima: yeah should be |
| 12:18.33 | mallum | PaxAnima: as long as the embedder explicitly states it does not want keyboard focus |
| 12:19.00 | PaxAnima | ok... some xlib magic? |
| 12:20.23 | mallum | a little, look at xkbd.c in the xkbd src and I think its faily obvious where it states it dont want focus |
| 12:20.35 | PaxAnima | ok, thanks |
| 12:48.26 | *** join/#gpe flypiper (~46-cub@168.215.193.70) |
| 13:47.23 | mallum | pb_: ping |
| 13:50.20 | pb_ | mallum: que? |
| 13:52.38 | mallum | pb_: is there an easy l18n way of formatting the system date ? |
| 13:53.35 | pb_ | mallum: strftime(nl_langinfo(D_FMT)) |
| 13:54.05 | mallum | pb_: do I have to call anything before that ? |
| 13:54.19 | pb_ | mallum: you need to have called setlocale at some point. |
| 13:54.56 | mallum | pb_: okey thought so - thanks :-) |
| 14:29.21 | *** join/#gpe pierre_ (~pierre@AClermont-Ferrand-201-1-3-58.abo.wanadoo.fr) |
| 15:08.35 | *** join/#gpe prpplague (~JoeBob1@12.148.134.9) |
| 16:22.44 | mallum | hmmmm |
| 16:25.45 | mallum | pb_: ping |
| 16:36.50 | pb_ | mallum: yo |
| 16:39.05 | mallum | pb_: hey, I plugged in some libfam code |
| 16:39.12 | mallum | pb_: to mbdesktop |
| 16:39.20 | mallum | pb_: but its not working :( |
| 16:39.20 | pb_ | cool. |
| 16:39.23 | pb_ | ah, not so cool. |
| 16:39.58 | mallum | pb_: I dubious to it being a powerpc / dnotify problem .. if I tarball up the source could you give it ago for me ? |
| 16:41.35 | pb_ | sure |
| 16:41.40 | pb_ | probably have to wait until I get home though. |
| 16:42.04 | mallum | pb_: okey |
| 16:42.24 | mallum | pb_: ley me post my select() stuff into #flood and see if it looks ok to you |
| 17:37.03 | mallum | pb_: ok I've uploaded mbdesktop-0.0.6 to my experiemental dir |
| 17:37.18 | pb_ | cool. |
| 17:37.31 | mallum | pb_: I have the code now recieving initial events from the fam server ( of what files exist ), but not actual changes once its started up |
| 17:37.40 | pb_ | right. |
| 17:37.44 | mallum | pb_: so Im 90% sure now its a problem with the server |
| 17:39.22 | mallum | pb_: what do you use to convert debs to ipkgs ? |
| 17:39.33 | mallum | pb_: I'll make a fam ipkg |
| 17:39.46 | pb_ | I have an, um, special script. |
| 17:40.06 | pb_ | handhelds.org/~pb/dpkg-distil |
| 17:40.51 | pb_ | One thing it doesn't currently do is replace the Maintainer: field in the control file. You will need to do that by hand, as the Debian maintainer will probably get upset if he starts receiving complaints from hh.org bugzilla. |
| 17:42.10 | mallum | pb_: aha cool |
| 17:44.42 | pb_ | mallum: also, you should run your eye over the dependencies: there are some packages that are named differently in familiar, or have gone from concrete to virtual, and the dependencies may need updating to match. |
| 17:55.22 | *** join/#gpe dc__ (~dctanner@modem-167-41-60-62.vip.uk.com) |
| 17:55.32 | pb_ | dc__: g'day |
| 17:55.47 | dc__ | pb_: afternoon :) |
| 17:56.28 | dc__ | pb_: are you aware of this error where xkbd doesn't show up in gpe-login? |
| 17:56.48 | pb_ | dc__: vaguely. |
| 17:57.01 | pb_ | dc__: I remember hearing there were problems with gpe-login2 in autolock mode. |
| 17:57.04 | dc__ | pb_: well it seems to happen after the 1st time you suspend and resume. |
| 17:57.14 | pb_ | dc__: yeah, that would be it. |
| 17:57.21 | dc__ | pb_: the first time you do works, but every time after that you just get no xkbd L. |
| 17:57.23 | dc__ | @ |
| 17:57.23 | dc__ | :/ |
| 17:57.29 | pb_ | dc__: I always use hard-keys mode these days so I keep forgetting about it. |
| 17:57.32 | dc__ | and it's not very helpful :} |
| 17:57.39 | dc__ | pb_: indeed. |
| 17:57.56 | pb_ | dc__: I don't know what would be going wrong there. I will have a look this evening and see if I can figure it out. |
| 17:58.20 | dc__ | pb_: good'o.. |
| 17:58.24 | dc__ | pb_: now, mime. |
| 17:58.36 | dc__ | pb_: what do you think about just adopting the gnome methods. |
| 17:58.52 | pb_ | dc__: I don't know much about the various competing methods. |
| 17:59.03 | pb_ | well, for "much", read "anything" really. |
| 17:59.31 | dc__ | pb_: basicly, gnome has some real weird method - but it seems to work okay for them. |
| 17:59.49 | pb_ | dc__: right. |
| 17:59.51 | dc__ | pb_: KDE uses .desktop files, but the arrangement is _very_ ugly, but that's just me :] |
| 18:00.21 | dc__ | pb_: then there's the even weirder fd.org idea, that uses, guess it....xml |
| 18:00.26 | pb_ | whoa. |
| 18:00.41 | dc__ | hehe |
| 18:00.50 | pb_ | well, hmm, with mallum here we must be very careful not to say anything bad about freedesktop.org. |
| 18:00.57 | dc__ | but aparently it also uses some mime update command like ours. |
| 18:01.05 | dc__ | pb_: well, xml is fun.... |
| 18:01.13 | dc__ | pb_: if you like that kind of stuff :D |
| 18:01.20 | pb_ | dc__: indeed. :-} |
| 18:01.43 | dc__ | personly after storm, I'm no longer on very good terms with it. |
| 18:01.44 | pb_ | dc__: I can't say I do particularly, though I guess I have used XML as a data storage format myself before now. |
| 18:02.03 | dc__ | but it does do the job. |
| 18:02.05 | pb_ | dc__: well, let me look at the different options and think about it a bit. |
| 18:02.16 | dc__ | what I've been doing :] |
| 18:02.27 | pb_ | heh. |
| 18:02.56 | pb_ | it might be worth investigating whether Debian has any plans for a unified mime handling system as well. |
| 18:03.06 | pb_ | it would seem unfortunate to have GNOME and KDE using different ones. |
| 18:03.20 | dc__ | heh, yeagh |
| 18:03.32 | pb_ | let me search my unread debian-devel mail for likely threads. |
| 18:03.42 | dc__ | ah, goody. |
| 18:03.54 | pb_ | the power of evolution :-) |
| 18:04.05 | dc__ | oh dear, yes. |
| 18:04.09 | pb_ | sadly nothing there. |
| 18:04.21 | dc__ | pb_: I reply, the power of squirrel mail! |
| 18:04.25 | pb_ | heh. |
| 18:08.12 | dc__ | hrm |
| 18:08.21 | dc__ | I may just stick with the current system :[ |
| 18:08.26 | dc__ | but maybe make it into a libmime |
| 18:08.29 | dc__ | later on. |
| 18:08.30 | pb_ | mmm. |
| 18:08.45 | pb_ | seems like a reasonable enough idea, I guess. |
| 18:08.46 | dc__ | so then we can easily change the actualy filesystem format and still have the same api. |
| 18:08.52 | pb_ | indeed. |
| 18:09.04 | dc__ | right, now, I'll first finish of the mime editor :) |
| 18:09.14 | dc__ | bbl food :) |
| 18:32.36 | *** join/#gpe ade|desk (~adrian@unimatrix.ex.ac.uk) |
| 18:35.19 | pb_ | dc__: on the gpe-login thing, the xkbd is still running for subsequent sessions but is apparently not getting mapped correctly. I'm not sure why that is yet. |
| 18:36.03 | pb_ | dc__: in the interests of memory usage, it is probably not desirable to keep xkbd hanging around all the time anyway. It might be better to start it anew each time. |
| 18:37.13 | moray | pb_: sounds a good idea, given it would presumably fix the bug by accident at the same time ;) |
| 18:37.50 | mallum | pb_: yeah, it starts up pretty quick |
| 18:38.01 | pb_ | moray: indeed. |
| 18:38.46 | mallum | pb_: I said to dc__ the fd.org mime stuff wasn't very nice ;-) |
| 18:39.02 | pb_ | mallum: ha. |
| 18:39.03 | mallum | pb_: I think the kde system is nicest and maybe most suitable |
| 18:40.47 | pb_ | right. |
| 18:53.32 | mallum | hmm, anyone know if familiar has start-stop-deamon ? |
| 18:57.12 | mallum | arg fam is all c++ :( |
| 19:05.23 | dc__ | hehe |
| 19:05.25 | dc__ | hum |
| 19:05.33 | dc__ | mallum: so you really think the kde method is best? |
| 19:07.52 | mallum | seems the most efficient |
| 19:08.01 | dc__ | so really really? |
| 19:08.08 | mallum | which is surely good on a 'limited' platform |
| 19:08.38 | dc__ | okay! |
| 19:08.46 | dc__ | so it is :) |
| 19:08.51 | PaxAnima | what's the kde method in this case? |
| 19:08.55 | dc__ | and I'll write it into a libmime aswell |
| 19:09.02 | dc__ | PaxAnima: no one needs to know ;] |
| 19:09.31 | PaxAnima | dc__: ah.. and I guess the alternative is gnome-method? |
| 19:09.52 | dc__ | PaxAnima: I guess :/ |
| 19:10.47 | dc__ | mallum: so now, then I guess both mbdesktop and gpe-filemanager will benifit from the .desktop files :} |
| 19:10.59 | mallum | dc__: also then pkg maintainer only need create a .desktop file, no a .desktop file and a mime file |
| 19:11.03 | mallum | dc__: indeed |
| 19:11.19 | dc__ | mallum: :) |
| 19:11.44 | dc__ | \j #opie |
| 19:11.45 | dc__ | hehe |
| 19:11.55 | dc__ | damn my lack of slashes. |
| 19:12.05 | mallum | dc__: I was kind of hoping gpe-filemanger could benefit for fam too, but a) it bigger than I epected b) its c++ c) I cant get it to work properly |
| 19:13.12 | dc__ | mallum: what is fam...? |
| 19:13.37 | mallum | dc__: http://oss.sgi.com/projects/fam/index.html |
| 19:18.06 | dc__ | oh yay. |
| 19:19.37 | dc__ | [Desktop Entry] |
| 19:19.43 | dc__ | Comment=A Media Player |
| 19:19.43 | dc__ | Comment[es]=Reproductor de ficheros multimedia |
| 19:19.43 | dc__ | Exec=opieplayer2 |
| 19:19.43 | dc__ | Icon=opieplayer2/MPEGPlayer |
| 19:19.43 | dc__ | Type=Application |
| 19:19.43 | dc__ | MimeType=audio/mpeg;audio/x-wav;audio/x-mpegurl;audio/x-ogg;video/x-ogm;video/x-ms-asf;video/quicktime;video/mpeg;video/x-msvideo;playlist/plain; |
| 19:19.48 | dc__ | MimeTypeIcons=SoundPlayer;Sound;SoundPlayer;SoundPlayer;MPEGPlayer;MPEGPlayer;MPEGPlayer;MPEGPlayer;MPEGPlayer;opieplayer2/playlist2 |
| 19:19.49 | dc__ | Name=Opie Media Player |
| 19:19.53 | dc__ | mallum: there we go |
| 19:20.12 | dc__ | hum |
| 19:20.19 | mallum | cool |
| 19:20.26 | dc__ | guess I have to move all my document icons to /usr/share/pixmaps :/ |
| 19:20.42 | mallum | dc__: I think you can specify them explicity |
| 19:20.49 | dc__ | mallum: so, if you make sure mbdesktop sticks to that... |
| 19:20.55 | dc__ | http://cvs.handhelds.org/cgi-bin/viewcvs.cgi/opie/apps/Applications/mediaplayer.desktop?rev=1.10&content-type=text/vnd.viewcvs-markup |
| 19:21.12 | dc__ | mallum: yeah, but that just gets too messy. |
| 19:21.15 | dc__ | mallum: I'll just move them :] |
| 19:21.37 | mallum | dc__: mbdesktop check for an inital '/' and opens absolute |
| 19:21.53 | dc__ | right |
| 19:22.01 | mallum | dc__: but better if everything isn't absolute, cos then it means we can switch themes ( in theory ) |
| 19:22.04 | dc__ | so that's good for the main icon. |
| 19:22.12 | dc__ | yeah |
| 19:22.22 | dc__ | so err for the document icons.. |
| 19:22.27 | dc__ | oh I know. |
| 19:22.40 | dc__ | I'll just make libmime search in the gpe-fm path... |
| 19:22.43 | mallum | dc__: are you parsing the ';' seperators ? |
| 19:22.45 | dc__ | hrm |
| 19:23.01 | dc__ | mallum: do you think libmime or gpe-fm should find the absolute filename for the icons? |
| 19:23.11 | dc__ | mallum: I will be. |
| 19:23.24 | mallum | dc__: well I just added it so older stuff would work |
| 19:23.34 | dc__ | oh cool. |
| 19:23.41 | mallum | dc__: when you do, pass me the code and I'll add it to my dotdesktop |
| 19:23.52 | mallum | dc__: as catogaries are ; split |
| 19:23.58 | dc__ | mallum: hum, do you wanna send me your dotdektop stuff again? I do have my own edited version. |
| 19:24.08 | dc__ | mallum: but it'd be kind of better to use yours. |
| 19:24.33 | mallum | dc__: sure mine is in matchbox cvs and will be part of libMB eventually |
| 19:24.43 | dc__ | hum |
| 19:24.56 | mallum | dc__: but its pretty small |
| 19:24.56 | dc__ | so I guess we use different source files though? |
| 19:25.07 | dc__ | libdotdesktop ;p |
| 19:25.30 | mallum | dc__: yeah could do |
| 19:25.38 | dc__ | mallum: can we do that :) |
| 19:25.39 | mallum | dinner time ... biab |
| 19:25.53 | dc__ | mallum: it makes sense, since both libMB and libmime will want to use it. |
| 19:25.59 | dc__ | mallum: I'll make it ;) |
| 19:29.49 | *** join/#gpe pierre (~pierre@AClermont-Ferrand-201-1-3-58.abo.wanadoo.fr) |
| 19:38.35 | dc__ | mallum: wow, so mbmenu does dotdesktop :) |
| 19:43.44 | mallum | dc__: yep, just about |
| 19:46.40 | dc__ | right, so I made myself a libdotdesktop, and I'll put it in cvs once I get libmime vaugely working. |
| 19:48.44 | *** join/#gpe pierre (~pierre@AClermont-Ferrand-201-1-3-58.abo.wanadoo.fr) |
| 19:49.12 | mallum | dc__: cool |
| 19:49.32 | mallum | dc__: BTW I got rid of the strsep parsing around the ='s |
| 19:51.30 | mallum | dc__: I got dnotify working :)) |
| 19:51.51 | dc__ | :) |
| 19:51.54 | dc__ | bbl, tv |
| 19:53.41 | *** join/#gpe pierre (~pierre@AClermont-Ferrand-201-1-3-58.abo.wanadoo.fr) |
| 20:16.17 | *** join/#gpe benmeyer (~ben@gatekeepernj3.sharpsec.com) |
| 20:28.52 | *** join/#gpe pb_ (~pb@pc2-cmbg4-3-cust239.cmbg.cable.ntl.com) |
| 21:00.13 | mallum | pb_: boo |
| 21:03.45 | mallum | pb_: I switched to pure dnotify and it works :-) |
| 21:04.41 | pb_ | ah, cool. |
| 21:07.12 | mallum | fam was all c++ and the daemon was 150k |
| 21:07.24 | pb_ | ugh. |
| 21:11.11 | mallum | dnotify was approx 10 lines |
| 21:11.49 | moray | hmm |
| 21:19.47 | *** join/#gpe HET2 (~diman@chello080110105196.509.15.vie.surfer.at) |
| 21:20.38 | mallum | moray: hmm ? |
| 21:21.44 | moray | mallum: 10 lines v. 150k. though if it had been a general comment, it would have been about this essay-writing with which I am bored |
| 21:27.01 | mallum | moray: unfotunalty its not portable across anything other than linux, whilst fam is ( but fam uses dnotify on linux ) |
| 21:28.24 | moray | ah right, I wondered if (hoped that) the 10 lineness was because it was using the kernel-provided API |
| 21:34.28 | *** join/#gpe mallum (~mallum@pc-80-193-218-32-hw.blueyonder.co.uk) |
| 21:47.45 | mallum | pb_: whats a SIG32 signal ? |
| 21:48.24 | pb_ | mallum: first RT signal. I think linuxthreads uses it, among other things. |
| 21:49.38 | pb_ | also SIGSWI for the Arthur emulation, though I doubt that's what you're seeing. :-} |
| 21:50.32 | mallum | pb_: hmm, maybe gdb is being wierd |
| 21:51.02 | pb_ | mallum: with most causes of sig32, the correct response is just to hit "c" in gdb. |
| 21:51.29 | pb_ | you can probably convince it not to stop at all for those signals, though I don't know how offhand. |
| 21:54.04 | mallum | http://www.pugs-are-us.co.uk |
| 21:54.30 | mallum | ^^^^^ where we got the pugs from |
| 21:55.31 | mallum | pb_: cool Im trying to fix a nasty seg on the signal callback |
| 21:55.45 | pb_ | mallum: ah, it really _is_ the middle of nowhere. |
| 21:56.13 | cmarqu | Anybody running a Cyrus IMAP server? I heard there will be a remote exploit soon. |
| 21:56.48 | mallum | pb_: yeah surrounded by millions of cabbages |
| 21:56.59 | pb_ | heh, |
| 21:57.20 | mallum | pb_: we have to actually pick them up nearer xmas, so maybe we could pop in and say hi on our way up if your about .. |
| 21:57.30 | mallum | cmarqu: yes I am |
| 21:57.42 | pb_ | mallum: sure, let me know what day you will be passing. |
| 21:58.10 | cmarqu | mallum: I know nothing more, but this guy knows what he is talking about |
| 21:58.11 | mallum | cmarqu: but its not externally excessable ( I think ) - try nmaping it mallum.homeip.net |
| 21:58.20 | mallum | pb_: cool |
| 21:58.27 | mallum | biab - alan partridge is on |
| 22:02.32 | *** part/#gpe NonToxic (joshua@h002078c680df.ne.client2.attbi.com) |
| 22:17.17 | *** join/#gpe BanditD (~BanditD@pD9006455.dip.t-dialin.net) |
| 22:26.38 | *** join/#gpe dc__ (~dctanner@async123-6.nas.onetel.net.uk) |
| 22:26.46 | dc__ | ack |
| 22:26.48 | dc__ | pb_: hum |
| 22:26.56 | dc__ | where's that mallum boy |
| 22:29.29 | PaxAnima | dc__: mallum just got disconnected... |
| 22:29.34 | *** join/#gpe flypiper (~pipercub@12-222-141-210.client.insightBB.com) |
| 22:29.47 | dc__ | PaxAnima: damn L. |
| 22:29.48 | dc__ | :/ |
| 22:30.07 | dc__ | it seems with the kde method, there are no actual names of the mime types stored anywhere |
| 22:30.28 | dc__ | so it has like image/jpeg with the extension .jpg and all |
| 22:30.49 | dc__ | but it doesn't store 'JPEG Image File' anywhere for me :{ |
| 22:31.32 | PaxAnima | hmm.. do you need that? |
| 22:31.45 | dc__ | well, if I want to display infomation about it. |
| 22:31.51 | dc__ | especialy in the mime editory. |
| 22:31.53 | dc__ | editor. |
| 22:32.22 | *** join/#gpe mallum (~mallum@pc-80-193-218-32-hw.blueyonder.co.uk) |
| 22:32.28 | dc__ | mallum: ! |
| 22:32.30 | dc__ | mallum: help :/ |
| 22:32.43 | dc__ | mallum: I can't seem to find out where the name of mime types is stored. |
| 22:32.57 | dc__ | mallum: e.g. 'JPEG Image File'.... |
| 22:33.06 | mallum | hmm, my machine locked up :( |
| 22:33.06 | dc__ | mallum: so I can display the name of a mime type in the mime editor. |
| 22:33.10 | dc__ | mallum: :[ |
| 22:33.24 | mallum | did cmarqu scan my machine ? |
| 22:33.38 | mallum | name of the mimetype ? |
| 22:33.48 | dc__ | yes |
| 22:35.20 | mallum | how do you mean ? |
| 22:35.25 | dc__ | well |
| 22:35.37 | pb_ | dc__: hum indeed. |
| 22:35.39 | dc__ | where do I get the name of say the video/mpeg mime type? |
| 22:35.57 | dc__ | <PROTECTED> |
| 22:36.08 | dc__ | and the .desktop file will contain the icon to use for video/mpeg |
| 22:36.22 | dc__ | but no where does it say that the file type is actualy called. |
| 22:36.43 | pb_ | dc__: on my machine I seem to have some stuff in /etc/mime-handlers with that information in. Is that the GNOME scheme? |
| 22:37.01 | dc__ | pb_: heh, err no, I think that's probly my stuff :/ |
| 22:37.06 | pb_ | dc__: ah. :-) |
| 22:37.11 | dc__ | pb_: :) |
| 22:37.31 | pb_ | dc__: I guess I must have been making install a bit too indiscriminately :-} |
| 22:37.44 | dc__ | indeed. |
| 22:39.01 | dc__ | okay... |
| 22:39.04 | dc__ | hm |
| 22:39.04 | dc__ | :/ |
| 22:42.08 | mallum | dc__: dont the .desktop that provides a particular minitype provide an executable to be run for that mime type ? |
| 22:44.07 | dc__ | mallum: yes. |
| 22:44.11 | dc__ | mallum: but I want the name of it :) |
| 22:44.23 | dc__ | mallum: it seems to opie guys think it's hardcoded :{ |
| 22:45.28 | dc__ | heh, |
| 22:45.35 | dc__ | mallum: think about that :) |
| 22:45.35 | dc__ | bbl.... |
| 22:49.23 | mallum | dc__: just use the .desktops Exec= key ? |
| 23:02.03 | mallum | pb_: woohoo, you can now remove, add new or edit the dotdesktop files and the changes appear on the fly in mbdesktop ( although a little inefficiently ) |
| 23:02.13 | pb_ | cool! |
| 23:02.29 | pb_ | btw, it would be neat if mbdesktop had a -bg option for inherit from root window. |
| 23:02.50 | mallum | pb_: how do you mean ? |
| 23:03.21 | pb_ | mallum: use the background pixmap that was set on the root. |
| 23:03.27 | pb_ | mallum: or can you do that already? |
| 23:04.35 | mallum | pb_: you can pass a background image as a parameter and it will stretch / tile it for you ( and deal with rotations ) |
| 23:04.47 | pb_ | mallum: right, but I want it to just pick up the one that is already set. |
| 23:04.56 | pb_ | mallum: ie, be psuedo-transparent. |
| 23:05.05 | mallum | pb_: yeah that'd be pretty easy to add in |
| 23:05.49 | pb_ | I guess in the particular case of mbdesktop you probably could actually make it use ParentRelative without worrying about the window manager getting in the way. |
| 23:06.33 | mallum | pb_: it cant use parent relative as it needs to get the background pixmap into an mbpixpuf so it can alphablend against it |
| 23:06.42 | pb_ | ah. |
| 23:07.20 | pb_ | so that would mean having two copies of the pixmap in memory, which would suck a bit. |
| 23:08.08 | mallum | yep |
| 23:11.15 | mallum | pb_: its probably safer just to ditch setting the root pixmap |
| 23:11.35 | pb_ | mallum: can you not use Render to display your images alpha blended onto the image behind them? |
| 23:11.42 | mallum | pb_: otherwise memory will be wasted on something you cant see |
| 23:13.13 | pb_ | mallum: indeed. but, from the point of view of configuring the background image, it is not especially desirable to need to know that mbdesktop is in use, and do things differently. |
| 23:14.15 | mallum | pb_: you could set an xsettings prop or summin |
| 23:15.03 | mallum | pb_: with matchbox just a background image is kind of a waste anyway cos you'll only see it when no apps are open |
| 23:15.38 | pb_ | mallum: sure, but people like to set one nonetheless. |
| 23:16.41 | mallum | pb_: if its an issue of people wanting to do the transparency hack, I can make mbdesktop export the root pixmap_id thing |
| 23:19.08 | mallum | pb_: Im not sure how nautilus or kdesktop handle this type of thing |
| 23:19.12 | pb_ | mallum: it's more an issue of apps that want to change the background (e.g. colin's gpe-setbg) |
| 23:20.50 | pb_ | mallum: seems that nautilus declares its own pixmap as the root one, which I guess it is in a sense. |
| 23:21.13 | mallum | pb_: well they could check if a desktop is running, if so change some XSetting, if not do the 'normal' thing |
| 23:21.36 | pb_ | mallum: sure, I would just rather the "normal" thing went on working the way it did before. |
| 23:21.37 | mallum | pb_: yeah thats what Im saying mbdesktop should do |
| 23:23.50 | mallum | pb_: it can work how it did, it just 320x240x2 bytes will be wasted :( |
| 23:24.19 | mallum | pb_: what happens if you run gpe-setbg on nautilus ? |
| 23:24.28 | pb_ | mallum: same, it fails. |
| 23:25.14 | pb_ | or, to be more specific, it changes the root background but you don't get to see the results until you manage to kill Nautilus (and make it stay dead, which appears to be almost impossible) |
| 23:25.18 | mallum | pb_: I'll make mbdesktop conflict with gpe-setbg ;-) |
| 23:25.51 | pb_ | mallum: is it really not possible for mbdesktop to operate without keeping a whole-screen pixmap to composite the icons onto? |
| 23:26.15 | mallum | pb_: welcome to X windows ;-) |
| 23:26.43 | pb_ | mallum: geez, the opie guys were right all along. |
| 23:27.29 | pb_ | mallum: but I thought this was one of the things XRender was meant to fix. It can render antialiased text, right? |
| 23:28.12 | mallum | pb_: I did a while back look and what I could do with render ( for exacly that problem ), there seems to be no easy way of drawing a rgb array, apart from drawing thoughsands of XRender 1x1 rectangles |
| 23:28.28 | pb_ | mallum: ah, hmm. |
| 23:28.42 | pb_ | mallum: that's a shame. |
| 23:29.43 | mallum | pb_: you'd expect you could create an xrender picture with arbituary data but looking through the headers I couldn't figure out an obvious way to do it |
| 23:30.31 | mallum | pb_: yet as you say it must be possible becuase at its core, all xft is doing is rendering freetype generated bitmaps to render pictures |
| 23:31.39 | mallum | pb_: and also even if that was possible it would still be a hack as I'd have to fake/hack transparency |
| 23:32.39 | mallum | pb_: I think that kind of thing is definetly coming in render though ... |
| 23:32.47 | pb_ | mallum: indeed. I don't really share your aversion to the transparency hacks, however :-) |
| 23:34.47 | mallum | pb_: I dont think the hacks show up so much in matchbox cos everything is so static |
| 23:35.00 | pb_ | mallum: true. |
| 23:36.34 | mallum | pb_: people are better setting there background image with mbdesktop anyway, cos then it will honor rotations ;-) |
| 23:41.46 | mallum | dc__: http://art.gnome.org/show_theme.php?themeID=258&category=gtk2 ! |
| 23:44.33 | pb_ | mallum: seems to be a common problem (lo going missing) |
| 23:44.47 | pb_ | mallum: I have no idea what causes it though :-} |
| 23:46.02 | mallum | pb_: yeah, it seems to have started happening since I set an up command in /etc/network/interfaces to mount an nfs dir when Im at home |
| 23:46.13 | mallum | pb_: portmap now seems to hang around on boot |
| 23:46.38 | pb_ | mallum: I think portmap problems are normally caused by missing lo. |
| 23:47.31 | mallum | pb_: hmm, how annoying :( |
| 23:47.53 | pb_ | indeed. |
| 23:48.00 | pb_ | does it work if you "ifup lo" manually? |
| 23:48.38 | mallum | yeah |
| 23:48.53 | mallum | I have to take it down then up and then its fine |
| 23:50.04 | mallum | pb_: I have to figure out how to block signals too |
| 23:50.26 | pb_ | sigprocmask, I guess. |
| 23:50.46 | mallum | with the sigaction 'advanced' stuff ? |
| 23:51.07 | mallum | I can block and they'll get queued till I unblock ? |
| 23:51.11 | pb_ | yeah. |
| 23:56.27 | *** join/#gpe ade|desk (~adrian@unimatrix.ex.ac.uk) |