| 00:52:08 | Pentad | I have a question on cross-compiling and X11, anybody got a sec? |
| 00:54:06 | PaxAnima | sorry, don't use crosscompiler |
| 00:54:35 | Pentad | NP, I'll try back here another day. :-) |
| 04:33:16 | colin_werk | Morning |
| 07:23:29 | jsmanrique | hello |
| 07:23:37 | jsmanrique | good morning (from Spain) |
| 07:36:54 | jsmanrique | I will back later... |
| 07:36:57 | jsmanrique | see you |
| 08:08:18 | mallum | hey |
| 08:11:25 | mallum | pb_: I think I figured out the mmenu and png issue |
| 08:19:28 | pb_ | hey cool, what was up with it? |
| 08:21:47 | mallum | well I played about with the icon last night on my laptop and it loaded fine |
| 08:22:11 | mallum | then I realised, 3100 has a staticgray visual and of course Im not attempting to handle that it any way |
| 08:22:43 | mallum | so its building a pixmap out of the png with a load of crap |
| 08:36:57 | pb_ | ah. |
| 08:37:31 | pb_ | mmm, I wouldn't be surprised if some of our other programs have the same problem. |
| 08:42:03 | mallum | yeah maybe, X Visuals are a knightmare |
| 08:43:27 | mallum | I've been messing with the 8bit one which gives a psuedo visual ( 'classic' colormap ) and a direct color one |
| 08:44:07 | mallum | the directcolor one gives me some plane masks for R, G and B, but when I put data in them the colors are all wrong :( |
| 08:44:45 | mallum | I dont know if my XServer is broke ar 8bit or my bit shifts are broke ( seem ok and work fine for 16 and 24 ) or Im missing something entirely |
| 08:46:49 | pb_ | Doesn't directcolour still go through a LUT of some kind? |
| 08:47:09 | pb_ | I always get directcolour and truecolour mixed up. %-/ |
| 08:49:31 | pb_ | The way I _think_ directcolour works is that the pixel value is split into R/G/B fields, and each one of them is looked up in a map table to get the actual colour value that will be used for display. |
| 08:50:02 | pb_ | I.e. it's still a bit like pseudocolour, but you can make it look like a very coarse truecolour by setting the LUTs to a 1:1 mapping. |
| 08:50:52 | mallum | LUT = colormap ? |
| 08:54:56 | pb_ | yeah. |
| 08:54:58 | pb_ | Look Up Table |
| 08:55:53 | mallum | yeah , but Im confused on how I set it for that specific case |
| 08:55:58 | pb_ | Dunno. |
| 08:56:19 | pb_ | I'm not sure that 8-bit directcolor is worth losing much sleep over. I think most people on 8bpp displays run pseudocolor. |
| 08:56:19 | mallum | ok, back to reaading posts rom the early 90's on google groups ;-) |
| 08:57:09 | pb_ | at least, I always used pseudocolour back in the days of 8bpp displays, but maybe I was weird. |
| 08:57:14 | mallum | pb_: yeah I want to go with directcolor as then my code fits better |
| 08:57:32 | pb_ | ah. |
| 08:58:13 | pb_ | is that really true? directcolour is usually more awkward, and/or less expressive. |
| 08:58:24 | mallum | ie I can just take my rgba[] array and shift things into place rather than attempting to match values against random colormap |
| 08:59:24 | pb_ | by driving it as truecolour you mean? Yeah, that makes it easy, but you reduce yourself to just a couple of bits of information on each colour channel. |
| 09:00:32 | mallum | yeah |
| 09:01:07 | mallum | Im not really that bothered about it looking really great as long as it looks ok |
| 09:01:14 | pb_ | fair enough. |
| 09:01:37 | mallum | Its only the tuxscreen that has an 8bit display that I know of |
| 09:01:59 | mallum | and most new machines now come with at least 12/16 |
| 09:02:03 | pb_ | yeah. |
| 09:02:16 | mallum | I would have probably thought its cheaper now to but a 16bit lcd rather than an 8bit one |
| 09:12:29 | jsmanrique | hello, again ;-) |
| 09:12:35 | jsmanrique | good morning |
| 09:13:19 | mallum | hi there |
| 09:20:24 | mallum | pb_: by reading the man page of XCreateColorMap, it appears I can pass the visual in and if its directcolor it will allocate the colormap for the RGB plane ( *I think* ) |
| 09:26:18 | jsmanrique | any news about "monochrome" icons? |
| 09:26:29 | jsmanrique | 3 days to arrive London... |
| 09:31:14 | jsmanrique | what about the weather forecast for this weekend? |
| 09:31:39 | mallum | well we can maybe hook up for lunch/beer if you want |
| 09:32:19 | jsmanrique | nice.. |
| 09:32:38 | mallum | what part of town are you staying ? |
| 09:32:47 | jsmanrique | I am newbie at linux, but even more at London ;) |
| 09:32:52 | colin_werk | jsmanrique: if you mean this: http://europa.armlinux.org/cgi-bin/bugzilla/show_bug.cgi?id=107, then it has an action item for you |
| 09:33:11 | jsmanrique | colin_werk . oops, I have to take a look |
| 09:33:23 | jsmanrique | mallum : Hyde park |
| 09:33:44 | jsmanrique | mallum : near Padington station, I think.. |
| 09:33:56 | mallum | jsmanrique: ok, I is not too far from there |
| 09:35:55 | jsmanrique | mallum : the conferences will be on Saturday in Imperial College |
| 09:36:22 | jsmanrique | mallum : the hotel is Hyde Park West |
| 09:36:59 | mallum | jsmanrique: okey , are you getting the eurostar ? |
| 09:39:07 | jsmanrique | eurostar? what's that? |
| 09:41:55 | pb_ | train through the Channel Tunnel |
| 09:41:55 | mallum | the channel tunnel |
| 09:42:25 | pb_ | don't think the Eurostar goes as far as Spain, does it? |
| 09:42:50 | mallum | I though jsmanrique was french ? |
| 09:43:11 | jsmanrique | no, I go by air |
| 09:43:19 | jsmanrique | no, I am spanish |
| 09:43:52 | jsmanrique | we (my girlfriend and me) will arrive to chatwick airport |
| 09:43:59 | mallum | gatwick |
| 09:44:07 | jsmanrique | yes, that is |
| 09:47:03 | jsmanrique | the agency will take us to the hotel.. |
| 09:55:12 | jsmanrique | we will look for some travelcards to visit it |
| 10:26:55 | ujsmanriqe | sorry |
| 10:27:07 | ujsmanriqe | again my server playing with me :( |
| 10:32:16 | ujsmanriqe | about monochrome icons problem, if you had seen the scaps I did, it seems it is a libgpewidget problem rescalling png files |
| 10:35:03 | ujsmanriqe | well, it seems that scaps have disappeared.. |
| 10:55:17 | mibus | hallo ppls |
| 10:56:29 | jsmanrique | hello again |
| 10:58:00 | pb_ | hi mibus |
| 10:58:16 | mibus | hey pb_ |
| 10:58:31 | mallum | hello mibus |
| 10:58:55 | mibus | hi mallum, jsmanrique |
| 10:59:05 | jsmanrique | I have to go to eat soon.. |
| 10:59:11 | mibus | has anyone had usbnet probs? |
| 10:59:31 | mibus | i upgraded to 0.6pre1 then upgraded my kernel via ipkg to the latest in unstable |
| 10:59:34 | jsmanrique | mallum : we will talk later about possible London date for a beer ;) |
| 10:59:45 | mibus | and now usbnet doesnt seem to like me without lots of pushing |
| 10:59:50 | mibus | :( |
| 10:59:56 | mibus | jsmanrique: ok ciao. have fun! |
| 11:00:13 | mibus | symptoms: usbf shows itself as up and IP'ed |
| 11:00:21 | mibus | <plugs in> |
| 11:00:57 | mibus | standard "usb device connected, assigned device #6" (but **not** usbnet messages, yes usbnet is loaded :) |
| 11:01:15 | mibus | if i run /etc/suspend-scripts/S10usbnet then /etc/resume-scripts/R90usbnet |
| 11:01:19 | mibus | then ifup usbf |
| 11:01:22 | mibus | then add an IP |
| 11:01:25 | mibus | it works OK |
| 11:02:38 | mallum | jsmanrique: ok |
| 11:02:57 | mibus | (actually i have to ifdown usbf before the suspend-script) |
| 11:04:50 | mibus | has successfully scared all others off the channel. Now it's onto World Domination (TM) |
| 11:06:04 | jsmanrique | mallum : ok, I must finish my little speak today.. |
| 11:07:29 | mibus | hey pb you still there? |
| 11:12:07 | mibus | um pb_ youre insane |
| 11:12:46 | mibus | remember how i showed you a mini-fix for gpe-nmf-ogg's title handling code? |
| 11:12:51 | mibus | and you uploaded a new version? |
| 11:13:02 | mibus | well you forgot to upload a new gpe-nmf-ogg, which is where the change was actually made :P |
| 11:37:47 | pb_ | ah, sorry, away from keyboard there. |
| 11:37:54 | mibus | pb_: i noticed :P |
| 11:38:29 | pb_ | mibus: my home PC is switched off at the moment, so I can't upload a package right now. |
| 11:38:35 | pb_ | mibus: I'll do it this evening though. |
| 11:38:46 | mibus | pb_: ah ok :P |
| 11:39:04 | mibus | i might be able to cross-compile gpe-nmf-ogg cos its not gtk-requiring IIRC... |
| 11:39:18 | pb_ | right. |
| 11:39:33 | pb_ | only the frontend needs gtk, though I think some of the backend stuff requires glib. |
| 11:40:01 | mibus | well i can probably hack up the makefiles |
| 11:40:20 | pb_ | yeah. |
| 11:40:21 | mibus | (or wait until tomorrow... hummm... depends when i get home from uni, cos there's a LUG meeting in town...) |
| 11:40:29 | pb_ | or build by hand, it's only one source file. |
| 11:40:37 | mibus | point |
| 11:40:38 | mibus | :D |
| 11:41:46 | mibus | umm... hang on... d-vorbis.so doesnt depend on just *one* source file... does it? |
| 11:41:52 | pb_ | yeah, I think it does. |
| 11:41:57 | pb_ | just vorbis.c as far as I remember. |
| 11:42:17 | mibus | so wheres playlist_ogg.c go? |
| 11:42:23 | mibus | cos its that that changed |
| 11:42:29 | pb_ | ah, hmm, dunno. |
| 11:42:38 | pb_ | maybe that's in there too, you'd have to check the makefile. |
| 11:42:40 | mibus | yeah both |
| 11:42:43 | mibus | just checked |
| 11:42:48 | pb_ | still, two source files is still not too bad. |
| 11:47:04 | mibus | no im just trying it now... |
| 11:47:17 | mibus | hmm |
| 11:47:40 | mibus | playlist_ogg.c includes ogg/ogg.h and vorbis/vorbis.h |
| 11:48:01 | mibus | but in ../../vorbis-fixp/lib there's no ogg/ and vorbis/ dirs... ogg.h & vorbis.h are in *that* directory |
| 11:48:23 | mibus | so you cant build against CVS fixp without installing it first?? |
| 11:48:27 | pb_ | right. |
| 11:48:41 | pb_ | or, more exactly, you need to install the libvorbis-dev package. |
| 11:49:03 | mibus | include fp vorbis and just link against fixp? |
| 11:49:05 | mibus | naughty ;-) |
| 11:50:10 | mibus | and i cant compile fixp vorbis from CVS anyway :-( |
| 11:50:20 | mibus | mibus: ~/ipaq/cvs/gpe/base/vorbis-fixp/lib> make CC=arm-linux-gcc CFLAGS= |
| 11:50:20 | mibus | arm-linux-gcc -c -o framing.o framing.c |
| 11:50:20 | mibus | arm-linux-gcc -c -o bitwise.o bitwise.c |
| 11:50:20 | mibus | arm-linux-gcc -c -o info.o info.c |
| 11:50:20 | mibus | arm-linux-gcc -c -o sharedbook.o sharedbook.c |
| 11:50:21 | mibus | arm-linux-gcc -c -o codebook.o codebook.c |
| 11:50:23 | mibus | arm-linux-gcc -c -o synthesis.o synthesis.c |
| 11:50:25 | mibus | arm-linux-gcc -c -o block.o block.c |
| 11:50:27 | mibus | arm-linux-gcc -c -o window.o window.c |
| 11:50:29 | mibus | arm-linux-gcc -c -o mdct.o mdct.c |
| 11:50:33 | mibus | arm-linux-gcc -c -o lsp.o lsp.c |
| 11:50:35 | mibus | lsp.c: In function `vorbis_lsp_to_curve': |
| 11:50:37 | mibus | lsp.c:86: warning: initialization from incompatible pointer type |
| 11:50:39 | mibus | lsp.c:89: parse error before `int' |
| 11:50:41 | mibus | lsp.c:97: `rln' undeclared (first use in this function) |
| 11:50:42 | mibus | lsp.c:97: (Each undeclared identifier is reported only once |
| 11:50:44 | mibus | lsp.c:97: for each function it appears in.) |
| 11:50:47 | mibus | make: *** [lsp.o] Error 1 |
| 11:50:49 | mibus | mibus: ~/ipaq/cvs/gpe/base/vorbis-fixp/lib> |
| 11:50:51 | mibus | grr |
| 11:50:53 | mibus | time to cheat, methinks |
| 11:50:55 | mibus | copies .so from the iPAQ |
| 11:54:11 | pb_ | hmm, I wonder what went wrong there. |
| 11:54:13 | mibus | grr - "cannot decode this stream" |
| 11:54:45 | mibus | ah |
| 11:54:50 | mibus | it wants libvorbis.so |
| 11:54:57 | mibus | which of course doesnt exist :( |
| 11:55:28 | pb_ | um, it should be linking against libvorbis-fixp.so (.0 maybe) |
| 11:56:28 | pb_ | something is wrong either with your build, or with the vorbis library you are using to link. |
| 11:56:59 | pb_ | It might well be the latter, the libraries on the ipaq are not really intended for development. |
| 11:57:40 | mibus | :P |
| 11:57:50 | mibus | i renamed libvorbis-fixp to libvorbis |
| 11:57:51 | mibus | silly me |
| 11:58:07 | mibus | but now when i link its against libvorbis-fixp.so not libvorbis-fixp.so.0 |
| 11:58:13 | mibus | so it STILL dont work on the ipaq :( |
| 11:58:39 | pb_ | I guess your libvorbis.so is broken then. |
| 11:59:00 | pb_ | You could rename it to libvorbis-fixp.so.0 and modify the final link for d-vorbis.so appropriately. |
| 11:59:12 | mibus | how would i modify the link? |
| 11:59:30 | pb_ | replace -lvorbis with libvorbis-fixp.so.0 |
| 11:59:34 | mibus | ive called it -fixp.so.0 and it wont link (-lvorbis-fixp not fount) |
| 11:59:38 | pb_ | I think that should be ok. |
| 12:00:13 | pb_ | or give up and make a symlink on the ipaq for the time being. |
| 12:00:17 | mibus | :D |
| 12:01:07 | mibus | does gtk2 use much more ram than gtk1? |
| 12:01:12 | pb_ | dunno |
| 12:01:27 | pb_ | wouldn't be surprised if it did. |
| 12:01:52 | mibus | its just i seem to run out of RAM more now... |
| 12:01:58 | mibus | i should check it out... |
| 12:02:05 | mibus | (poor 32MB iPAQ :) |
| 12:02:35 | mibus | <waiting for playlist files to be added> |
| 12:02:42 | mibus | well it didnt crash |
| 12:02:55 | mibus | but of course |
| 12:02:59 | mibus | i havent cvs updated |
| 12:03:00 | mibus | duh! |
| 12:03:02 | mibus | GRRR |
| 12:03:03 | mibus | :P |
| 12:03:04 | pb_ | ! |
| 12:03:31 | mibus | you at yer ipaq ATM? |
| 12:03:41 | mibus | duh, im running the DEBUG appmgr |
| 12:03:44 | mibus | *moron* |
| 12:05:00 | mibus | hmm |
| 12:05:04 | mibus | still didnt pick up the title |
| 12:05:38 | pb_ | yeah, I have my ipaq here. |
| 12:05:43 | pb_ | hmm. |
| 12:06:11 | mibus | syslog is running, that wouldnt be helping |
| 12:06:54 | mibus | damn busybox's ps applet not showing ram %age ;-) |
| 12:07:35 | pb_ | yeah. :-/ |
| 12:07:41 | pb_ | complain to sandman I guess. |
| 12:07:42 | mibus | hmm |
| 12:07:45 | mibus | i dont have a libgpewidget |
| 12:07:46 | mibus | uhoh |
| 12:08:16 | mibus | ~ # ipkg files libgpewidget0 |
| 12:08:16 | mibus | ~ # |
| 12:08:17 | mibus | hmm |
| 12:08:25 | mibus | ~ # ipkg install libgpewidget0 |
| 12:08:25 | mibus | Package libgpewidget0 (0.24) installed in root is up to date |
| 12:08:28 | mibus | hmmmmmmm |
| 12:08:33 | pb_ | looks bad. |
| 12:08:51 | mibus | "odd" more to the point |
| 12:09:00 | mibus | it mustva been there before... |
| 12:09:12 | mibus | reinstall worked ;-) |
| 12:09:36 | jsmanrique | hello |
| 12:09:52 | mibus | hey pb_ theres a bug report on appmgr now for a background image ;) |
| 12:10:45 | mibus | hmm |
| 12:10:49 | mibus | whilst at login screen: |
| 12:10:51 | mibus | ~ $ cat /proc/meminfo |
| 12:10:54 | mibus | total: used: free: shared: buffers: cached: |
| 12:10:54 | mibus | Mem: 31490048 29339648 2150400 0 335872 8404992 |
| 12:10:54 | mibus | Swap: 0 0 0 |
| 12:10:54 | mibus | MemTotal: 30752 kB |
| 12:10:54 | mibus | MemFree: 2100 kB |
| 12:10:54 | mibus | MemShared: 0 kB |
| 12:10:56 | mibus | Buffers: 328 kB |
| 12:10:58 | mibus | Cached: 8208 kB |
| 12:11:00 | mibus | SwapCached: 0 kB |
| 12:11:02 | mibus | Active: 8556 kB |
| 12:11:04 | mibus | Inactive: 15208 kB |
| 12:11:06 | mibus | HighTotal: 0 kB |
| 12:11:10 | mibus | HighFree: 0 kB |
| 12:11:12 | mibus | LowTotal: 30752 kB |
| 12:11:14 | mibus | LowFree: 2100 kB |
| 12:11:16 | mibus | SwapTotal: 0 kB |
| 12:11:18 | mibus | SwapFree: 0 kB |
| 12:11:20 | mibus | ~ $ |
| 12:11:41 | pb_ | hmm, the suggestion of making appmgr transparent is an interesting one. |
| 12:11:53 | mibus | mmm |
| 12:11:57 | mibus | but IMHO evil |
| 12:12:06 | mibus | cos surely then the scrollbars etc. would be too? |
| 12:12:15 | mibus | how is it done in gpe-login etc.? |
| 12:12:27 | pb_ | no, only the window background is usually transparent. |
| 12:12:39 | pb_ | see ownerinfo for example, it has scrollbars too. |
| 12:13:26 | mibus | hmm |
| 12:13:30 | mibus | i gotta check that out then |
| 12:13:34 | mibus | gtk2 will do it too? |
| 12:14:39 | mibus | yay, new nmf module |
| 12:14:42 | mibus | well that only took an hour ;-) |
| 12:15:31 | mibus | got any good keywords to look for? |
| 12:15:32 | pb_ | yeah, gtk2 can be transparent also, it's an X feature. |
| 12:15:38 | pb_ | keywords for what? |
| 12:15:44 | mibus | dont worry |
| 12:16:06 | mibus | gdk_window_set_back_pixmap (GTK_VIEWPORT (widget)->view_window, NULL, TRUE); |
| 12:16:10 | mibus | seems to be the magic code |
| 12:16:21 | pb_ | Yeah. |
| 12:16:34 | pb_ | The TRUE is for parent relative, that's the important bit. |
| 12:19:08 | mibus | meh |
| 12:19:20 | mibus | im gonna have to make the viewport myself in the tab creation now |
| 12:19:22 | mibus | :( |
| 12:21:11 | mibus | widget->style->bg_pixmap[GTK_STATE_NORMAL] = (GdkPixmap*) GDK_PARENT_RELATIVE; |
| 12:21:12 | mibus | gtk_widget_set_style (GTK_WIDGET (widget), widget->style); |
| 12:21:14 | mibus | whats that for then? |
| 12:21:38 | mibus | and do i have to make the containing window PARENT_REL... now? |
| 12:21:58 | pb_ | All the parent windows have to be parentrelative. |
| 12:22:35 | pb_ | The style stuff is a different, possibly more correct, way to set the background to transparent. |
| 12:23:36 | mibus | 'cos ownerinfo does *both* |
| 12:24:05 | pb_ | yeah |
| 12:24:14 | mibus | crap i cant do it :-( |
| 12:24:26 | pb_ | there is no way to get at the viewport view_window other than the gdk_window_set_back_pixmap thing. |
| 12:25:10 | mibus | you arent supposed to realize() a widget until its in a toplevel container |
| 12:25:18 | mibus | and mine isnt until after the tab is filled :( |
| 12:25:29 | mibus | by which time its hard to get at the viewport :( |
| 12:25:54 | mibus | so surely ill have to set the table, viewport, notebook, and window all to PARENT_REL |
| 12:26:04 | mibus | in which case almost everything will be transparent, no? |
| 12:27:58 | pb_ | if you aren't using a viewport, you can ignore that. |
| 12:30:31 | mibus | well i still have to set it all |
| 12:30:37 | mibus | cos i have to put the gtktable in a viewport |
| 12:30:46 | mibus | either manually or with gtk_scrolledwindow_add_with_viewport |
| 12:34:52 | mibus | pb_: well i can make the background behind the notebook black ;-) |
| 12:35:17 | mibus | pb_: and if i switch tabs |
| 12:35:25 | mibus | pb_: it makes the area behind the icons black too... |
| 12:36:56 | mibus | pb_: so its a start ;-) |
| 12:38:34 | mibus | pb_: well now i can make it all black to start with... |
| 12:38:57 | mibus | so WHAT AM I MISSING?? GRR |
| 12:41:01 | mibus | PHIL |
| 12:41:07 | mibus | GET YER BUT BACK HERE! :P |
| 12:41:37 | mallum | mibus: parentrelative is still a hack, X does not really do transparency |
| 12:42:18 | mibus | mallum: but its BLACK. why cant it be blue? ;-) |
| 12:42:35 | mibus | mallum: and whatever happened to the alpha stuff keith (?) was working on? |
| 12:42:48 | mallum | um, its still happening ? |
| 12:42:58 | mibus | mallum: but i want it NOOOOWWWWW ;-) |
| 12:43:05 | mibus | well gtk1 doesnt help any :D |
| 12:43:14 | mallum | patience is a virtue ;-P |
| 12:43:33 | mibus | noone ever said i had virtues ;-) |
| 12:44:17 | mibus | mallum: any ideas on the whole black thing tho? |
| 12:44:27 | mallum | nope |
| 12:44:27 | mibus | mallum: ? |
| 12:44:51 | mallum | prolly matchbox box's fault ;) |
| 12:44:58 | mibus | yeah :P |
| 12:45:13 | mibus | so does fspanel show tasks on CVS matchbox now? |
| 12:45:23 | mallum | maybe ;) |
| 12:45:31 | mibus | !!!:P |
| 12:45:38 | mallum | eggnog |
| 12:45:42 | mibus | !!!!!!!!!!!! |
| 12:45:45 | mallum | :) |
| 12:45:48 | mallum | :) |
| 12:45:50 | mallum | :) |
| 12:45:52 | mibus | if i lived near you id come over and bop you one :P |
| 12:45:55 | mallum | um yes it does |
| 12:45:57 | mibus | :D |
| 12:46:06 | mallum | :{) |
| 12:46:11 | mibus | so why did the task-getting code not work? |
| 12:46:17 | mallum | >:{] |
| 12:46:17 | mibus | was it just the workspaces thing? |
| 12:46:20 | mallum | yeah |
| 12:46:20 | mibus | :P |
| 12:46:29 | mibus | how would that affect my xsi.c? |
| 12:46:39 | mallum | know shouldn't do |
| 12:46:54 | mallum | you dont really need to worry about workspaces I guess |
| 12:47:31 | mibus | mmm but afaik the code id written didnt |
| 12:47:34 | mibus | hence my confusion |
| 12:47:40 | mallum | let me have a look |
| 12:47:48 | mibus | thanks muchly |
| 12:47:53 | mibus | >;) |
| 12:48:36 | mibus | mallum: its matchbox's fault!!! displayed on my desktop, gpe-ownerinfo does nothing special |
| 12:48:41 | mibus | with transparency on i mean |
| 12:48:46 | mibus | but on my iPAQs display |
| 12:48:48 | mibus | its all black! |
| 12:48:48 | mibus | aha! |
| 12:48:53 | mibus | its YOUR fault :P |
| 12:49:14 | mallum | yeah its cos your settin g your background to that of the parent |
| 12:49:19 | mallum | the parent is a matchbox window |
| 12:49:23 | mallum | it is black |
| 12:49:25 | mibus | ummm |
| 12:49:27 | mibus | ah |
| 12:49:37 | mibus | so... how can i make it transparent too? |
| 12:50:09 | mallum | other wm's set there 'frame' windows to parentrelative too ( IIRC ) so the hack 'works' |
| 12:50:16 | mibus | ah |
| 12:50:18 | mibus | mmm |
| 12:50:30 | mibus | well its probably possible to set a pixmap to a gtkwidget somehow |
| 12:50:35 | mibus | so it shows thru to that instead |
| 12:50:41 | mibus | hmmm, gotta look it up... |
| 12:51:13 | pb_ | um? |
| 12:51:38 | mibus | pb_: transparency doesnt work on the iPAQ anyway |
| 12:51:43 | mibus | pb_: its all mallum's fault :D |
| 12:52:01 | pb_ | well, fix it. |
| 12:52:15 | mibus | pb_: i dont think he's gonna |
| 12:52:23 | mibus | pb_: i think he's got a new canned rant to test out ;-) |
| 12:52:24 | mibus | :P |
| 12:52:40 | pb_ | no, I meant _you_ fix it. |
| 12:52:50 | mibus | pb_: can i assign a pixmap to the background of a gtkwindow? |
| 12:52:53 | mibus | pb_: um, huh? |
| 12:52:55 | mibus | pb_: ME? |
| 12:52:58 | pb_ | yes. |
| 12:53:00 | mallum | mibus: yeah YOU |
| 12:53:03 | mibus | pb_: how? what? who? where? when? why? whichc? |
| 12:53:11 | mibus | aaah, theyre ganging up on me! |
| 12:53:12 | pb_ | mibus: cvs co apps/matchbox |
| 12:53:16 | mibus | runs and hides in a corner |
| 12:53:41 | mibus | hmm |
| 12:53:52 | mibus | GtkStyle has a GdkPixmap called *bg_pixmap[5] |
| 12:54:02 | mibus | sounds appropriate? |
| 12:54:18 | pb_ | <mibus> widget->style->bg_pixmap[GTK_STATE_NORMAL] = (GdkPixmap*) GDK_PARENT_RELATIVE; |
| 12:54:23 | pb_ | remember when you said that? |
| 12:54:25 | mibus | now i just gotta get a pixmap from somewhere |
| 12:54:26 | mallum | mibus: it did used to have the hack in it but IIRC I removed it cos it caused a little extra flicker and I thought 'nobody will ever want to run a trans aterm as it will look silly' |
| 12:54:31 | mibus | yeah yeah :P |
| 12:54:43 | mibus | mallum: well it looks sillier black :P |
| 12:54:51 | mallum | mibus: your app does |
| 12:54:52 | pb_ | mallum: why does it cause any more flicker? |
| 12:54:57 | mibus | theres no point if i can set a proper pixmap |
| 12:55:45 | mallum | pb_: cant remember, this was ages ago before a big cleanup ... it may have even just been cleaned up |
| 12:56:49 | mibus | hey wow |
| 12:56:52 | mibus | it works ;-) |
| 12:57:03 | mibus | just set it to an actually loaded GdkPixmap |
| 12:57:06 | mibus | i suck :D |
| 12:57:17 | mibus | and it makes a tiled background image |
| 12:57:22 | mibus | now i just gotta find a nice PNG |
| 12:57:22 | mibus | :D |
| 12:57:42 | mibus | libgpewidget makes the icons alpha-blended on, too... :D |
| 12:57:56 | mibus | rockin rockin rockin |
| 12:57:56 | mibus | :D |
| 12:58:27 | pb_ | heh |
| 13:00:15 | mibus | ill scap a pic in a few mins |
| 13:00:33 | mibus | (cant wait to show this to my buddy who's a linux-on-ipaq convert) |
| 13:01:19 | mallum | well its pretty trivial for me to add parentrelative back into matchbox, though its probably a better idea to grab the rootwindows pixmap id and use that as a background to your app ( as this is all parentrelative will do anyway ) |
| 13:02:04 | mibus | can i use the rootwindow pixmap as a GdkPixmap or do i have to convert first? |
| 13:04:06 | mibus | hmm |
| 13:04:25 | mibus | setting the bg of the viewport seems to have evil effects on things well outside the viewport |
| 13:05:17 | mibus | mallum: actually itd be good to have transparency cos then the recent-apps list can blend in to the dock :D |
| 13:23:01 | gazer | looks in. |
| 13:23:35 | gazer | grr. gtk_scrolled_* is being annoying. |
| 13:24:33 | mibus | hi gazer |
| 13:24:35 | mibus | 'night ppls |
| 13:24:37 | mibus | im going to bed |
| 13:24:53 | mibus | im posting a scap of a background-capable appmgr first |
| 13:25:05 | mibus | be warned, its hideous 'cos i cant make it look the way i want :( |
| 13:25:29 | mibus | ok scaps finished |
| 13:25:31 | mibus | bye ppls |
| 13:30:23 | gazer | anyone know how to add a key listener to a gtk-scrolled widget? |
| 13:30:32 | gazer | will it not act as a key source? |
| 13:34:38 | pb_ | pardon? |
| 13:36:20 | gazer | so here's what I want: |
| 13:36:54 | gazer | the standard GPE app has a toolbar on top and then a gtk_scrolled widget with the information within it. |
| 13:37:03 | gazer | (cf gpe-todo and gpe-memo) |
| 13:37:26 | pb_ | yup |
| 13:37:39 | gazer | the problem is that the gtk_scrolled widget doesn't look at the joypad. |
| 13:37:57 | gazer | I want to press "down" on joypad and scroll down a line. |
| 13:38:18 | gazer | gpe-appmgr does this by setting a global listener, which is ugly. |
| 13:38:44 | gazer | So my approach is to wrap the gtk_scrolled_new call so that it returns a gtk_scrolled widget with an installed keylistener. |
| 13:38:49 | gazer | but that isn't working. |
| 13:39:23 | gazer | I'm guessng that gtk_scrolled doesn't accept a key listener. |
| 13:39:28 | gazer | guessing, rather. |
| 13:39:49 | gazer | any clue? |
| 13:55:18 | gazer | hmm. how do I determine what type a gtk widget is pragmatically. |
| 14:13:52 | gazer | grr. this isn't working. I'm doing something wrong. |
| 14:17:16 | pb_ | GTK_OBJECT_TYPE (object) will get you the type, I think. |
| 14:17:29 | pb_ | (sorry, was on the phone for ages there) |
| 14:18:21 | gazer | I got the type, and figured out how to print it. |
| 14:18:44 | gazer | I want to put a key handler on the gtk_scrollable, but the handler is never getting called. |
| 14:18:45 | pb_ | ah right |
| 14:19:04 | gazer | I tried putting a global key snooper (cutting and pasting from gpe-appmgr/main.c), but that isn't working either. |
| 14:19:13 | pb_ | hmm. |
| 14:19:30 | gazer | I'm probably doing something dumb. Can the gtk_scrollabe use a keyhandler at all? |
| 14:19:44 | pb_ | Dunno. You'd have to check the source. |
| 14:24:48 | gazer | yeah, I'll poke at it later. Should get some writing done today. |
| 14:50:16 | mallum | pb_: have you done anything more with the evil usb plans ? |
| 15:01:44 | pb_ | mallum: yeah, waiting for the pcb layout guy to start work. |
| 15:02:43 | mallum | pb_: ooh cool :) |
| 18:35:27 | pb_ | ibot, message for mibus: gpe-nmf-ogg_0.06 is uploaded |
| 18:36:49 | cmarqu_ | goes and installs it |
| 18:37:01 | pb_ | ha. |
| 18:37:23 | pb_ | the only change is a tweak to do with id3 information (or whatever the ogg equivalent is called). |
| 18:37:43 | cmarqu_ | mibus' scap is really awkward |
| 18:37:54 | pb_ | which one's that? |
| 18:37:56 | pb_ | checks scap |
| 18:38:07 | cmarqu_ | the transparent gpe-appmgr |
| 18:38:19 | pb_ | ooh heck |
| 18:38:21 | pb_ | yes |
| 18:38:45 | pb_ | kind of nasty, the old alpha-blended icons strike again. |
| 18:38:57 | cmarqu_ | Yeah... |
| 18:39:18 | pb_ | hmm. |
| 18:39:25 | pb_ | kind of an insoluble problem, in general. |
| 18:39:39 | cmarqu_ | :( |
| 18:39:44 | pb_ | I could add an option to turn alpha off, but that will make the icons look ragged round the edges. |
| 18:40:04 | pb_ | Or appmgr could do "proper" compositing, but that would require it to have a huge pixmap. |
| 18:40:15 | cmarqu_ | The OPIE guys will laugh at us... |
| 18:40:39 | pb_ | The OPIE guys laugh at me all the time, I'm used to it by now. |
| 18:40:55 | cmarqu_ | GTK2 would still be the same I suppose? |
| 18:41:01 | pb_ | Yeah. |
| 18:41:23 | pb_ | It would need an X server extension to fix it properly. |
| 18:41:34 | pb_ | Maybe one already exists, you would have to check with one of the X-Men. |
| 18:41:44 | cmarqu_ | How does Gnome 2 do its desktop icons, for instance? |
| 18:42:03 | pb_ | Good question. I have no idea. |
| 18:42:17 | pb_ | It might just keep one massive pixmap for the whole desktop. If memory is no object, it's easy. |
| 18:42:30 | cmarqu_ | Hmm |
| 18:42:55 | pb_ | Would be interesting to find out, if someone wanted to go digging in Nautilus or whatever. |
| 18:44:40 | pb_ | Oh, maybe you could do it with XRender, actually. |
| 18:50:14 | cmarqu_ | mibus will figure it out |
| 18:50:31 | BBrox | Well, the giant pixmap thing is the easiest way to fix it :-) |
| 18:50:44 | BBrox | And you can also quickly refresh the screen using the pixmap backbuffer |
| 18:51:07 | cmarqu_ | sees his free RAM vanish |
| 18:52:30 | BBrox | Well, proper compositing would also make RAM vanish pretty quickly... |
| 18:54:38 | BBrox | And it would also mean that you need to display things always in the same order :-) |
| 20:20:28 | PaxAnima | hello |
| 20:25:53 | prpplague | howdy |
| 20:26:17 | PaxAnima | prpplague: are you a gpe user or developer? |
| 20:37:07 | pb_ | dc_! |
| 20:37:08 | dc_ | hum de dum |
| 20:37:10 | dc_ | pb_: :/ |
| 20:37:19 | pb_ | ha. |
| 20:37:46 | dc_ | to much work :\ |
| 20:37:55 | pb_ | school-type work? |
| 20:38:34 | prpplague | PaxAnima actually neither |
| 20:38:43 | dc_ | and now I have to wake up at 4am on wendsday to race some dodgy gokart |
| 20:38:51 | pb_ | prpplague: just some kind of groupie, huh? |
| 20:39:11 | prpplague | pb_: lol, well sorta, i'm more into using matchbox i guess |
| 20:39:20 | pb_ | dc_: crumbs, karting in the middle of the night? |
| 20:39:37 | prpplague | i've got alot of custom fltk apps that previously used icewm with |
| 20:39:42 | prpplague | now i use mb |
| 20:39:51 | pb_ | ah. |
| 20:40:01 | dc_ | pb_: at 7am, but it's a few hours away :/ |
| 20:40:10 | pb_ | dc_: ah I see. |
| 20:40:20 | pb_ | dc_: still, might be fun I guess. |
| 20:40:25 | dc_ | plus school work. |
| 20:40:26 | prpplague | pb_: i remember a rather interesting late night conversation with mallum about the problems with icewm |
| 20:40:30 | dc_ | but yeah, it shoul be... |
| 20:40:44 | prpplague | pb_: a couple weeks later mallum pops up again with matchbox |
| 20:40:51 | pb_ | heh :-) |
| 20:41:05 | dc_ | hehe |
| 20:41:07 | pb_ | dc_: think I'm supposed to be going to some kind of go-kart event in November sometime. |
| 20:41:18 | dc_ | heh cool. |
| 20:44:22 | pb_ | dc_: did you see mibus' dubious transparent appmgr on scap? |
| 20:45:56 | dc_ | oh cool. |
| 20:45:59 | dc_ | pb_: looks good :) |
| 20:46:05 | dc_ | qwll |
| 20:46:07 | dc_ | well |
| 20:46:10 | dc_ | 'interesting' |
| 20:46:19 | pb_ | yeah. |
| 20:46:28 | pb_ | the same old problem with the alpha blended icons. :-} |
| 20:46:37 | dc_ | it seems :[ |
| 20:48:38 | pb_ | still, colin seems optimistic. |
| 20:48:41 | pb_ | <cmarqu_> mibus will figure it out |
| 20:48:53 | cmarqu_ | :) |
| 20:49:49 | dc_ | hehe |
| 20:49:54 | dc_ | he's a good guy |
| 20:50:00 | pb_ | yeah. |
| 20:50:11 | pb_ | did you tell him where to send your jumper? |
| 20:50:16 | pb_ | he seemed to be getting quite upset about that. |
| 20:50:25 | dc_ | oh dear. |
| 20:50:29 | dc_ | I totaly forgot :/ |
| 20:50:33 | pb_ | ha. |
| 20:50:51 | pb_ | well, the weather seems to be turning cold, you'll need it soon. :-/ |
| 20:51:50 | dc_ | ibot message for mibus: hey, my address is Funnels Farm, Down Street, Nutley, TN22 3LG, England :-D send the jumper when you get some time. |
| 20:52:01 | dc_ | there we go |
| 20:52:08 | pb_ | heh |
| 20:52:13 | pb_ | good-o |
| 20:52:18 | dc_ | oh dear. |
| 20:52:25 | pb_ | what kind of address is "Funnels Farm" though? :-) |
| 20:52:42 | dc_ | ibot message for mibus: and my county is Easy Sussex. |
| 20:52:46 | pb_ | ! |
| 20:52:53 | dc_ | ibot message for mibus: East Sussex!! |
| 20:53:01 | pb_ | :-) |
| 20:53:05 | dc_ | oops ;D |
| 20:53:06 | dc_ | hehe |
| 20:53:18 | pb_ | ah, top comedy. |
| 20:53:21 | dc_ | if only it really easy 'easy' sussex :] |
| 20:53:32 | pb_ | mmm. |
| 20:53:41 | dc_ | if only it really was 'easy' sussex :] |
| 20:53:42 | dc_ | bah |
| 20:53:47 | pb_ | hey, you should update your coordinates on colin's map too. |
| 20:53:51 | dc_ | I lack rice at this time or night |
| 20:54:02 | dc_ | yes rice.... |
| 20:54:08 | dc_ | pb_: hmm, I need a GPS... |
| 20:54:18 | pb_ | dc_: use a map? |
| 20:54:35 | cmarqu_ | Use mapblast or mapquest |
| 20:54:48 | cmarqu_ | If it's not on that link in that file |
| 20:54:49 | dc_ | pb_: but, but...that could be too easy.. |
| 20:54:52 | dc_ | and clever. |
| 20:55:20 | dc_ | heh, maby on thursday I'll have some time |
| 20:55:21 | pb_ | dc_: at the moment the contacts page has you further north than mallum. |
| 20:55:26 | dc_ | need siome sleep now. |
| 20:55:30 | dc_ | pb_: heh, not good. |
| 20:56:02 | dc_ | talk on thursday :] |
| 20:56:07 | pb_ | ok. |
| 20:56:09 | pb_ | night. |
| 20:56:10 | dc_ | have a nice day etc... |
| 21:37:36 | jsmanrique | hi! |
| 21:38:32 | moray | hello... |
| 21:39:31 | PaxAnima | hi |
| 21:41:29 | moray | hm, has anyone had midi playing working on their ipaq? |
| 21:41:54 | moray | presumably timidity ought to work, though I don't know if it would be too slow |
| 21:42:03 | PaxAnima | never tried |
| 21:42:42 | PaxAnima | but woldn't the patches for timitidy suck up space? |
| 21:43:03 | moray | yes, but I'd only need one... |
| 21:43:31 | moray | I need to play a couple of MIDI files in a presentation tomorrow, is all, and was wondering if I could just use the ipaq, since it would be lighter/more amusing... |
| 21:43:47 | PaxAnima | he |
| 21:43:53 | PaxAnima | why not play an ogg? |
| 21:44:03 | PaxAnima | render the midi file to wave first? |
| 21:46:18 | moray | possible, but I have a couple of hundred MIDI files I'd *potentially* want to play, though only 3 definite ones |
| 21:46:47 | PaxAnima | oh |
| 21:47:16 | PaxAnima | what kinda of midi files is it, since you only need 1 patch? |
| 21:47:32 | moray | http://www.srcf.ucam.org/~mma29/2002/harmony/ |
| 21:48:22 | PaxAnima | moray: timitidy is in the arm debian distro |
| 21:49:27 | PaxAnima | I want a serial port for the ipaq, so that I can hook it to my synth module |
| 21:50:11 | moray | PaxAnima: your ipaq does have a serial port, surely? |
| 21:50:42 | PaxAnima | moray: yes, but I need the adapter from the ipaq plug to serial |
| 21:51:08 | moray | ah yes, and presumably to serial the other way round from on the back of a computer |
| 21:52:12 | PaxAnima | I know there is an adapter for 36xxx, but haven't seen one for 38xx |
| 21:52:22 | moray | right |
| 21:53:00 | PaxAnima | what's the website thing? |
| 21:53:28 | PaxAnima | (just read "Hiden Markow Models" and got a bit interested..;) |
| 22:11:45 | moray | PaxAnima: that's what I was working on this summer... |
| 22:11:57 | PaxAnima | what does it? |
| 22:12:12 | moray | PaxAnima: machine learning stuff (=artificial intelligence...) - the particular problem is getting the computer to learn to harmonise music |
| 22:12:28 | PaxAnima | for what purpose? |
| 22:13:14 | moray | PaxAnima: to see if you can, really... - research rather than anything directly useful (though more fun than some things, since it does produce music) |
| 22:13:21 | PaxAnima | oh? |
| 22:14:07 | PaxAnima | so give the system a MIDI file and it adds harmony to it? |
| 22:15:11 | moray | PaxAnima: it doesn't take MIDI files directly as input usually, but you can do that, yes |
| 22:15:30 | PaxAnima | moray: are there any good, easy to read, papers or web pages on hidden markow models? |
| 22:15:34 | moray | but the point isn't so much that it can harmonise (which it can), but that it learns to do so from looking at examples |
| 22:16:01 | PaxAnima | aha |
| 22:16:59 | moray | PaxAnima: erm, I'm not sure about easy to read... - a good thing to look at is Rabiner 1989 'A tutorial on hidden markov models and selected applications in speech recognition.' |
| 22:17:16 | moray | though I think my description in that dissertation of the formula needed is clearer than his one ;) |
| 22:17:30 | moray | ('formulae', sorry, there are quite a few) |
| 22:18:40 | PaxAnima | I'm on the last year of my master, and often read papers which use HMM in one way or the other... I've had some in statistical image processing course, but not that I understand it too well from that |
| 22:24:02 | moray | there's a copy of the paper I mentioned at http://www.cs.berkeley.edu/~murphyk/Bayes/rabiner.pdf - that's about the best thing I know of |
| 22:24:34 | moray | (only the first part is relevant; then it talks more specifically about speech stuff) |
| 22:24:49 | PaxAnima | thanks |
| 22:25:59 | PaxAnima | moray: your sony laptop looks cute |
| 22:26:51 | moray | PaxAnima: yes, it is, though I'm fed up with sony laptops for being unreliable (all the ones I know of have had their batteries die very quickly, and several have had other problems too) |
| 22:27:34 | PaxAnima | oh? how poor is the battery performance? |
| 22:28:42 | moray | the battery in that one is completely dead - over the first year or so it gradually stopped holding charge, then died. to start with it would last maybe a couple of hours if you were lucky |
| 22:28:51 | PaxAnima | (the sony laptop makes my lifebook look big;) |
| 22:29:13 | PaxAnima | sounds bad |
| 22:29:30 | moray | of course, the battery isn't covered by the warranty ;) |
| 22:34:10 | PaxAnima | who is doing gpe-calendar? mccarthy? |
| 22:37:23 | cmarqu_ | PaxAnima: pb, but mccarthy did some work on it recently |
| 22:37:32 | PaxAnima | ok |
| 22:39:57 | PaxAnima | just wanted to get some ideas on the handling of repeating evnts.... kinda "stuck" on the calendar app in my "little" PIM project |
| 22:42:37 | PaxAnima | maybe some sensless shooting around in Return to castle wolfenstein can clear my mind.... |
| 22:43:10 | cmarqu_ | can't get the UT2003 demo to run |
| 22:52:00 | PaxAnima | buhu... the game crashed |
| 22:52:19 | PaxAnima | cmarqu_: what graphics card do you have? |
| 22:55:06 | cmarqu_ | Geforce 1 |
| 22:56:21 | PaxAnima | oh.. need a faster card than GF1 to run the game |
| 22:56:59 | cmarqu_ | Guess so. I never really play anyway though, I just wanted to look at the graphics |
| 22:57:10 | cmarqu_ | RTCW runs fine OTOH |
| 22:58:13 | PaxAnima | on and old GF1? |
| 22:58:28 | cmarqu_ | Yes. With the nvidia binary drivers |
| 22:58:48 | cmarqu_ | It's not balzing speed, but playable. |
| 22:58:57 | PaxAnima | ok |
| 22:59:16 | cmarqu_ | UT2003 will be unplayable I'm sure |
| 22:59:26 | cmarqu_ | If I ever get it to run |
| 22:59:45 | cmarqu_ | Anyway, bedtime |
| 22:59:49 | cmarqu_ | Night all |
| 22:59:52 | PaxAnima | good night |