| 00:26.40 | dc_ | sleep :] |
| 00:27.08 | PaxAnima | good idea |
| 00:27.17 | PaxAnima | ;) |
| 00:27.22 | dc_ | :D |
| 00:27.24 | dc_ | night then |
| 00:27.29 | pb_ | dc_: night |
| 00:27.32 | moray | dc_: ha - good night. |
| 00:27.47 | dc_ | moray: =P |
| 00:28.09 | moray | it appears that in right-to-left locales GTK just reverses the packing order, in case I wasn't the only one not to have spotted this yet. |
| 00:28.31 | moray | so you don't need to make any special effort about labels etc. for things to work |
| 00:29.26 | pb_ | yes, indeed. |
| 00:29.38 | pb_ | I need to figure out how to make that happen dynamically now, for gpe-ownerinfo and the like. |
| 00:30.10 | moray | yes - I suppose the display migration code might help here, if it means we already know how to rebuild an existing window? |
| 00:30.29 | moray | (although I don't know at what level that works?) |
| 00:30.59 | pb_ | The display migration stuff operates at a lower level, really. It just unrealizes all the widgets, and then realizes them again on the new screen. |
| 00:31.19 | moray | right |
| 00:31.51 | pb_ | There's a gtk_widget_set_default_direction (or some such thing), which looks like it might be able to change the packing order for boxes on the fly. |
| 00:32.01 | moray | aha |
| 00:32.26 | pb_ | So, possibly, we just need a way to figure out what the right direction is for the current locale (which I guess you can get from nl_langinfo somehow) and set the flag appropriately. |
| 00:34.25 | pb_ | I haven't invested too much effort into this yet because gpe-login doesn't actually have a working locale selector at the moment. I think paul was planning to work on that next week, though, so things may change. |
| 00:34.36 | moray | ok. |
| 00:57.39 | moray | pb_: more specifically, we probably want gtk_widget_set_default_direction_recurse |
| 00:57.57 | pb_ | moray: sounds like it. |
| 00:58.31 | moray | it sets the private flag then emits a signal, which I assume gets on with changing things (just following that up though) |
| 00:58.42 | pb_ | right, cool. |
| 02:53.21 | nelson | Anybody know why gpe2 doesn't install ssh? |
| 03:26.19 | *** join/#gpe cantona (~cantona@vp185248.hk.uac1.hknet.com) |
| 03:26.32 | pigeon | ... |
| 03:27.49 | cantona | ... |
| 03:27.57 | pigeon | :o |
| 03:28.27 | *** join/#gpe walters (walters@verbum.org) |
| 03:29.40 | cantona | so wonder u are not idling |
| 03:29.57 | pigeon | Just came back from lunch. |
| 03:30.39 | cantona | oh.. |
| 03:30.43 | cantona | lunch time too... |
| 03:30.49 | cantona | cya |
| 04:04.06 | *** join/#gpe flypiper (~netwinder@12-222-141-210.client.insightBB.com) |
| 04:15.27 | *** join/#gpe Turbonium (~W8TVI@1114107486.mi.dial.hexcom.net) |
| 04:36.12 | *** join/#gpe pattieja (~pattieja@www.pcxperience.com) |
| 07:28.48 | *** join/#gpe Turbonium_ (~W8TVI@1114104869.mi.dial.hexcom.net) |
| 08:15.20 | *** part/#gpe Turbonium_ (~W8TVI@1114104869.mi.dial.hexcom.net) |
| 08:34.51 | *** join/#gpe florian (~fuchs@dialin-145-254-099-182.arcor-ip.net) |
| 08:34.53 | florian | hi all! |
| 09:35.29 | *** join/#gpe florian (~fuchs@dialin-145-254-099-241.arcor-ip.net) |
| 09:51.11 | *** join/#gpe dc_ (~dc@modem-72-63-60-62.vip.uk.com) |
| 10:13.48 | *** join/#gpe PaxAnima (~petter@v058b.studby.ntnu.no) |
| 10:14.56 | PaxAnima | morning |
| 10:21.09 | *** join/#gpe mallum (~mallum@pc-80-193-218-21-hw.blueyonder.co.uk) |
| 10:21.49 | *** join/#gpe florian (~fuchs@dialin-145-254-097-254.arcor-ip.net) |
| 10:21.57 | florian | hi |
| 10:24.07 | mallum | hey |
| 10:24.10 | *** join/#gpe paul ([a7okYHH0F@hibernia.jakma.org) |
| 10:25.15 | *** join/#gpe pb_ (~pb_@dsl-62-3-66-201.zen.co.uk) |
| 10:25.54 | paul | hi pb |
| 10:26.51 | pb_ | hi paul |
| 10:28.09 | florian | pb_: hello... |
| 10:28.09 | *** join/#gpe TheMasterMind1 (foobar@h-69-3-1-47.MCLNVA23.covad.net) |
| 10:28.13 | pb_ | hi florian |
| 10:28.58 | florian | pb_: filed this ipkg issue in bugzilla :-) |
| 10:29.21 | pb_ | florian: cool. |
| 10:29.42 | florian | now i'll downgrade libgpewidget "by hand" |
| 10:30.02 | pb_ | florian: if you want to make it work quickly, just changing gpe-conf to pass in valid argc and argv pointers would probably do the trick. |
| 10:30.42 | pb_ | or, yes, downgrading libgpewidget will do the trick as well, though you may find that ipkg immediately upgrades it again for you. |
| 10:32.26 | florian | hmm... not if i don't tell him i downgraded something :-) |
| 10:32.54 | florian | but you're right - i'll look at gpe-conf first |
| 10:34.10 | pb_ | oh, I see, you were planning to just overwrite libgpewidget.so.1 behind ipkg's back? yes, that would work. |
| 10:34.52 | paul | is this to do with gpe-conf being segv happy at the moment? |
| 10:35.03 | pb_ | yup |
| 10:35.20 | pb_ | it's calling gpe_application_init (NULL, NULL) which the latest libgpewidget doesn't like. |
| 10:36.37 | paul | ah |
| 10:37.01 | paul | who maintains gp-conf wlan? |
| 10:38.01 | mallum | pb_: hi, do you know of any info on building a custom kernel netinst cd ? |
| 10:38.05 | pb_ | florian, I think |
| 10:38.41 | pb_ | mallum: there are some instructions somewhere, maybe even in the SYSLINUX menus. |
| 10:38.52 | pb_ | mallum: if you hold on a few minutes I will try to find them. |
| 10:38.57 | mallum | pb_: cool, thanks |
| 10:39.16 | mallum | pb_: the mandrake installer works with the keyboard on this tablet PC :( |
| 10:39.52 | florian | paul: i think ole, but nils and i know something about it too |
| 10:40.12 | dc_ | hum |
| 10:40.18 | dc_ | mallum: c700 :] |
| 10:40.27 | dc_ | mallum: oh dear. |
| 10:40.35 | dc_ | mallum: not good news for debian then. |
| 10:40.37 | paul | florian: i havnt looked too closely (cause gconf isnt working) but i think it may not always write out correct wireless.opt file. |
| 10:40.44 | pb_ | mallum: aha. |
| 10:41.04 | mallum | dc_: I *think* the mandrake installer has ehci support turned on, whilst debian does not |
| 10:41.14 | dc_ | mallum: ahh :) |
| 10:41.36 | pb_ | mallum: ehci is generally only used for USB 2.0 devices, and I doubt your keyboard is one of those. |
| 10:41.52 | paul | florian: eg, start out with an empty wireless.opt file - run wlancfg - create a few schemes/confs - save, and have a look at the file. its missing the 'case $ADDRESS in' line (iirc). |
| 10:43.24 | mallum | pb_: yes, but mandrake seems to also have a stock kernel and something is going wrong in the handover from bios kbd control to kernel usb control - missing ehci ( the tablet pc uses usb2.0 ) is the only thing I can think of thats an obvious difference |
| 10:43.32 | pb_ | mallum: I think the information you need is in the installation manual. Look for the chapter entitled "Technical Information on the Boot Floppies". |
| 10:43.52 | mallum | pb_: cool, thanks will check |
| 10:44.15 | florian | paul: this is possible... ole told me he is the same opinion... |
| 10:44.30 | mallum | dc_: so did you order a c700 btw ? |
| 10:44.43 | florian | paul: but if you start with the default config file it should work |
| 10:44.46 | paul | florian: and it doesnt always correctly parse existing wireless.opt files. |
| 10:45.08 | florian | paul: do you know details? |
| 10:46.15 | paul | florian: i ran into this problem after hand editing wireless.opts then wlancfg and things not working. then i nulled out the file and ran wlancfg again and it didnt work because of the missing case. |
| 10:46.16 | florian | paul: it devides some sections... that ones with multiple ranges |
| 10:46.36 | paul | florian: i dont have details on the misparsing - was in a hurry to get it working really. :) |
| 10:47.50 | florian | paul: if the case construct is not there/commented out it definitely doesn't work. if you find another case, just mail me the config file |
| 10:48.11 | paul | its simply not there - wlancfg obviously expects it to be there. |
| 10:48.15 | dc_ | mallum: 'hopefully' today I will :D |
| 10:51.54 | florian | paul: yep, definitely... if you want to make it work at easy as possible use the default file and delete all entrys except the one with *,*,*,*, therefore you can use wlancfg, no need to do this by hand :-) |
| 10:52.19 | paul | florian: :) |
| 10:53.16 | *** join/#gpe dc_ (~dc@modem-72-63-60-62.vip.uk.com) |
| 10:54.23 | florian | bbl |
| 10:56.02 | paul | florian: though... perhaps wlancfg parser could be made less reliant on existing contents of file. ie read in and parse into an appropriate struct, then on write out from the contents of the structure. less fragile. (but you'd lose comments, etc..) |
| 11:14.44 | *** join/#gpe cmarqu_werk (www@vws1.kens.com) |
| 11:18.56 | pb_ | hi cmarqu_werk |
| 11:24.34 | cmarqu_werk | Hi. |
| 11:28.48 | mallum | cmarqu_werk: hello, I noticed reading the x4.3 release notes, it appears you maybe dont need to use ttmkfontdir any more and the included X font directory tools should do the job |
| 11:30.11 | cmarqu_werk | mallum: Ah. I'll have to read up more then. Still, setting the font path should be working. |
| 11:31.25 | cmarqu_werk | "mkfontscale", hmm. |
| 11:31.39 | cmarqu_werk | Now I need that as a binary. |
| 11:31.39 | mallum | cmarqu_werk: yes are you sure its just not working cos the font.dir file or whatever is in the wrong format ? |
| 11:33.24 | PaxAnima | mallum: my request for doc on the multisync mailinglist has resulted in an updated, and more documented, empty plugin example... but as I don't have a clue about how data is organised in GPE, and lack of vCard i/vCal, I can't do much.. maybe I'll try to do a plugin for my old, dusty PIM, for testing... |
| 11:34.21 | cmarqu_werk | mallum: Well, it works on the desktop at least. |
| 11:35.00 | cmarqu_werk | And the fonts.dir at least looks the same as my other fonts.dir files. |
| 11:40.47 | mallum | PaxAnima: I guess you need to hassle pb_ for that info |
| 11:42.46 | PaxAnima | but as I said, I could do a plugin for my PIM, as it has vCard/vCal as well a known (and simple) DB structure... |
| 11:44.58 | mallum | PaxAnima: is your PIM packaged ? |
| 11:45.04 | PaxAnima | no.. |
| 11:45.21 | PaxAnima | haven't bothered to do it.. my I guess I'll do it one day.. |
| 11:45.29 | PaxAnima | my = bur |
| 11:45.30 | PaxAnima | but |
| 11:45.47 | mallum | PaxAnima: :(, more people would get use then if it was plugged into gpe .. |
| 11:46.27 | PaxAnima | but it would be interesting to compare my DB schema to that of GPE... |
| 11:46.33 | dc_ | :} |
| 11:47.19 | PaxAnima | but I have a dream, that one day I'll sit down, clean up the code, make the data abstraction better and port it to GTK2.... ;) |
| 11:48.59 | mallum | ok ... :/ |
| 11:49.25 | PaxAnima | but at the moment, it's just a dream... |
| 11:50.49 | PaxAnima | but my PIM differ quite from the GPE PIM.. it's more Palm-ish |
| 11:50.57 | *** join/#gpe dc__ (~dc@62.60.119.16) |
| 11:51.40 | PaxAnima | and it's quite ugly, to be honest... |
| 11:57.32 | cmarqu_werk | PaxAnima: Probably doesn't follow the HIG, huh? |
| 11:57.50 | pb_ | cmarqu_werk: hehe |
| 11:58.56 | PaxAnima | cmarqu_werk: HIG? what HIG? ;) |
| 11:59.00 | *** join/#gpe TGG__ (~trillian@host213-122-57-109.in-addr.btopenworld.com) |
| 11:59.06 | mallum | uh oh , here we go ..... |
| 12:04.41 | TGG__ | +n |
| 12:05.01 | TGG__ | anyone know if GPE runs on the Zaurus? |
| 12:05.46 | pb_ | TGG__: not at present, but it is planned. |
| 12:06.26 | TGG__ | ok - the big Q , any idea how long........... |
| 12:08.39 | pb_ | Not exactly. It's mostly contingent on OE getting an updated XFree86. |
| 12:12.43 | *** part/#gpe TGG__ (~trillian@host213-122-57-109.in-addr.btopenworld.com) |
| 12:13.08 | *** join/#gpe florian (~fuchs@dialin-145-254-099-052.arcor-ip.net) |
| 12:23.58 | mallum | pb_: Im gonna start playing with xfree for OZ today |
| 12:31.23 | *** join/#gpe dc___ (~dc@modem-117-41-60-62.vip.uk.com) |
| 12:33.08 | pb_ | mallum: ah, cool |
| 12:34.17 | mallum | pb_: obviously Im gonna build new fam packages too |
| 12:34.49 | pb_ | great. |
| 12:47.30 | mallum | pb_: do the 3900 kernel have XSCALE_CACHE_ERRATA turned on ? |
| 12:47.55 | pb_ | Yes. |
| 12:48.51 | mallum | pb_: and they make quite a speed difference, but can be unstable - or is that untrue ? |
| 12:52.29 | pb_ | The errata mostly relate to writeback caching, and the workaround is basically to use writethrough at all times. So yeah, there is a speed hit, and yes, it can be somewhat unstable without. Exactly what the impact of the bug would be in real life is difficult to quantify. |
| 12:53.05 | mallum | pb_: ok, but you'd recommend turning it on ? |
| 12:53.16 | mallum | pb_: it appears the c700 has it off |
| 12:59.18 | pb_ | depends what chip the c700 uses. |
| 12:59.34 | pb_ | if it's a pxa255, the workarounds should be unnecessary. |
| 13:00.19 | pb_ | if it's a pxa250 then they may be required; personally I'd turn it on, but leaving it off is a valid choice as well. |
| 14:31.06 | *** join/#gpe dc__ (~dc@modem-140-44-60-62.vip.uk.com) |
| 14:55.54 | *** join/#gpe dc___ (~dc@modem-24-60-60-62.vip.uk.com) |
| 15:16.41 | *** join/#gpe prpplague (~JoeBob1@12.148.134.9) |
| 15:35.21 | *** join/#gpe florian (~fuchs@dialin-145-254-093-158.arcor-ip.net) |
| 15:35.25 | florian | re |
| 15:35.28 | pb_ | re florian |
| 15:50.10 | *** join/#gpe dc_ (~dc@modem-77-50-60-62.vip.uk.com) |
| 15:50.20 | dc_ | :] |
| 15:50.28 | pb_ | dc_: heh |
| 15:50.35 | dc_ | damn this php |
| 15:50.50 | pb_ | dc_: any progress on putting your popup stuff in libgpewidget? I will make a new release this evening to fix the gpe-conf breakage. |
| 15:51.06 | dc_ | pb_: sure, how about by 8pm? |
| 15:51.19 | dc_ | pb_: I'm currently on a working half term. |
| 15:51.22 | pb_ | dc_: cool |
| 15:51.33 | dc_ | wish it was ;0 |
| 15:51.38 | pb_ | dc_: no rush, just ping me when you are ready |
| 15:51.41 | dc_ | but atleast I get money |
| 15:51.42 | dc_ | :) |
| 15:51.47 | pb_ | heh |
| 15:51.47 | dc_ | and atleast it's code. |
| 15:51.56 | pb_ | quite so. |
| 15:52.10 | dc_ | pb_: I was kind of anoyed last time when working on the popup stuff. |
| 15:52.22 | dc_ | pb_: I guess I'm to young to worry about licences ;) |
| 15:52.53 | spung | dc_: wassup? |
| 15:53.08 | pb_ | dc_: :-} |
| 15:53.36 | dc_ | spung: I have to 'contact'...yes contact! the author of the gtk color selecton widget before it can be alowed to do into libgpewidget. |
| 15:54.00 | dc_ | pb_: but...can't you take lgpl code and put it in another lgpl lib without doing that? |
| 15:54.09 | pb_ | dc_: yes, but not gpl code |
| 15:54.18 | dc_ | pb_: isn't gtk lgpl then? |
| 15:54.20 | pb_ | dc_: not sure |
| 15:54.23 | dc_ | :\ |
| 15:54.28 | pb_ | dc_: I thought it was, though, yes. |
| 15:54.29 | florian | dc_: do you contact hom, or shall i? |
| 15:54.47 | florian | s/hom/him |
| 15:55.17 | dc_ | <PROTECTED> |
| 15:55.17 | dc_ | <PROTECTED> |
| 15:55.17 | dc_ | <PROTECTED> |
| 15:55.17 | dc_ | <PROTECTED> |
| 15:55.23 | dc_ | gtkcolorsel.c |
| 15:55.24 | dc_ | so... |
| 15:55.33 | dc_ | florian: we shouldn't need to then |
| 15:55.59 | dc_ | pb_: am I right? |
| 15:56.32 | pb_ | dc_: so it seems, but florian is the expert on these issues. |
| 15:56.46 | dc_ | florian: ;] |
| 15:57.38 | florian | ehh... what does this have to do with gtkcolorsel? |
| 15:58.09 | dc_ | florian: arn't the widgets from there? |
| 15:58.47 | florian | dc_: nope, taht's it - * Released under GNU GPL, read the file 'COPYING' for more information |
| 15:59.07 | florian | ... from sp-color-slider.c |
| 15:59.07 | dc_ | wtf. |
| 15:59.09 | dc_ | oh |
| 15:59.10 | dc_ | heh |
| 15:59.37 | dc_ | florian: where did that come from? |
| 16:00.02 | florian | dc_: from sodipodi |
| 16:00.06 | dc_ | doh! |
| 16:00.17 | dc_ | florian: thennnn.... |
| 16:00.27 | dc_ | florian: who don't we steal the gtk ones instead ;) |
| 16:00.27 | dc_ | hehe |
| 16:01.44 | florian | dc_: we have them... but just create an app with a GtkColorsel and try to use it on your ipaq... |
| 16:02.14 | dc_ | I know...but the sliders from gtkcolorsel |
| 16:02.18 | florian | i'll drop him a short mail |
| 16:02.38 | dc_ | hehehe |
| 16:02.47 | dc_ | florian: nevermind all this then :) |
| 16:03.14 | dc_ | florian: anyhow, once we can put the silder color selector into libgpewidget. |
| 16:03.40 | dc_ | florian: I intend to put a button at the bottom of the color palette dropdown menu. |
| 16:03.48 | dc_ | florian: like 'Custom...' |
| 16:03.52 | florian | dc_: btw.. i just updated GPE.opml in cvs... i think there is some work for you ;-) |
| 16:04.09 | dc_ | florian: then put that and the font dropdown in libgpewidget for gpe-word and gpe-conf theme |
| 16:04.14 | dc_ | florian: :D |
| 16:05.03 | florian | dc_: this sounds good, tell me if it looks so :-) |
| 16:06.14 | florian | this ipkg makes me crazy! |
| 16:06.35 | dc_ | ! |
| 16:07.16 | florian | it doesn't try to install a local package... |
| 16:08.45 | florian | oh, even worse.... it installed and then downloaded an older version and tried to install :-/ |
| 16:11.00 | pb_ | doh |
| 16:12.58 | florian | strange... |
| 16:15.26 | florian | mallum: what's the current cvs status of matchbox? is the font changing code in? |
| 16:20.17 | mallum | florian: no, not as yet |
| 16:20.51 | mallum | florian: for now can we just set the font color ?, rather than change the actual font itself ? |
| 16:21.41 | *** join/#gpe mccarthy (~mccarthy@granular.che.pitt.edu) |
| 16:21.46 | mccarthy | hi all |
| 16:21.56 | spung | yo mccarthy |
| 16:22.02 | spung | mccarthy: got my e-mail? |
| 16:22.15 | mccarthy | pb_: I dont remember my bugzilla email, but I will be looking at that bug today |
| 16:22.36 | mccarthy | oops. bugzilla password (not email) |
| 16:22.48 | mccarthy | spung is using the jedi mind trick on me |
| 16:22.58 | spung | lol |
| 16:23.33 | dc_ | spung: ;) |
| 16:23.41 | mccarthy | spung: I can never remember peoples *real* names, so I have no clue if I read your email ;) |
| 16:23.54 | spung | mccarthy: Luis Oliveira :) |
| 16:24.04 | mccarthy | spung: ahh. then yes. |
| 16:24.16 | spung | mccarthy: any news on that? |
| 16:24.17 | mccarthy | I am not convinced you are correct though. |
| 16:24.28 | spung | mccarthy: correct about the fix or the bug? |
| 16:24.45 | pb_ | hi mccarthy |
| 16:24.49 | mccarthy | spung: about the fix (I am not sure about whether there is a bug or not either, I guess) |
| 16:24.56 | mccarthy | spung: how does this bug manifest? |
| 16:25.08 | pb_ | mccarthy: there is an option on the bugzilla login screen to email your password to you. |
| 16:25.09 | florian | mallum: you told me some keys for both font and tolor for future implementation... gpe-conf now does both :-) |
| 16:25.10 | mccarthy | pb_: spring break, yeah! |
| 16:25.10 | spung | mccarthy: efence complaings |
| 16:25.15 | pb_ | mccarthy: aha, cool |
| 16:25.28 | spung | mccarthy: and IIRC it segfaults without it |
| 16:25.33 | mccarthy | spung: ahhh. ok, I will look at it that way... |
| 16:25.43 | mallum | florian: yeah ok I'll add it |
| 16:26.02 | mallum | florian: am really tired and stressed today so probably do it tommorow |
| 16:26.36 | spung | mallum: you need sun and kittens on your lap :D |
| 16:26.38 | florian | mallum: great... at the moment font color changing makes matchbox die soon |
| 16:26.56 | mccarthy | spung: if your fix actually fixes *something* then there IS a bug ;) since I dont think that it *should* have helped with any event_db_stop() issues |
| 16:27.22 | florian | mallum: yes ok, just tell me if it works - i want to know if my code works :-) |
| 16:27.29 | spung | mccarthy: well, efence stops complaining with my fix... why do you say it doesn't help? |
| 16:27.40 | mccarthy | pb_: have you looked at the event-ui "double click" bug? |
| 16:27.46 | pb_ | mccarthy: nope, not yet |
| 16:27.55 | mccarthy | pb_: ok. I will look there today too. |
| 16:27.55 | pb_ | mccarthy: do you have a url for that? |
| 16:28.07 | *** join/#gpe jg (~jg@dhcp-r17.crl.dec.com) |
| 16:28.17 | spung | jg: it's all your fault, anyway |
| 16:28.27 | pb_ | mccarthy: I started rewriting the day-view a couple of weeks back, but haven't done much on gpe-calendar since then. |
| 16:28.29 | mccarthy | pb_: I am not sure that anyone actually posted it to bugzilla I just remember an email (maybe florian?) that talked of it (and I see it too) |
| 16:28.36 | pb_ | mccarthy: ah, ok |
| 16:28.56 | spung | pb_: are you getting rid of the GtkList ? |
| 16:29.01 | jg | spung: of course.... Almost anything on your screen is my fault... |
| 16:29.02 | mccarthy | pb_: I will have to look at what you did to the day-view |
| 16:29.03 | pb_ | mccarthy: bugs that aren't in bugzilla tend to disappear from my radar pretty rapidly. :-/ |
| 16:29.15 | spung | jg: :) |
| 16:29.17 | pb_ | mccarthy: I haven't checked my changes in yet; the new code isn't really functional at the moment. |
| 16:29.21 | pb_ | spung: GtkCList, yes |
| 16:29.27 | mccarthy | pb_: would it be hard to make the week view "clickable"? |
| 16:29.33 | pb_ | mccarthy: probably not |
| 16:29.39 | spung | pb_: are you thinking about a drawing area or something? |
| 16:29.42 | pb_ | spung: yes |
| 16:29.47 | spung | pb_: cool! |
| 16:29.47 | pb_ | spung: same as the week and month views |
| 16:30.20 | cmarqu_werk | BBL |
| 16:30.26 | *** part/#gpe cmarqu_werk (www@vws1.kens.com) |
| 16:30.32 | spung | mccarthy: why do you say my fix doesn't help the event_db_stop() issue? |
| 16:30.35 | mccarthy | pb_: I am confused about the conv. you got going with spung? whats up with Clists? and how does it related to wekk/month views? |
| 16:30.55 | pb_ | mccarthy: the current day-view uses a GtkCList. I am rewriting it to use a drawingarea, like the week and month views do. |
| 16:31.09 | mccarthy | spung: because there shouldn't be any "clones" in the one_shot_events glist |
| 16:31.38 | mccarthy | pb_: so you are finally erasing all of my original contributions :) |
| 16:31.58 | pb_ | mccarthy: ah :-) |
| 16:32.03 | mccarthy | pb_: once you rewrite the recurrence stuff then I will officially be obsolete ;) |
| 16:35.19 | spung | mccarthy: hmmm.. right.. doesn't make sense then |
| 16:36.56 | mccarthy | where is this event_db__free_event routine? pb? |
| 16:37.21 | spung | mccarthy: event-db.h |
| 16:37.29 | spung | it's a macro |
| 16:42.22 | jg | mallum: could you build/package/release a server with the new resource extension (XRes, I beleive it is called), and its client? I'm still getting set up to build myself, but due to RMS pain, I've lost yet more time. If you can't, I'll try to on Wednesday, but the font stuff continues to soak my time more than expected.... |
| 16:43.58 | jg | mallum, pb_: dillo it appears is leaking X resources madly: we need to be able to check other clients as well. |
| 16:44.26 | jg | mallum, pb_: not nice when your X server gets to 17 megabytes.... Keeps oom really busy :-(. |
| 16:44.40 | pb_ | jg: right. :-/ |
| 16:44.45 | mallum | jg: sure I plan to do that |
| 16:44.52 | dc_ | thank god I don't use dillo ;D |
| 16:44.57 | pb_ | jg: the current dillo in familiar is rather old (0.6.6; current is 0.7) so it may already be fixed upstream. |
| 16:44.59 | mallum | jg: I have been looking and the X stuff today |
| 16:45.06 | jg | mallum: great. I'll let you know as soon as I can build again; today is really busy..... |
| 16:45.15 | mallum | jg: yeah probably wont be today |
| 16:45.16 | pb_ | jg: however, regrettably there are no gtk2 patches for dillo 0.7 yet. |
| 16:45.19 | jg | dealing with RMS has cost yet more time.... |
| 16:45.27 | jg | pb_: that's too bad. |
| 16:45.36 | jg | pb_: are they planning a port soon? |
| 16:45.36 | florian | bbl |
| 16:45.40 | mallum | jg: I want to package it up for the OZ guys as well and would like to use more or less the same code |
| 16:45.53 | mallum | jg: and switch to tslib |
| 16:46.03 | pb_ | jg: I'm not sure. I've asked a few times, and received non-committal responses. |
| 16:46.30 | pb_ | jg: the original stated intention was to port to gtk2 when "the major distributions start shipping it". I think that point has now arrived, so in theory this would be the time. |
| 16:46.44 | jg | pb_: yup. |
| 16:46.50 | jg | 2.2 is even out.... |
| 16:46.58 | pb_ | quite. |
| 16:47.20 | jg | and there are at least screen shots for the remaining major gtk1 apps up (e.g. evolution). |
| 16:48.05 | jg | not to mention mozilla.... |
| 16:48.14 | pb_ | It wouldn't be all that hard to do a gtk2 port for dillo 0.7 based on the old gtk2 patches for dillo 0.6.6; just the old problem of not enough hours in the day. |
| 16:49.13 | jg | pb_, mccarthy: yup. I did get a query from blizzard asking which ipaq to buy, cause someone in netscape/aol wants to do a port of gecko and hack... |
| 16:49.43 | pb_ | heh. |
| 16:49.45 | mallum | jg: yes , that would be cool |
| 16:49.58 | mallum | jg: I wonder what ever happened to that pure xlib port |
| 16:51.40 | mccarthy | ls |
| 16:51.43 | mccarthy | oops |
| 16:51.47 | dc_ | :} |
| 16:51.49 | PaxAnima | mccarthy: ;) |
| 16:52.04 | mccarthy | c'mon guys cant you tell me whats in this dir? |
| 16:52.13 | PaxAnima | . |
| 16:52.14 | PaxAnima | .. |
| 16:52.29 | dc_ | .rice |
| 16:52.49 | mccarthy | spung: do you recall exactly what you did to get efence to complain and/or to get things to crash? |
| 16:53.16 | spung | ahh |
| 16:53.20 | spung | duh |
| 16:53.48 | spung | mccarthy: i hadn't thought about this for a long time |
| 16:54.03 | spung | mccarthy: the existence of clones is irrelevant |
| 16:54.05 | *** topic/#gpe by pb_ -> GPE Palmtop Environment | http://gpe.handhelds.org | http://www.handhelds.org/z/wiki/CurrentThingsToDo | http://europa.debian.org/bugzilla | bootgpe2 images now at http://familiar.handhelds.org/releases/v0.7/boot/ | Familiar 0.7 issues: http://www.handhelds.org/bugzilla/showdependencytree.cgi?id=99 | Please every body hammer matchbox, reporting bugs to bugzilla | http://handhelds.org/z/wiki/GPEDevWeekend |
| 16:54.10 | spung | event_db_list_destroy() is trying to access the event_t's after they've been |
| 16:54.12 | spung | event_db_destroy()ed. |
| 16:54.17 | spung | mccarthy: it's as simple as that |
| 16:54.33 | spung | if (ev->flags & FLAG_CLONE) |
| 16:54.42 | spung | ev has been freed |
| 16:54.49 | spung | this is what causes efence to complain |
| 16:56.17 | mccarthy | spung: but I am deleting like crazy and efence is perfectly happy :( |
| 16:56.37 | spung | mccarthy: what's your point? |
| 16:56.49 | mccarthy | spung: I cannot reproduce your bug |
| 16:57.09 | spung | mccarthy: try with gpe-today, IIRC gpe-calendar only calls event_db_stop() on exit |
| 16:59.04 | spung | mccarthy: or does it? |
| 17:00.14 | spung | luis@nhop:~/Projects/gpe/base/gpe-calendar$ cat *.c | grep event_db_stop |
| 17:00.15 | spung | luis@nhop:~/Projects/gpe/base/gpe-calendar$ |
| 17:00.23 | spung | actually, it doesn't call it at all |
| 17:04.30 | mccarthy | spung: gpe-today doent seem to do anything here :( |
| 17:04.43 | mccarthy | I get a blank window (is it hung?) |
| 17:04.48 | spung | hmmm |
| 17:04.53 | spung | mccarthy: latest cvs? |
| 17:04.58 | mccarthy | but efence isnt complaining yet ;) |
| 17:05.07 | mccarthy | spung just 2 minutes ago! |
| 17:05.11 | spung | mccarthy: damn |
| 17:05.15 | dc_ | hehe |
| 17:06.29 | spung | mccarthy: it works here :\ |
| 17:07.12 | spung | mccarthy: gpe-today is quite slow with efence |
| 17:07.56 | mccarthy | spung: I am restarting it now.....12:06.... |
| 17:08.41 | mccarthy | BTW, I added a event_db_stop() to the exit portion of gpe-calendar and it still doesnt barf... |
| 17:08.51 | mccarthy | ack. wait.....I might know the answer.... |
| 17:09.37 | mccarthy | are you saying you have a problem updating or something? (i.e., after calling event_db_stop at least once?) |
| 17:10.02 | spung | mccarthy: yeah, i think that was the problem.. |
| 17:11.09 | mccarthy | spung: hmmm. well the clones never get put on any permanent list, so it makes sense that you are having issues if you try to access them after essentially "restarting" the event_db routines |
| 17:11.24 | paul | mccarthy: http://handhelds.org/bugzilla/show_bug.cgi?id=601 |
| 17:11.28 | spung | mccarthy: hum? |
| 17:11.40 | paul | mccarthy: you vote for my bug, i'll vote for your bug. (do a pincer movement on pb_ :) ) |
| 17:11.55 | mccarthy | spung: but I guess I might have a memory leak though, if I dont clean out the clones at the end.... |
| 17:12.10 | mccarthy | spung: gpe-today...still wating 12:10 |
| 17:12.19 | spung | mccarthy: event_db_stop() calls event_db_destroy(iter->data); then it calls list_destroy() that tries to access this free'd event |
| 17:12.25 | spung | mccarthy: cleary this is a bug |
| 17:12.32 | spung | s/cleary/clearly |
| 17:13.37 | mccarthy | paul: why do we need to tap-and-hold instead of simply tap? |
| 17:14.04 | paul | mccarthy: when i wrote that i thought perhaps there was a reason for it not being just a tap. |
| 17:14.13 | paul | mccarthy: but tap would do fine. |
| 17:14.19 | paul | or i assumed there was a reason |
| 17:15.22 | mccarthy | spung: I might have to look at gpe-today to know if this is a bug in libeventdb or a bug in my *wonderful* explaination of how to *use* libeventdb (which there was none, of course :) |
| 17:15.50 | mccarthy | spung: In any case, I am thinking of changing the way that libeventdb works anyhow.... |
| 17:16.19 | spung | mccarthy: but.. event_db_stop() is reading free'd memory... that's a bug! |
| 17:16.53 | mccarthy | at 12:14:30 I get ElectricFence Exiting: mmap() failed: Cannot allocate memory and gpe-today crashes |
| 17:17.40 | spung | mccarthy: can you get a bt? |
| 17:18.16 | paul | double free() is most definitely a bug. (nasty one too) |
| 17:18.26 | mccarthy | spung I didnt run it in gdb, but I will now... |
| 17:19.44 | spung | mccarthy: cool.. but, you do see that event_db_stop() is causing event_db_list_destroy() to read free'd memory! |
| 17:20.04 | mccarthy | spung: I havent *seen* any such thing, but yes, I have heard you say that |
| 17:20.36 | spung | mccarthy: hmm.. but, by reading the code.. you see that happens! don't you? |
| 17:20.40 | mccarthy | like I said adding an event_db_stop to gpe-calendar does not reproduce the bug no matter how I try |
| 17:20.58 | mccarthy | spung: actually, no I dont see it in the code |
| 17:21.23 | mccarthy | spung: event_db_stop erases the once_shot events, no CLONES are ever in that list |
| 17:21.30 | paul | (nasty esp cause it might only bite rarely) |
| 17:21.32 | spung | mccarthy: event_db_stop() destroys the events right? then calls event_db_list_destroy() which tries to read the list data which has been destroyed |
| 17:21.39 | mccarthy | the CLONES are the only ones erased previously |
| 17:21.59 | spung | mccarthy: yes.. but to check if there are CLONES, list_destroy tries to read the data |
| 17:22.30 | mccarthy | spung: AHHH. NOW I see what you are saying.....duh |
| 17:22.35 | mccarthy | ok, you are correct |
| 17:22.35 | spung | mccarthy: :) |
| 17:22.44 | spung | cool |
| 17:22.57 | mccarthy | I thought you were saying that the clones were erased and then erased AGAIN which is not true |
| 17:23.07 | spung | mccarthy: indeed |
| 17:23.23 | mccarthy | ok. let me think if your solution is the best way to fix it.... |
| 17:25.30 | mccarthy | spung: I think we still want to "destroy" the list, but we should distinguish between destroying a "temporary" list (which might contain clones) and a "permanent" list (which does not) |
| 17:26.04 | spung | mccarthy: yeah, guess we should have separate destroy_list functions then |
| 17:26.10 | mccarthy | so I think I will add a destroy_permanent_list and make event_db_stop call that (since I think that is the only place it is ever used) |
| 17:26.40 | mccarthy | (while the temporary lists are destroyed al over the place...at least in gpe-calendar) |
| 17:26.42 | spung | mccarthy: maybe make a destroy_temp_list() would look better? |
| 17:27.16 | spung | mccarthy: but that's between you and pb_ :) as long as it works :) |
| 17:27.21 | mccarthy | spung: but then I need to change gpe-calendar (and maybe gpe-today ;), whereas a permanent only requires a change to libeventdb |
| 17:27.37 | mccarthy | pb_: is that ok with you? |
| 17:27.47 | pb_ | mccarthy: sounds ok |
| 17:28.15 | spung | luis@nhop:~/Projects/gpe/base/gpe-today$ cat *.c | grep list_destroy | wc -l |
| 17:28.16 | spung | <PROTECTED> |
| 17:28.43 | spung | mccarthy: yeah, just a small change.. |
| 17:29.33 | mccarthy | spung: nope. actually, you were right the first time... all the "permanent" destroy will do now is g_slist_free anyhow.... |
| 17:29.45 | mccarthy | so I guess I will take your patch ;) |
| 17:29.49 | spung | mccarthy: cool |
| 17:30.46 | spung | mccarthy: can the return list of a event_db_list_for_period(); have clones? |
| 17:31.28 | mccarthy | spung: yep |
| 17:31.42 | mccarthy | spung: so apps should always use the "destroy" |
| 17:31.51 | spung | mccarthy: right, makes sense then |
| 17:31.53 | mccarthy | rather than a simple free |
| 17:32.04 | mccarthy | (which is what stop will now do) |
| 17:33.01 | mccarthy | spung: I cannot get a backtrace on gpe-today...it simple says exited with code 7 (whatever that means) |
| 17:33.38 | spung | weird |
| 17:34.04 | mccarthy | I will let you know if I get anything different with the fixed libeventdb |
| 17:34.35 | spung | mccarthy: it doesn't sound like a libeventdb thing |
| 17:36.59 | mccarthy | spung: you are correct, same thing |
| 17:38.05 | spung | mccarthy: sounds like missing error detection somwhere :\ |
| 17:43.19 | spung | mccarthy: well, if you find anything about that please bugzilla it |
| 17:43.31 | mccarthy | spung: will do |
| 18:02.24 | mccarthy | pb_: the bug I mentioned earlier is easy. a widget_show_all (rather than a show) fixed it |
| 18:02.32 | pb_ | mccarthy: ah right |
| 18:02.39 | mccarthy | pb_: there is another small bug I wann fix too... |
| 18:03.03 | mccarthy | pb_: clicking on a month-view day does *not* open an event with that date! |
| 18:03.31 | pb_ | mccarthy: hrm |
| 18:04.06 | pb_ | mccarthy: where is this widget_show that needs to be changed to show_all? |
| 18:04.42 | mccarthy | pb_: in dayview |
| 18:05.13 | pb_ | mccarthy: ok. be careful though, the GtkDateSel doesn't play well with gtk_widget_show_all. |
| 18:05.33 | mccarthy | pb_: hmm |
| 18:05.44 | pb_ | it may not be a problem, just something to watch out for. |
| 18:05.50 | mccarthy | pb_: but the place that I put it was one that *had* one already |
| 18:05.56 | pb_ | ok |
| 18:06.22 | mccarthy | pb_: in selection_made, it would show_all if the selection had an ev associated with it, but *not* if it was building one from scratch |
| 18:06.38 | pb_ | oh right, got it. |
| 18:06.45 | pb_ | yeah, that should be safe enough to change. |
| 18:07.03 | mccarthy | pb_: this is a weird bug though since it wouldn't show up if you had already clicked an existing event (since it had gotten a show_all that last time, and you cached the window) |
| 18:07.13 | pb_ | ah, heh. |
| 18:07.31 | mccarthy | all better now though....I am gonna track down the date thing and then quit hacking for today |
| 18:08.00 | pb_ | righto. |
| 18:08.02 | mccarthy | (I made one more small eventdb change on the recurring stuff that should make it slightly faster) |
| 18:08.13 | mccarthy | (it is still too damn slow though) |
| 18:08.32 | pb_ | mccarthy: btw, don't suppose you were planning to attend the handhelds.org mini-conference at CRL in April? |
| 18:08.48 | mccarthy | pb_: nope |
| 18:08.51 | pb_ | shame |
| 18:09.27 | mccarthy | it is, but I travel too much for my "real" work. (need to hang with the kiddies now and then) |
| 18:09.35 | pb_ | yeah, that's fair. |
| 18:10.09 | mccarthy | I will certainly stop and drink beer next time in the UK though (I love UK beer) |
| 18:10.25 | mccarthy | (and maybe with jamey if I am ever near CRL ;) |
| 18:10.36 | mccarthy | (as long as we can find a bar with good brit beer) |
| 18:10.36 | pb_ | Yeah, that'd be cool. |
| 18:11.39 | pb_ | right, going home; bbl |
| 18:14.30 | *** join/#gpe ade|indy (~adrian@unimatrix.ex.ac.uk) |
| 18:34.14 | *** join/#gpe pb_ (~pb@pc2-cmbg4-3-cust239.cmbg.cable.ntl.com) |
| 18:35.00 | mccarthy | pb_: I still cannot build new gpe-calendar packages :( |
| 18:35.15 | mccarthy | I have asked to have libgtk 2.2 put on the ipaqs.... |
| 18:36.08 | pb_ | mccarthy: ah. :-/ |
| 18:36.44 | pb_ | mccarthy: I can crank out a new package if you want. |
| 18:37.09 | pb_ | or, I guess I can make you an account on my new arm box and you can build it there. |
| 18:37.36 | mccarthy | pb_: either way. I wanted to test the package on my ipaq before uploading it.... |
| 18:37.48 | pb_ | mccarthy: ok, I'll make an ipk and send it to you for testing. |
| 18:38.00 | mccarthy | (I need both a new gpe-calendar and a new libeventdb) |
| 18:38.08 | pb_ | right |
| 18:38.38 | pb_ | is everything checked in and ready to go? |
| 18:40.45 | mccarthy | pb_: yeah, but you know what....if you do email it to me you need to send it directly to mccarthy@granular.che.pitt.edu rather than any of my other emails since the University (in its infinite wisdom) has chosen to knock down our email servers for another 4 hours! |
| 18:40.55 | pb_ | doh :-) |
| 18:41.18 | mccarthy | what on earth could take them 8 full hours? are they building a new machine from scratch?! |
| 18:41.59 | pb_ | mccarthy: looks like the version numbers have not been bumped. could you do that? |
| 18:42.27 | mccarthy | pb_: sure. do we do that in a Makefile now? (that discussion confused me :) |
| 18:42.31 | pb_ | yeah |
| 18:42.43 | pb_ | just change where it says "VERSION = ..." |
| 18:42.55 | pb_ | and if you want to do "make tag" too, that would be even better. |
| 18:46.18 | mccarthy | pb_: do I need to do a commit after the tag? |
| 18:46.36 | pb_ | mccarthy: nope, tag works on the repository anyway |
| 18:46.44 | mccarthy | pb_: ok. all done |
| 18:46.45 | *** join/#gpe florian (~fuchs@p50928C22.dip.t-dialin.net) |
| 18:46.46 | pb_ | cool |
| 18:46.48 | pb_ | hi florian |
| 18:46.55 | florian | pb_: hi |
| 18:46.58 | florian | hi all! |
| 18:53.05 | florian | did someone try to capture wireless lan frames using an orinoco driver? |
| 19:16.52 | cmarqu | Ah, nils called. |
| 19:19.16 | *** join/#gpe dan_t (~dan_t@212-41-206-101.adsl.galactica.it) |
| 19:19.24 | mccarthy | pb_: if/when you zap something to me let me know, since they may have messed with the routers as well.... |
| 19:19.38 | mccarthy | df -k |
| 19:19.43 | mccarthy | damn |
| 19:19.47 | mccarthy | I gotta stop that |
| 19:24.30 | florian | cmarqu: great - sounds like they finished work for today ;-) |
| 19:28.11 | cmarqu | :) |
| 19:28.15 | cmarqu | Booking tomorrow... |
| 19:28.33 | pb_ | Cool. |
| 19:29.56 | florian | great |
| 19:30.44 | florian | noone wants to write a frontend for ipkg ;-( |
| 19:48.14 | *** join/#gpe dc_ (~dc@modem-60-63-60-62.vip.uk.com) |
| 19:48.22 | dc_ | we |
| 19:48.24 | dc_ | :] |
| 19:53.57 | pb_ | florian: :-( |
| 19:56.40 | florian | today is an ugly day ;-/ |
| 19:56.56 | dc_ | :[ |
| 19:57.19 | florian | thes GtkCombos behave ugly too :-/ |
| 19:58.02 | dc_ | that's why we have my new popup menus! |
| 19:58.03 | dc_ | ;] |
| 19:58.29 | pb_ | quite right. |
| 19:58.33 | florian | http://handhelds.org/scap/port.18049.png |
| 19:58.51 | dc_ | :{ |
| 19:58.57 | florian | yes, but this doesn't help much porting gpa... |
| 20:00.06 | pb_ | florian: seems like a theme issue. |
| 20:00.14 | pb_ | florian: do you have the latest Angelistic? |
| 20:01.29 | florian | pb_: don't know - is your latest image with recent unstable stuff and todays cvs matchbox |
| 20:01.52 | pb_ | florian: do "xst read", see what Net/ThemeName is set to. |
| 20:02.51 | florian | pb_: this i know without xst :-) it's "blondie" |
| 20:02.58 | pb_ | florian: ah, then you will lose. |
| 20:03.12 | pb_ | florian: your theme need to include something like |
| 20:03.18 | pb_ | style "entry" |
| 20:03.20 | pb_ | { |
| 20:03.22 | pb_ | <PROTECTED> |
| 20:03.24 | pb_ | } |
| 20:03.26 | pb_ | otherwise your GtkEntries will be really wide. |
| 20:03.47 | pb_ | oh, er, and |
| 20:03.50 | pb_ | class "GtkEntry" style "entry" |
| 20:05.22 | florian | ok, we should add this in cvs... or not? |
| 20:05.31 | pb_ | it's in cvs already. |
| 20:05.54 | pb_ | out of the box, you should get the Angelistic theme which includes the right stuff. |
| 20:06.22 | pb_ | I'm not sure what more we can do, really. |
| 20:07.23 | florian | setting some ugly size requests... :-/ |
| 20:07.45 | dc_ | :\\ |
| 20:08.06 | dc_ | florian: okay, I'm getting my 1Mb+ GPE.opml file ;D |
| 20:08.13 | dc_ | lets of work. |
| 20:08.16 | dc_ | tehh |
| 20:08.18 | pb_ | heh |
| 20:08.44 | dc_ | mmmmm |
| 20:08.53 | cmarqu | The theme is the thing that is read before the user's gtkrc file? |
| 20:09.04 | pb_ | cmarqu: yeah, I think so. |
| 20:09.44 | pb_ | I'm not quite certain what the interaction is between /etc/gtkrc. ~/.gtkrc and the theme. |
| 20:16.05 | *** join/#gpe BBrox (~ulmer@bbrox.org) |
| 20:16.41 | *** join/#gpe BBrox (~ulmer@bbrox.org) |
| 20:30.10 | pb_ | dc_: not keeping you awake I hope |
| 20:30.21 | pb_ | mccarthy: new gpe-calendar built (at last) |
| 20:30.21 | dc_ | hehe |
| 20:30.30 | dc_ | I just don't feel to well :\ |
| 20:30.44 | dc_ | which in turn makes me bored. |
| 20:30.48 | dc_ | and so on |
| 20:30.53 | pb_ | ah. :-/ |
| 20:30.58 | dc_ | pb_: tell me to do something.. |
| 20:31.03 | pb_ | dc_: fix some bugs |
| 20:31.04 | dc_ | I can't decide / |
| 20:31.09 | dc_ | pb_: which one |
| 20:31.32 | pb_ | good question. let me mail this stuff to mccarthy, then I'll look |
| 20:31.37 | dc_ | :} |
| 20:32.27 | cmarqu | It would be good if there was a site-gtkrc that was read after the theme, but before the user's gtkrc. |
| 20:32.32 | mccarthy | dc_: you could fix my memory leak in gpe-mapviewer |
| 20:32.46 | pb_ | mccarthy: ah, what was your preferred email address again? |
| 20:32.50 | dc_ | mccarthy: I'm no good at fixing leeks |
| 20:33.08 | mccarthy | mccarthy@granular.che.pitt.edu |
| 20:33.10 | dc_ | mccarthy: especialy the edible kind |
| 20:33.15 | florian | dc_: what about the filemanager stuff?! |
| 20:33.17 | pb_ | mccarthy: mail sent |
| 20:34.13 | mccarthy | pb_: cool. I will let you know if it is good. |
| 20:34.40 | mccarthy | pb_: should I tell you whether to upload it? or should I just sign and upload if I like it? |
| 20:34.49 | pb_ | mccarthy: just sign and upload it if it does the trick |
| 20:35.23 | mccarthy | none of you guys is good at OpenGL are you? |
| 20:35.33 | mccarthy | none...are? |
| 20:36.59 | pb_ | dc_: don't suppose you want to work on xcalibrate? |
| 20:37.14 | dc_ | pb_: I dunno, what does that involve? |
| 20:37.23 | pb_ | dc_: well, basically making it less ugly. |
| 20:37.44 | pb_ | dc_: it's the first thing you see on a new install, and it looks pretty bad. |
| 20:37.56 | dc_ | true |
| 20:38.36 | pb_ | fr'example, the Opie calibrator is much more attractive. |
| 20:38.46 | dc_ | do we have a picture of it? |
| 20:39.02 | pb_ | dunno, let me look. |
| 20:39.55 | pb_ | hm, not that I can find right away. |
| 20:40.04 | dc_ | quite. |
| 20:40.06 | dc_ | either can I |
| 20:40.11 | pb_ | but, the main differences are, it has a background image rather than just plain black; |
| 20:40.12 | mccarthy | last I looked there was one on the trolltech site.... |
| 20:40.25 | pb_ | it doesn't have an X cursor getting in the way; |
| 20:40.38 | pb_ | and the crosshairs slide from place to place rather than jumping. |
| 20:40.49 | dc_ | ahh |
| 20:41.02 | dc_ | I guess that can be a start |
| 20:41.07 | dc_ | and a nice background maybe |
| 20:41.10 | pb_ | yeah |
| 20:42.03 | mccarthy | trolltech redid their site...cant find it |
| 20:42.12 | dc_ | heh |
| 20:42.51 | dc_ | is Qtopia'a tree closed source? |
| 20:43.01 | pb_ | the Opie tree is available. |
| 20:43.13 | pb_ | I think Qtopia proper is too, though I'm not quite sure. |
| 20:44.10 | dc_ | " I think Qt is the biggest breakthrough in software technology since the C programming language! |
| 20:44.16 | dc_ | Nigel Orchard |
| 20:44.16 | dc_ | Pilot Systems |
| 20:44.16 | dc_ | " |
| 20:44.16 | dc_ | doh |
| 20:44.42 | pb_ | doh indeed. |
| 20:46.55 | mccarthy | pb_: I think somehow the University has even managed to munge *my* ability to get email directly to my machine...any chance you wanna stick it in hh.org~pb? |
| 20:47.18 | pb_ | mccarthy: done, look for mccarthy.tar |
| 20:47.24 | mccarthy | pb_: thanks. |
| 20:48.51 | dc_ | pb_: so where do I get my fun xcalibrate source from? |
| 20:48.53 | mccarthy | pb_: where does one get a knew Packages script that knows how to handle ar files? |
| 20:48.59 | pb_ | dc_: in handhelds.org cvs, apps/xcalibrate |
| 20:49.04 | dc_ | right |
| 20:49.06 | pb_ | mccarthy: ipkg cvs |
| 20:49.17 | pb_ | mccarthy: presumably you mean ipkg-{un}build, right? |
| 20:49.33 | pb_ | or ipkg-makeindex, or something |
| 20:49.38 | dc_ | I guess it's time to learn how do use xlib then? |
| 20:49.46 | pb_ | dc_: guess so! |
| 20:49.47 | mccarthy | pb_: yeah, I guess I mean an ipkg-makeindex |
| 20:50.00 | dc_ | hrm, is it safe to run xcalibrate on my desktop? |
| 20:50.19 | pb_ | dc_: don't think it'll work, it looks for /dev/h3600_ts or some such |
| 20:50.30 | dc_ | how borring |
| 20:50.34 | pb_ | :-/ |
| 20:50.39 | dc_ | can I just trick it? |
| 20:50.56 | pb_ | not easily |
| 20:50.59 | pb_ | at least, not that I can think of. |
| 20:51.06 | dc_ | :\ |
| 20:51.33 | pb_ | I guess there's a reason nobody else is rushing to work on xcalibrate. |
| 20:51.58 | dc_ | yes |
| 20:52.13 | dc_ | and I am even more diabled without a cross compiler |
| 20:52.20 | pb_ | ah yes, suck. |
| 20:52.25 | pb_ | maybe xcalibrate not such a good idea then. |
| 20:54.32 | pb_ | dc_: in which case, wanna have a go at writing a panel clock program that can do spiffy things like set alarms and change the timezone? |
| 20:55.10 | mccarthy | pb_: damn. I think I broke the alarms.... |
| 20:55.20 | pb_ | mccarthy: darn |
| 20:56.12 | mccarthy | pb_: but I *did* fix the display issues |
| 20:56.21 | mccarthy | pb_: but it really isnt any faster (that I notice) |
| 20:56.25 | pb_ | ah. |
| 20:56.55 | dc_ | actualy I have a better idea =P |
| 20:56.59 | dc_ | hehe |
| 20:56.59 | mccarthy | pb_: would you be *strongly* against making the "clone" list semi-permanent? (i.e., keeping it around for the duration of the session) |
| 20:57.03 | dc_ | nothing involving X |
| 20:57.06 | pb_ | dc_: aha, what's that? |
| 20:57.06 | dc_ | ! |
| 20:57.15 | pb_ | mccarthy: fairly strongly, yes |
| 20:57.21 | dc_ | pb_: make a clock app that has moving hands! |
| 20:57.27 | pb_ | mccarthy: but you are welcome to convince me. |
| 20:57.30 | pb_ | dc_: ah, good idea |
| 20:57.55 | dc_ | it's just what I've always wanted |
| 20:57.57 | mccarthy | pb_: do you notice gpe-calendar being *very* slow to change views? or days in day-view (more specifically) |
| 20:57.58 | pb_ | though, I guess we already have xclock, oclock, etc, I'm not sure the world needs too many more clocks. |
| 20:57.59 | dc_ | ;] |
| 20:58.21 | pb_ | mccarthy: it does seem a bit sluggish. |
| 20:58.27 | dc_ | do they dock? |
| 20:58.43 | pb_ | dc_: ah, no. |
| 20:58.44 | mccarthy | pb_: well, do you have any recurring events? I think they make it even *worse*! |
| 20:58.49 | dc_ | pb_: :) |
| 20:59.02 | pb_ | dc_: is there room in the dock for a clock with moving hands? |
| 20:59.16 | nelson | uh-oh. |
| 20:59.21 | dc_ | pb_: we'll make some |
| 20:59.34 | nelson | Matchbox isn't displaying its topbar when the desktop is running. |
| 20:59.35 | mccarthy | pb_: I think it is sluggish because it recreates the clones each time you change views (or days) |
| 20:59.47 | nelson | So when I switch back to mbdesktop from an application, I have no way to get back. |
| 20:59.49 | pb_ | dc_: or you could make a generic "swallow this other app" thing, which might be cute. |
| 20:59.51 | dc_ | nelson: that's what the Tasks folder if for |
| 21:00.10 | dc_ | pb_: a-whataa-swallow-whatta? |
| 21:00.20 | mccarthy | pb_: if we made the clones at the initialization (which I think we do anyway) and *saved* them, then this should be fast again |
| 21:00.25 | pb_ | dc_: a wrapper thing that would allow you to dock any application |
| 21:00.32 | nelson | dc_: sigh. That is a broken idea. |
| 21:00.43 | nelson | dc_: users are stupid. |
| 21:00.50 | dc_ | nelson: :) |
| 21:00.52 | nelson | dc_: you should only ever give them ONE way of doing something. |
| 21:00.54 | pb_ | mccarthy: saved them in the database, or just kept them around in RAM? |
| 21:01.12 | mccarthy | pb_: I wouldn't make them directly clones anymore though, but would instead have "events" which are as they currently are, but then also have "event_pointers" or some such that would include clones |
| 21:01.15 | nelson | the "tasks" folder must go. Who do I talk to about it? |
| 21:01.18 | mccarthy | pb_: just keep them in RAM |
| 21:01.18 | dc_ | pb_: but...then I will just complain about not being able to do the geine effect that OS-X does. |
| 21:01.23 | pb_ | nelson: mallum |
| 21:01.29 | dc_ | nelson: mallum |
| 21:01.32 | pb_ | dc_: "geine" effect? |
| 21:01.37 | nelson | nelson: mallum |
| 21:01.52 | nelson | Okay, mallum it is. |
| 21:01.55 | mccarthy | nelson: mallum ;) |
| 21:02.03 | pb_ | mccarthy: hrm |
| 21:02.23 | dc_ | pb_: doh, I lack spelling skills |
| 21:02.32 | pb_ | mccarthy: wanna send me an email explaining how that would work? |
| 21:02.46 | mccarthy | pb_: all we really need for display is the time string and the uid of the event data |
| 21:03.06 | pb_ | dc_: sorry, I couldn't (still can't) figure out what you meant at all. |
| 21:03.20 | pb_ | mccarthy: indeed |
| 21:03.21 | mccarthy | pb_: most of the code to work with that already exists since I made the find_by_uid or whatever routine |
| 21:03.36 | dc_ | pb_: hang... |
| 21:03.38 | mccarthy | pb_: so we can easily grab the other event info that we might need |
| 21:03.56 | mccarthy | yeah, I am getting email again! |
| 21:04.01 | dc_ | pb_: ah ha, genie. |
| 21:04.04 | pb_ | mccarthy: find_by_uid is pretty slow though. |
| 21:04.06 | dc_ | pb_: in a bottle... |
| 21:04.13 | pb_ | dc_: aha. |
| 21:04.26 | mccarthy | pb_: hmmm... I didnt know that...I guess it would be |
| 21:04.32 | dc_ | pb_: OS-X does this effect where is really 'sucks' the windows into the task bar |
| 21:04.39 | pb_ | mccarthy: it has to scan the entire event list. |
| 21:04.49 | dc_ | or rather genie is 'General Electric Network for Information Exchange' |
| 21:04.56 | dc_ | doh |
| 21:05.01 | pb_ | dc_: ah yes, we should indeed implement that |
| 21:05.01 | dc_ | why didn't we know that. |
| 21:05.21 | pb_ | certainly an oversight. |
| 21:05.30 | mccarthy | pb_: I guess there isnt a way to "index" a g_slist? so that we could get ev[uid]? |
| 21:05.34 | dc_ | pb_: :} |
| 21:05.37 | pb_ | mccarthy: not really. |
| 21:05.43 | dc_ | pb_: okay....moving one, any more ideas. |
| 21:05.55 | pb_ | mccarthy: the whole point of the event_t structure is that it is what you use to address an event. |
| 21:06.13 | mallum | nelson: if you dont like the task folder, you can use a dock app to get back to your last app |
| 21:06.17 | pb_ | mccarthy: gpe-calendar should not be operating on UIDs internally much if at all. |
| 21:06.34 | nelson | mallum: it's not that I don't like it. It's that nobody is going to like it. |
| 21:06.53 | pb_ | dc_: hrm |
| 21:06.54 | nelson | mallum: folders hold files, not programs. |
| 21:07.17 | mccarthy | mallum: what the heck is the task folder? |
| 21:07.31 | nelson | Besides which, a "Task" is something which isn't done. It's not something that's running. |
| 21:07.46 | nelson | mccarthy: that's where your running programs disappeared to. |
| 21:08.10 | mccarthy | pb_: ok. so maybe clones should really be exactly that (full clones). This would certainly make it faster, but might make it a big memory hog... |
| 21:08.17 | nelson | mccarthy: did you notice that matchbox's menu bar disappeared? |
| 21:08.19 | nelson | mallum: LOL! |
| 21:08.33 | nelson | mallum: you've got email, too. |
| 21:08.45 | pb_ | mccarthy: the thing is that a clone only really makes sense in the context of a time period. |
| 21:08.46 | mccarthy | nelson: I might be running an older matchbox (shhh, dont tell on me) |
| 21:09.04 | pb_ | mccarthy: if you have an event that recurs all the way off to infinity, you can't pre-compute all the possible clones even if you wanted to. |
| 21:09.16 | mccarthy | pb_: yes, I agree, but there should still be a way that we can make this dog run faster ;) |
| 21:09.39 | nelson | florian: don't give me that bullshit. |
| 21:09.45 | pb_ | mccarthy: have you tried profiling it to see where the time actually goes? |
| 21:09.50 | mccarthy | pb_: nope |
| 21:09.54 | nelson | florian: a bad idea is a bad idea, even if Windows manages to do the opposite from it. |
| 21:09.59 | pb_ | mccarthy: I think that has to be the first line of attack. |
| 21:10.08 | mccarthy | pb_: mostly because on my desktop it runs fast ;) |
| 21:10.18 | mccarthy | (and I cannot remember how to profile things) |
| 21:10.21 | pb_ | mccarthy: you can still profile it on the desktop, just all the numbers will be smaller. |
| 21:10.35 | mccarthy | pb_: hence the smiley wink |
| 21:10.49 | pb_ | aha |
| 21:10.53 | *** join/#gpe ade|indy (~adrian@unimatrix.ex.ac.uk) |
| 21:11.09 | nelson | mallum: are you serious? YOu're not going to fix this problem? |
| 21:11.18 | pb_ | dc_: I guess you looked at the CurrentThingsToDo page, right? |
| 21:12.29 | mccarthy | pb_: point me to a good reference on profiling? |
| 21:12.47 | dc_ | pb_: heh....my clock is already in action. |
| 21:13.02 | dc_ | pb_: I am sworn to finish that first ;] |
| 21:13.07 | mccarthy | dc_: too bad you couldn't fix my leak that fast ;) |
| 21:13.16 | mallum | nelson: please suggest a solution |
| 21:13.20 | dc_ | ! |
| 21:13.34 | pb_ | dc_: aha. |
| 21:13.43 | dc_ | I had leek in my soup for diner an hour ago. |
| 21:13.48 | dc_ | it wasn't very nice. |
| 21:13.57 | pb_ | ibot, search google for gprof |
| 21:14.04 | pb_ | mccarthy: check those links |
| 21:14.13 | mccarthy | pb_: ok. gprof helps... |
| 21:15.06 | florian | nelson: don't take so serious what i said - we need a cool solution... |
| 21:17.17 | florian | nelson: a bar has the disadvantage of using space, this is ok in a desktop invironment but a real matter on a pda |
| 21:18.26 | pb_ | florian: we already have bars on all other windows, and I don't think the desktop is particularly space-starved. |
| 21:18.46 | nelson | mallum: the solution is to go back to what you were doing before, where the application switcher is always there. |
| 21:19.27 | nelson | florian: I agree that using up screen space is bad. On the other hand, you can *always* double-click it to make it disappear. |
| 21:19.34 | mallum | nelson: not that easy Im afraid |
| 21:19.44 | nelson | mallum: why not? |
| 21:19.55 | nelson | isn't mbdesktop just another program? |
| 21:21.20 | mallum | nelson: yes exactly but its a special window |
| 21:21.32 | nelson | I just ran another copy of mbdesktop, and it acts like a program. |
| 21:21.49 | mallum | nelson: the easiest fix is to add a switch ( oh no ) to mbdesktop that makes it run just as a regular window |
| 21:21.51 | florian | pb_: in this way you're right - for the desktop this is no real problem |
| 21:22.14 | nelson | mallum: why not fix it so that it always does the right thing? |
| 21:22.43 | nelson | mallum: why do you need to have a way to force it to run broken? |
| 21:22.49 | mallum | nelson: but I like it how it is |
| 21:23.01 | florian | nelson: hmm, yes - i think that is ok for desktop usage |
| 21:23.09 | nelson | mallum: I can see that. Do you understand that it's going to drive users buggy? |
| 21:23.20 | pb_ | nelson: I guess we could adjust the session scripts to run mbdesktop in this alternate mode. |
| 21:23.54 | mallum | nelson: you can always run gpe-appmgr instead |
| 21:24.39 | nelson | mallum: Or .... I can convince you of how awful an idea this is. |
| 21:25.25 | nelson | mallum: my first thought, when I saw the task bar disappear was "Oh damn, matchbox crashed". |
| 21:25.35 | nelson | mallum: don't you think users are going to think the same thing? |
| 21:25.37 | mallum | nelson: but if I make it a normal window, its not really a desktop anymore |
| 21:26.37 | nelson | mallum: my task switcher doesn't disappear when I expose my desktop machine's desktop. |
| 21:26.52 | nelson | mallum: I have to exlicitly tell it to slide to one side, or tell it to hide itself. |
| 21:27.24 | mccarthy | nelson mallum: why not make the task-switcher a separate app entirely? |
| 21:27.24 | nelson | mallum: just like I can double-click on the task bar to make it slide up. |
| 21:28.03 | nelson | mccarthy: and then we can have a program which lets you switch from the program you're running now, to the task switcher! |
| 21:28.11 | mccarthy | pb_: I get weird results from this gprof thing (maybe I dont know how to use it well) |
| 21:28.23 | mallum | mccarthy: yes, I've been waiting ages for someone to do that ... |
| 21:28.59 | mccarthy | nelson: but you dont "switch" from a program to the dock!? Its simply there, or its not (when you tell it to go away) |
| 21:29.07 | mccarthy | mallum: why not use fspanel? |
| 21:29.31 | mccarthy | nelson: the dock behaves exactly like you seem to be suggesting that a task-switcher should... |
| 21:29.39 | mccarthy | and its a separate app |
| 21:29.45 | nelson | mccarthy: I was being facetious. |
| 21:30.06 | mallum | mccarthy: becuase fspanel is not designed for a 240 pixel wide display ... |
| 21:30.13 | dc_ | yay |
| 21:30.17 | dc_ | I have a ....line! |
| 21:30.21 | dc_ | my first clock hand |
| 21:30.37 | mccarthy | pb_: this thing claims that I called data_start 2636 times...do we *have* a data_start routine? |
| 21:32.54 | mallum | nelson: monolaunch -l some.png mbcontrol -desktop |
| 21:32.56 | nelson | mccarthy: the problem is that there are two ways of getting to running programs: either through the titlebar, or through the tasks folder. |
| 21:33.53 | mallum | nelson: or using a themes arrow keys |
| 21:34.02 | mallum | nelson: or using a hardware key |
| 21:34.23 | mallum | nelson: or using a dockapp with .desktop files |
| 21:34.34 | nelson | mallum: I can see the logic that got you to the current solution, but it's gonna drive newbies crazy. You've got to do a better job of leading them to develop the mental model you *want*. Otherwise they're doing to develop a different mental model which will lead them down the primrose path. |
| 21:34.34 | mallum | nelson: or using mbcotrol |
| 21:35.44 | dc_ | okay, enough of this |
| 21:35.49 | mallum | nelson: originally I had a little dockapp which would toggle the desktop view on and off ( with no desktop in the titlebar task menu ) maybe that would make more ssense |
| 21:35.52 | dc_ | pb_: give me a bug to fix |
| 21:36.01 | cmarqu | dc_: gpe-gallery! :) |
| 21:36.18 | dc_ | cmarqu: ! to you too |
| 21:36.34 | dc_ | okay I know |
| 21:36.41 | dc_ | cmarqu: just to make you happy |
| 21:36.47 | cmarqu | dc_: Thanks :) |
| 21:36.56 | dc_ | a list view! |
| 21:36.58 | mallum | nelson: except then the user will never find the damn dockapp |
| 21:36.59 | nelson | mallum: okay, how about this solution: don't give the users a list of programs. When they click on the downarrow, .... but no, that doesn't help either, because then their programs disappear into the task folder |
| 21:38.24 | cmarqu | Hmm, what would be a good module for the flyer source and pdf? gpe/marketing or /gpe/flyer? /marketing is more generic. |
| 21:38.50 | nelson | mallum: how about this: when they click on the downarrow, draw a set of boxes starting from the window border and travelling smaller and smaller onto the tasks folder. |
| 21:39.12 | mallum | nelson: ?? |
| 21:39.13 | nelson | mallum: and change the tasks folder image if there are running programs. |
| 21:39.26 | mallum | nelson: yes I plan to do that anyway |
| 21:39.33 | nelson | mallum: to give them a clue that their program has gone into the tasks folder. |
| 21:39.49 | dc_ | ahh nooo |
| 21:39.54 | dc_ | gpe-gallery went mad |
| 21:39.57 | mccarthy | pb_: i finally got your email |
| 21:39.57 | dc_ | god help us |
| 21:40.02 | mallum | nelson: yes, I was going to add a little 'emblem' to that folder |
| 21:40.48 | mccarthy | dc_: aha! I got it.... |
| 21:40.52 | mallum | nelson: and how about I change tasks to 'running programs' |
| 21:40.54 | dc_ | mccarthy: huray :D |
| 21:40.58 | mccarthy | dc_: you should work on porting xpdf to gpe! |
| 21:41.01 | nelson | mallum: actually, if you wanted to be *really* cool, take a screencap of their program, and use it as the png for their program. |
| 21:41.07 | dc_ | mccarthy: C++ |
| 21:41.11 | dc_ | mccarthy: end of story |
| 21:41.11 | mccarthy | dc_: damn |
| 21:41.32 | nelson | mallum: yes, that's also a good idea. |
| 21:41.42 | mccarthy | I think we need a pdf viewer for GPE 1.0 |
| 21:42.11 | dc_ | are we on gpe1 for gtk2? |
| 21:42.34 | mallum | nelson: I cant really make it a normal window cos that opens all other kinds of worms |
| 21:43.13 | mallum | nelson: I cant keep the titlebar menu there as its a seperate program and the code to draw that little drop down menu is infact pretty complicated |
| 21:43.30 | nelson | mallum: can't make what a normal window? |
| 21:43.42 | mallum | nelson: nope |
| 21:44.01 | mallum | nelson: if you want a normal window app launcher use gpe-applauncher |
| 21:44.10 | nelson | mallum: I'm suggesting that you get rid of the little drop down menu entirely. |
| 21:44.31 | mccarthy | gpe-applauncher?! |
| 21:44.32 | nelson | mallum: just have a downarrow pixmap. If they click it, the desktop comes to the top. |
| 21:44.55 | mallum | nelson: oh I see |
| 21:45.45 | mallum | nelson: well I could add a show desktop button to matchbox titlebars. Then you could just build a theme with that and no titlebar taskmenu |
| 21:45.47 | nelson | mallum: I'm sorry, I was confusing you. |
| 21:45.55 | mallum | nelson: no Im with you now |
| 21:46.04 | nelson | mallum: my desktop window manager has a downarrow button that brings up a menu. |
| 21:46.21 | nelson | Okay, so don't have a downarrow button. Just have a minimize button in the top right. |
| 21:46.36 | mallum | nelson: it doesn't have to be a down arrow, it can be what ever you want |
| 21:46.48 | mallum | nelson: the theming is very flexable |
| 21:49.22 | pb_ | mccarthy: data_start just indicates the start of the data area. |
| 21:49.23 | *** join/#gpe dc__ (~dc@modem-80-54-60-62.vip.uk.com) |
| 21:49.28 | pb_ | mccarthy: sounds like gprof is a little confused. |
| 21:49.56 | mccarthy | pb_: yep. It doesnt seem to check any libeventdb stuff either although I recompiled that too :( |
| 21:50.45 | pb_ | dc_: ah yes, a list view in gpe-gallery would be cool |
| 21:53.54 | florian | dc_: in gpe-filemgr too :-) |
| 21:54.11 | pb_ | yes. |
| 21:54.17 | dc__ | heh where? |
| 21:54.25 | dc__ | oh doh |
| 21:54.27 | dc__ | florian: right |
| 21:54.28 | dc__ | :] |
| 21:54.31 | dc__ | list view night tonight |
| 21:54.41 | cmarqu | Okay, we have a marketing/ module now. |
| 21:55.07 | dc__ | cmarqu: what does this mean then? |
| 21:55.44 | cmarqu | dc__: Well, you could play with the source of the GPE flyer if you wanted. And jsmanrique wanted to do a translation. |
| 21:56.01 | mallum | nelson: okey I've added support for desktop button in themes |
| 21:56.04 | cmarqu | Other stuff could be there too, e.g.. nils poster for CeBit. |
| 21:56.17 | dc__ | cmarqu: hehe |
| 21:56.34 | nelson | mallum: thanks! well done! |
| 21:56.56 | mallum | nelson: and tasks is now running programs |
| 21:59.27 | mallum | nelson: note: someone will have to make an image and add the desktop button definition to a theme |
| 21:59.30 | florian | Just received a mail from Lauris Kaplinski: |
| 21:59.32 | pb_ | dc__: did you lose interest in the clock then? |
| 21:59.45 | florian | "Statement: |
| 21:59.46 | florian | I allow the distribution of SPColorSlider code under GNU Lesser |
| 21:59.46 | florian | General Publick License." |
| 21:59.54 | pb_ | florian: Ah, great. |
| 22:00.12 | florian | very very nice... |
| 22:04.59 | cmarqu | florian: Make sure to check that mail into CVS. |
| 22:05.06 | *** join/#gpe walters (walters@verbum.org) |
| 22:05.54 | florian | cmarqu: ok, i'll do |
| 22:05.55 | nelson | mallum: and then set that theme to be the default. |
| 22:06.01 | dc__ | pb_: yeah |
| 22:06.04 | dc__ | pb_: to ugly |
| 22:06.28 | dc__ | ! |
| 22:07.11 | mallum | hey, how about a function call in the fb driver which paints the sharp logo ? |
| 22:07.48 | dc__ | oh yay |
| 22:08.12 | mallum | and lets just indent every function differently for the fun of it :-) |
| 22:08.21 | dc__ | hehehe |
| 22:17.38 | cmarqu | irc://freenode.net:<port>/<channel> -- would that be the official notation? |
| 22:20.09 | nelson | Shouldn't gpeirc have default entries that point here, and handhelds.org? |
| 22:20.59 | cmarqu | It has at least to freenode. |
| 22:21.12 | cmarqu | No channels though. |
| 22:21.25 | florian | good night all! |
| 22:21.29 | cmarqu | Night florian |
| 22:21.40 | pb_ | florian: night! |
| 22:28.17 | *** join/#gpe dc__ (~dc@modem-117-49-60-62.vip.uk.com) |
| 22:33.50 | nelson | cmarqu: yes, it should have channels defaulting as well. |
| 22:36.28 | cmarqu | nelson: ACK |
| 22:37.12 | nelson | I'm gonna nuke the other irc client out of unstable. |
| 22:37.29 | pb_ | nelson: olirc? sure, do it. |
| 22:37.45 | mallum | nelson: the folder emblem thing is gonna have to wait till after 0.5 is released |
| 22:38.40 | cmarqu | Oh, while we are at nuking stuff... the translation project has asked if I could rename gpe-ownerinfo2 to gpe-ownerinfo -- which exists in unstable already. |
| 22:38.59 | cmarqu | I would have to create a circular Replaces: loop now :/ |
| 22:39.20 | pb_ | cmarqu: ah, heh. |
| 22:39.24 | mallum | dc__: are you still gonna put togeather a wonderland theme ? |
| 22:39.31 | cmarqu | Unless the old gpe-ownerinfo is nuked -- or renamed to gpe-ownerinfo1 maybe. |
| 22:39.33 | dc__ | mallum: oh yes....tomorrow |
| 22:39.33 | pb_ | cmarqu: well, we can just quietly delete the "old" gpe-ownerinfo2 from unstable if you want. |
| 22:40.04 | pb_ | cmarqu: I don't think there is any need to keep the old gpe-ownerinfo around any longer, is there? |
| 22:40.07 | cmarqu | pb_: gpe-ownerinfo rather -- if I change stuff, I'd better do a total rename. |
| 22:40.12 | pb_ | cmarqu: ok |
| 22:40.15 | cmarqu | pb_: No, there is not. |
| 22:40.37 | pb_ | cmarqu: okay, let's do that then. |
| 22:40.39 | cmarqu | I'd even change the CVS module name, or is there anybody uncomfortable with that? |
| 22:40.46 | pb_ | yes, good idea. |
| 22:40.56 | mallum | dc__: you should drop the task menu from the titlebar ( just set its size to zero or dont have that frame ) and put a button in with action="desktop" instead ( to keep pb_, nelson happy ) |
| 22:40.58 | cmarqu | cvs mv then? |
| 22:41.14 | pb_ | You have to log into handhelds.org and move it directly in the repository. |
| 22:41.40 | cmarqu | pb_: Ah, okay, yes, I am allowed to do that, I forgot. |
| 22:41.49 | mallum | dc__: add the desktop button to the main frame |
| 22:41.51 | cmarqu | No such thing with projects at SF... |
| 22:42.07 | pb_ | Ha, indeed. |
| 22:42.15 | cmarqu | pb_: cvs rm the old gpe-ownerinfo foirst? |
| 22:42.49 | pb_ | cmarqu: I would rename that to something else (in the repository again) just in case we ever want to get the source back for some reason. |
| 22:43.10 | cmarqu | pb_: Well, it would still be in the Attic directory, right? |
| 22:43.11 | pb_ | cmarqu: what I tend to do in that situation is move the source directory out of the repository and into a "morgue" directory in my $HOME. |
| 22:43.26 | cmarqu | Hmm, okay. |
| 22:43.48 | pb_ | cmarqu: Or, yeah, you could just "cvs rm" all the files, but that would still leave the directories hanging around. |
| 22:44.03 | cmarqu | Hmm. |
| 22:45.10 | cmarqu | So... I'm renaming it to gpe-ownerinfo-gtk1 or gpe-ownerinfo1? |
| 22:45.21 | cmarqu | Or gpe-ownerinfo.old... |
| 22:45.30 | pb_ | gpe-ownerinfo1 would get my vote. |
| 22:46.25 | cmarqu | Okay. |
| 22:48.30 | cmarqu | Moved. |
| 22:48.39 | cmarqu | Let's see how that looks when I cvs up. |
| 22:48.41 | pb_ | Cool. |
| 22:49.42 | *** join/#gpe dan_t (~dan_t@212-41-206-101.adsl.galactica.it) |
| 22:51.20 | cmarqu | Argh. So, mv gpe-ownerinfo2/ gpe-ownerinfo/ was not a good idea. |
| 22:51.32 | cmarqu | cvs [server aborted]: could not find desired version 1.44 in /cvs/gpe/base/gpe-ownerinfo/ChangeLog,v |
| 22:51.51 | pb_ | Ah, you'll need to delete your old working directories. |
| 22:52.11 | cmarqu | So will anybody else? Bad... but okay. |
| 22:52.29 | pb_ | Yeah, I hadn't thought of that. It does seem bad, but probably unavoidable. |
| 22:54.01 | cmarqu | I'll send a message about that to the list at least. |
| 22:54.19 | pb_ | Okay. |
| 22:57.25 | cmarqu | Will CVS be confused if I move files around? like gpe-ownerinfo/gpe-ownerinfo2.c,v to gpe-ownerinfo/gpe-ownerinfo.c,v? |
| 22:57.42 | mallum | nelson: you do realise with the desktop button, its going to take 3 taps to just switch to another application ... |
| 23:50.07 | *** join/#gpe ibot (ibot@rikers.org) |
| 23:50.07 | *** topic/#gpe is GPE Palmtop Environment | http://gpe.handhelds.org | http://www.handhelds.org/z/wiki/CurrentThingsToDo | http://europa.debian.org/bugzilla | bootgpe2 images now at http://familiar.handhelds.org/releases/v0.7/boot/ | Familiar 0.7 issues: http://www.handhelds.org/bugzilla/showdependencytree.cgi?id=99 | Please every body hammer matchbox, reporting bugs to bugzilla | http://handhelds.org/z/wiki/GPEDevWeekend |
| 23:50.11 | pb_ | cmarqu: "Replaces" just tells ipkg that it should remove the replaced package when installing the replacing one. It won't proactively seek out packages that have been replaced, though; it needs to be told to install the replacing one by other means. |
| 23:50.26 | cmarqu | I see. |
| 23:50.26 | mallum | pb_: okey, phew :-) |
| 23:51.06 | pb_ | mallum: I think it's just the fact that matchbox fixes the size of "main" windows that stops the iconlist bug from appearing on the ipaq. |
| 23:53.34 | dan_t | i'm trying x2x, how i can make x show the cursor on the ipaq ? |
| 23:53.49 | dc___ | doh |
| 23:53.51 | dc___ | hehe |
| 23:55.45 | mallum | dan_t: xst write MATCHBOX/CURSOR str true |
| 23:57.32 | dan_t | mallum: it doesn't work |
| 23:59.24 | mallum | dan_t: try logging n / out and see it that fixes it |
