| *** dhbiker has quit IRC | 00:06 | |
| *** joshgillies has joined #maemo-ssu | 00:06 | |
| luf | freemangordon: http://repository.maemo.org/pool/maemo5.0/free/h/hildon-fm-l10n-public/ | 00:06 |
|---|---|---|
| luf | freemangordon: so we can rebuild all nokia "closed" hildon-fm-l10n-<locale> packages ... | 00:09 |
| *** dhbiker has joined #maemo-ssu | 00:12 | |
| luf | But I'm not sure if it's ok for layers ... | 00:22 |
| luf | I changed my cs_CZ hildon-fm.mo file and now I see Bluetooth instead of sfil_li_bluetooth | 00:23 |
| luf | :) | 00:23 |
| *** dafox has quit IRC | 00:24 | |
| *** nox- has joined #maemo-ssu | 00:33 | |
| *** Pali has quit IRC | 00:42 | |
| luf | I pushed changes into libhildonfm git. | 00:44 |
| *** Martix has quit IRC | 00:45 | |
| *** luf has quit IRC | 01:18 | |
| *** luf has joined #maemo-ssu | 01:18 | |
| *** arcean has quit IRC | 01:33 | |
| *** arcean has joined #maemo-ssu | 01:34 | |
| *** nox- has quit IRC | 02:20 | |
| *** int_ua has joined #maemo-ssu | 02:33 | |
| freemangordon | lus - yes, we can RE all closed localisation packages, it's been already done for osso-pdf-viewver, modest, camera-ui, etc | 02:37 |
| freemangordon | luf: ^^^ | 02:37 |
| freemangordon | (ENAME) | 02:37 |
| luf | Ok :( Any volunteer? :D | 02:38 |
| luf | freemangordon: where can I find such localizations? In extras? | 02:38 |
| freemangordon | merlin1991: what was the magic command to RE .mo to .po? I know you wrote it somewhere last time. :D | 02:39 |
| luf | Just I'm curious where put them (once I find time for it). | 02:39 |
| luf | msgunfmt ;) | 02:39 |
| luf | freemangordon: ^^^ | 02:39 |
| freemangordon | luf: hmm, those should be available in SDK | 02:39 |
| freemangordon | luf: aah, yes | 02:39 |
| freemangordon | exactly, msgunfmt :) | 02:39 |
| luf | I downloaded all hildon-fm packages ... But when rebuild where to store? | 02:40 |
| freemangordon | luf: what? hmm, lemme think... ummm.. cssu-devel? | 02:40 |
| freemangordon | :P | 02:40 |
| luf | sources? | 02:40 |
| freemangordon | aah, on gitorious | 02:41 |
| luf | Ugh, you're right I again overlooked something ... | 02:41 |
| freemangordon | check that one for example - https://gitorious.org/community-ssu/osso-pdf-viewer-l10n | 02:41 |
| luf | Yeah I see now. | 02:42 |
| freemangordon | what you've missed? | 02:42 |
| luf | Nothing now. | 02:43 |
| luf | Just time? :D | 02:43 |
| freemangordon | luf: BTW i looked at the commit. it looks ok to me. not that I've reviewed it thoroughly, but at first glance it seems ok | 02:43 |
| freemangordon | luf: time? hmm, RE l10n takes about 30-40 minutes :P | 02:44 |
| luf | freemangordon: what commit are you talking about? libhildonfm? | 02:45 |
| luf | I think you'll like the libhildonfm change :D | 02:45 |
| freemangordon | no, gnomevfs-extras | 02:46 |
| freemangordon | did you commit it? | 02:46 |
| luf | Yes I did also libhildonfm ;) | 02:47 |
| freemangordon | so you don;t use dbus? | 02:48 |
| freemangordon | to get the name that is | 02:48 |
| freemangordon | I wonder if it can bring troubles, though I can't think of any :) | 02:48 |
| luf | freemangordon: yes, that's it :) | 02:49 |
| luf | freemangordon: Maybe the name change will be a little bit slower but it get the icon from gconf so why not name also ;) | 02:50 |
| luf | Simple and bluez change independent :D | 02:50 |
| freemangordon | sure, though I wander what will happen if device get re-paired | 02:53 |
| freemangordon | *a device | 02:53 |
| freemangordon | device class most probably will remain the same (thus the icon), while the name could get changed | 02:54 |
| freemangordon | onthe other hand... | 02:54 |
| freemangordon | gconf key should be changed too, ain't? | 02:54 |
| luf | Sure it can and I think it is. | 02:54 |
| freemangordon | well, that is easy to be verified | 02:54 |
| luf | We'll see ;) | 02:54 |
| freemangordon | yep :) | 02:54 |
| luf | freemangordon: It seems to me that nobody wants to test it :) Even kerio ... maybe he's ill. | 02:58 |
| *** arcean has quit IRC | 03:00 | |
| *** jonwil has joined #maemo-ssu | 03:01 | |
| luf | freemangordon: do you know gnomevfs? | 03:08 |
| luf | I don't like it's prereading the "directories" so when you go into Bluetooth (sfil_li_bluetooth) it connects to all devices and reads their content. | 03:10 |
| DocScrutinizer05 | gnomevfs is a real annoyance | 03:12 |
| DocScrutinizer05 | the funny part is even poettering dislikes it | 03:13 |
| *** M4rtinK has quit IRC | 03:23 | |
| *** LaoLang_cool has joined #maemo-ssu | 03:39 | |
| *** LaoLang_cool has quit IRC | 03:43 | |
| *** luf has quit IRC | 04:03 | |
| *** int_ua has quit IRC | 04:03 | |
| *** FlameReaper has quit IRC | 04:06 | |
| *** kolp_ has joined #maemo-ssu | 05:31 | |
| *** kolp_ has quit IRC | 05:33 | |
| *** kolp has quit IRC | 05:34 | |
| *** amiconn has quit IRC | 05:43 | |
| *** amiconn_ has joined #maemo-ssu | 05:43 | |
| *** amiconn_ is now known as amiconn | 05:43 | |
| *** DocScrutinizer05 has quit IRC | 06:00 | |
| *** DocScrutinizer05 has joined #maemo-ssu | 06:00 | |
| *** ekze has quit IRC | 07:41 | |
| *** ekze has joined #maemo-ssu | 07:41 | |
| *** Jade is now known as jaded | 07:57 | |
| *** amiconn_ has joined #maemo-ssu | 08:06 | |
| *** amiconn has quit IRC | 08:06 | |
| *** amiconn_ is now known as amiconn | 08:06 | |
| *** DocAvalanche has joined #maemo-ssu | 08:08 | |
| *** jonwil has quit IRC | 08:11 | |
| *** dhbiker has quit IRC | 08:14 | |
| *** ZogG has quit IRC | 08:14 | |
| *** DocScrutinizer51 has quit IRC | 08:14 | |
| *** DocAvalanche is now known as DocScrutinizer51 | 08:14 | |
| *** dhbiker has joined #maemo-ssu | 08:20 | |
| *** dhbiker_ has joined #maemo-ssu | 08:33 | |
| *** dhbiker has quit IRC | 08:34 | |
| *** DocScrutinizer51 has quit IRC | 08:40 | |
| *** DocScrutinizer51 has joined #maemo-ssu | 08:40 | |
| *** Pali has joined #maemo-ssu | 08:43 | |
| *** joshgillies has quit IRC | 08:54 | |
| *** Pali has quit IRC | 09:18 | |
| *** luf has joined #maemo-ssu | 10:11 | |
| freemangordon | luf: will test it when I have some spare time | 10:38 |
| freemangordon | luf: does i still segfault? as I will test it on my primary, don;t want to risk a rebootloop | 11:28 |
| freemangordon | *does it | 11:28 |
| freemangordon | luf: Bad news :). using gconf to retreive device names don't work, I have 4 devices shown, but only 2 with names | 11:57 |
| freemangordon | i chacked in /var/lib/bluetooth/names and missing (in filemanager) devices are there | 11:58 |
| freemangordon | so you should revert hildonfm commit and use bluez dbus API to get device names. BTW the same should happen for the icon | 11:59 |
| freemangordon | as I have missing icons for those to devices too | 11:59 |
| *** M4rtinK has joined #maemo-ssu | 12:07 | |
| *** jonwil has joined #maemo-ssu | 12:10 | |
| *** wmarone has joined #maemo-ssu | 12:27 | |
| *** wmarone_ has quit IRC | 12:28 | |
| *** arcean has joined #maemo-ssu | 12:37 | |
| *** lizardo has joined #maemo-ssu | 12:38 | |
| *** Martix has joined #maemo-ssu | 13:05 | |
| *** kolp has joined #maemo-ssu | 13:16 | |
| luf | freemangordon: It's not funny :( However I don't have gnome icons names to change also icons. | 13:26 |
| luf | freemangordon: do you remember how you add such devices? Maybe two with names you added from N900 and the other two you added from remote devices. | 13:27 |
| luf | freemangordon: do you see segfaults? | 13:28 |
| *** arcean has quit IRC | 13:44 | |
| *** arcean has joined #maemo-ssu | 13:47 | |
| *** arcean_ has joined #maemo-ssu | 14:15 | |
| *** arcean has quit IRC | 14:15 | |
| *** arcean_ is now known as arcean | 14:15 | |
| *** wmarone has quit IRC | 14:22 | |
| *** wmarone has joined #maemo-ssu | 14:22 | |
| *** M4rtinK has quit IRC | 14:34 | |
| *** M4rtinK has joined #maemo-ssu | 14:58 | |
| *** M4rtinK has quit IRC | 14:59 | |
| *** M4rtinK has joined #maemo-ssu | 15:02 | |
| *** Pali has joined #maemo-ssu | 15:06 | |
| *** LaoLang_cool has joined #maemo-ssu | 16:06 | |
| *** iDont has joined #maemo-ssu | 16:09 | |
| *** NIN101 has joined #maemo-ssu | 16:21 | |
| *** LaoLang_cool has quit IRC | 16:25 | |
| *** luf has quit IRC | 16:48 | |
| *** Martix has quit IRC | 16:49 | |
| *** ekze has quit IRC | 17:11 | |
| * DocScrutinizer05 reports flawless update of T | 17:24 | |
| DocScrutinizer05 | haven't checked my certs yet though | 17:24 |
| DocScrutinizer05 | merlin1991: chem|st: could you take care about that X-MAEMO-changelog or whatever it's called section in pkg .deb sources, to have some description of what's new in this release, so HAM does show that | 17:26 |
| DocScrutinizer05 | for S | 17:26 |
| merlin1991 | hit chemist that he does it this time | 17:27 |
| DocScrutinizer05 | you know what i'm talking about? | 17:28 |
| Pali | DocScrutinizer05, do you mean my patch? https://gitorious.org/community-ssu/mp-fremantle-community-pr/commit/65fda141c8b92f4b5364b42e8de5e1452cf2b027 | 17:28 |
| DocScrutinizer05 | Pali: no | 17:29 |
| Pali | and what is X-MAEMO-changelog?? | 17:29 |
| DocScrutinizer05 | XB-maemo-upgrade-description in debian/control | 17:33 |
| merlin1991 | Pali: yep he means your patch (in a way) | 17:34 |
| Pali | DocScrutinizer05, my above patch doing it :-) | 17:34 |
| DocScrutinizer05 | overrides description if somebody doesn't install for first time | 17:35 |
| Pali | it generating maemo-upgrade-description from debian/changelog directly to debian/package/DEBIAN/control | 17:35 |
| DocScrutinizer05 | oooh | 17:35 |
| DocScrutinizer05 | anyway, please take care for HAM showing some "we fixed the TURKTRUST...bla bla bla" instead of standard gossip, for those who do upgrade. OTOH I don't know, maybe every fresh CSSU isnatll is already an upgrade for HAM? | 17:41 |
| DocScrutinizer05 | install* | 17:41 |
| DocScrutinizer05 | or rather, don't do anything like that for CSSU-S this time, since we didn't trst in T | 17:42 |
| DocScrutinizer05 | given we want to play safe for S, we probably shouldn't change *anything* form T to S | 17:43 |
| merlin1991 | DocScrutinizer05: any cssu install is already upgrade | 17:43 |
| merlin1991 | because the enabler postinst injects the fake mp so ham can upgrade it | 17:43 |
| DocScrutinizer05 | :nod: occured to me | 17:43 |
| DocScrutinizer05 | so we don't want pali patch I guess, not for now and not for future, rather edit "Description:" section in debian/control to include some info about this very upgrade and what it does | 17:45 |
| DocScrutinizer05 | and omit XB-maemo-upgrade-description completely | 17:45 |
| DocScrutinizer05 | right? | 17:45 |
| merlin1991 | nope | 17:45 |
| merlin1991 | because description also goes into apt-cache search and friends | 17:46 |
| merlin1991 | and there we don't need the update specific notes | 17:46 |
| DocScrutinizer05 | umpf | 17:46 |
| Pali | maemo pr1.3 metapackage does not show pr1.1 changes too | 17:46 |
| DocScrutinizer05 | then copy description section content to XB-maemo-upgrade-description section and augment it by some notes about what's new in current update | 17:47 |
| Pali | so I do not see reason to show changelog for first CSSU | 17:47 |
| DocScrutinizer05 | Pali: merlin just explained there's no such thing like a first CSSU | 17:47 |
| DocScrutinizer05 | (install) | 17:48 |
| Pali | yes | 17:48 |
| Pali | there is only update to last CSSU version | 17:48 |
| DocScrutinizer05 | yes, even for first time install | 17:48 |
| Pali | and debian/changelog in cssu metapackage is good place for *user* changelog | 17:49 |
| DocScrutinizer05 | *sigh* | 17:49 |
| DocScrutinizer05 | I don't think info in HAM for "Maemo CSSU system upgrade" should be "fixed TURKTRUST" for those who install CSSU first time | 17:50 |
| DocScrutinizer05 | it has to be a tad more verbose | 17:50 |
| DocScrutinizer05 | for now what I see in HAM "description" is fine for first time installing users | 17:53 |
| DocScrutinizer05 | once we get XB-maemo-upgrade-description, this will vanish even for those first time installers | 17:54 |
| DocScrutinizer05 | so pretty please take care that the info shown by HAM is appropriate for both first-timers and upgraders. The former wnat to learn what's a maemo5 CSSU system upgrade, the latter want to know what's new in T7.2 | 17:56 |
| merlin1991 | DocScrutinizer05: srly | 17:56 |
| merlin1991 | the former have to run the enabler | 17:56 |
| merlin1991 | hence either they learned already or something is really really wrong | 17:57 |
| DocScrutinizer05 | that's kinda flawed user experience | 17:57 |
| Pali | DocScrutinizer05, I think that HAM show on first line package name and on second line package description | 18:00 |
| Pali | and upgrade notes are shown tab when you click details | 18:00 |
| Pali | you have buttons like backup and details when you going to install cssu upgrade | 18:01 |
| DocScrutinizer05 | not afaik | 18:01 |
| kerio | add an X-maemo-install-description that takes priority over X-maemo-upgrade-description if it exists and has the information Doc wants | 18:01 |
| Pali | kerio, not possible | 18:02 |
| Pali | HAM is part of CSSU | 18:02 |
| kerio | ...oh, right | 18:02 |
| kerio | heh | 18:02 |
| Pali | so you need to upgrade metapackage for new HAM | 18:02 |
| Pali | but I think that HAM really show package name & description in update window | 18:02 |
| Pali | and not maemo-upgrade-description | 18:03 |
| DocScrutinizer05 | "XB-maemo-upgrade-description:" overrides "description:" | 18:05 |
| DocScrutinizer05 | usually for upgrades only, for CSSU though unconditionally | 18:05 |
| DocScrutinizer05 | if merlin1991's info been correct | 18:06 |
| DocScrutinizer05 | so if something overrides some other thing unconditionally, it's basically useless | 18:06 |
| merlin1991 | well usually you don't dpkg -i a basepackage before you fire up ham to install(upgrade) it | 18:07 |
| merlin1991 | but that's what our enabler does | 18:07 |
| merlin1991 | in order to bootstrap the mp | 18:07 |
| Pali | DocScrutinizer05, cssu is installed via enabler | 18:08 |
| Pali | so you will see install info in enabler package | 18:08 |
| kerio | hm, can't HAM directly *install* a system package? | 18:08 |
| Pali | so here is info what CSSU is | 18:08 |
| merlin1991 | 10 points for Pali :) | 18:08 |
| *** iDont has quit IRC | 18:08 | |
| Pali | kerio, problem is that you can have only one mp-fremantle metapackage | 18:09 |
| kerio | so? make HAM do its thing | 18:09 |
| kerio | if it can | 18:09 |
| Pali | so we need enabler, which uninstall all metapackages and install that from cssu | 18:09 |
| DocScrutinizer05 | Pali: I don't care when the content of my microwave food is listed on manufacturer's webpage, I want it on package of the food | 18:10 |
| Pali | ham does not support something like provides & replaces | 18:10 |
| Pali | so ham cannot do it | 18:10 |
| kerio | Pali: are you sure? it works for kernel, kernel-flasher and kernel-modules | 18:10 |
| Pali | kerio, not working | 18:10 |
| Pali | we have also problem with operator name widget update | 18:11 |
| DocScrutinizer05 | and friggin hell, how hard can it be to concatenate content of "description:2 and changelog into "XB-maemo-upgrade-description:" ? | 18:11 |
| kerio | Pali: why does fmg claim that it works then? | 18:11 |
| Pali | also on irc there was info that first is needed to update operator name widget and then cssu metapackage | 18:11 |
| Pali | DocScrutinizer05, why? | 18:12 |
| Pali | if you want to install CSSU you need: | 18:12 |
| Pali | 1. open web package and click on install button | 18:12 |
| Pali | 2. HAM will show that you are going to install package enabler which installing CSSU, there is written what CSSU is .... | 18:13 |
| Pali | 3. open menu and click on continue cssu installation | 18:13 |
| Pali | 4. HAM will show update and you will update system packages | 18:13 |
| kerio | btw, why can't we ship mp-fremantle-generic-pr? | 18:14 |
| Pali | kerio, becuase generic is on for global FW | 18:14 |
| kerio | really? the other ones have special metapackages? :o | 18:14 |
| Pali | there are variants: mp-fremantle-<code>-pr | 18:14 |
| Pali | yes | 18:14 |
| Pali | DocScrutinizer05, steps 1, 2 show cssu info | 18:15 |
| DocScrutinizer05 | Pali: I don't get your reasoning why you refuse to add the very info we got on *every* update until now | 18:18 |
| DocScrutinizer05 | it's poor product maintenance to not provide user with basic info about the opject he just looks at | 18:19 |
| Pali | now we have: | 18:20 |
| Pali | Maemo-Display-Name: Maemo 5 Community SSU Package | 18:20 |
| Pali | Description: Community SSU package, includes latest fixes from Gitorious and the community | 18:20 |
| DocScrutinizer05 | and til now we did provide such info in first installation, now you say you want to ditch that to *replace* it with changelog. Why? Can't you show both?? | 18:20 |
| Pali | and scripts will generate: | 18:21 |
| kerio | DocScrutinizer05: hm, adding both would be fine | 18:21 |
| Pali | Maemo-Upgrade-Description: Changelog for $(VERSION): | 18:21 |
| Pali | DocScrutinizer05, do you really want to see that Description line in Maemo-Upgrade-Description? | 18:21 |
| kerio | will HAM display it, otherwise? | 18:21 |
| Pali | If yes write some short first line for Maemo-Upgrade-Description: | 18:22 |
| DocScrutinizer05 | YES! rathe that only, than changelog ONLY | 18:22 |
| kerio | Pali: i'll write it | 18:22 |
| kerio | "Community SSU package, includes latest fixes from Gitorious and the community" | 18:22 |
| kerio | there :D | 18:22 |
| Pali | ok, we can test *what* and *where* will be displayed in next update | 18:26 |
| Pali | and then we can change lines which will be problematic | 18:26 |
| DocScrutinizer05 | fine with me | 18:27 |
| Pali | if HAM is really non deterministic with those descriptions (and lines) we can duplicate some lines between description and upgrade-desc. | 18:28 |
| DocScrutinizer05 | my general suggestion would be: copy content of "Description:" to "XB-maemo-upgrade-description:", then *append* changelog | 18:28 |
| DocScrutinizer05 | simple as that | 18:28 |
| Pali | full description? or only first line? | 18:29 |
| Pali | and where to copy it? at begining? or to append at end? | 18:29 |
| DocScrutinizer05 | err, the relevant info | 18:29 |
| DocScrutinizer05 | to beginning, then append changelog | 18:29 |
| Pali | DocScrutinizer05, better wait for next cssu-t update and see how and what do change | 18:29 |
| DocScrutinizer05 | Pali: basically next update tells me nothing since this is about those users who install firsttime | 18:30 |
| Pali | there is no "first time" installation of metapackage | 18:31 |
| Pali | only upgrade | 18:31 |
| DocScrutinizer05 | but i'm able to figure what stuff will look like, for firsttime and for update (which is no differnece according to merlin1991 and my estimation), and for differnet way to generate that text | 18:32 |
| Pali | first time is installing cssu enabler | 18:32 |
| Pali | which was not touched | 18:32 |
| *** jonwil has quit IRC | 18:32 | |
| Pali | cssu enabler installing via dpkg -i in terminat cssu metapackage | 18:32 |
| DocScrutinizer05 | Pali: sorry I'm a tad upset, so a last time: I'm not talking about enabler! | 18:36 |
| DocScrutinizer05 | frankly spoken, average lusers tend to forget that they have to run that upgrade in HAM after installing enabler | 18:37 |
| DocScrutinizer05 | and when they look into HAM and see maemo CSSU system upgrade, they get puzzled when it says "this is bringing you a fix for TURKTRUST" | 18:38 |
| DocScrutinizer05 | and during enjoying that puzzling feeling, they forget that they have run enabler 20 min ago (yes it will be 20 min since they did) | 18:39 |
| freemangordon | guys, Pali's patch was tested with latest -thumb, and IMO this is how all this shit have to look like | 18:45 |
| freemangordon | first tab - package description ("latest gratest studd") | 18:46 |
| freemangordon | *stuff | 18:46 |
| DocScrutinizer05 | freemangordon: I beg to differ | 18:46 |
| freemangordon | second tab: changelog | 18:46 |
| Pali | so I remember correctly that changelog is on separate tab? | 18:47 |
| freemangordon | yes | 18:47 |
| Pali | DocScrutinizer05, so why did you think that it wasnt? | 18:47 |
| DocScrutinizer05 | look, I did all that shit with starhash enabler (and i guess I was the first to ever use XB-maemo-upgrade-description), and for all I can recall XB-maemo-upgrade-description overrides Description: | 18:48 |
| freemangordon | DocScrutinizer05: is it a bit different: | 18:48 |
| freemangordon | Maemo-Upgrade-Description: | 18:48 |
| DocScrutinizer05 | maybe HAM chnaged since then | 18:48 |
| freemangordon | and it is on a second tab called "description" | 18:49 |
| *** ZogG has joined #maemo-ssu | 18:49 | |
| *** ZogG has joined #maemo-ssu | 18:49 | |
| freemangordon | so we have 3 tabs | 18:49 |
| * DocScrutinizer05 headdesks a bit and puts it on "bash them later" | 18:50 | |
| freemangordon | works gor me :D | 18:50 |
| freemangordon | *for | 18:50 |
| freemangordon | the fuck, obexftp works in osso-file-manager, kudos to luf :) | 18:54 |
| DocScrutinizer05 | freemangordon: than name in file */starhash-enabler/debian/control is "XB-Maemo-Upgrade-Description:" | 18:57 |
| DocScrutinizer05 | s/than/the/ | 18:57 |
| infobot | DocScrutinizer05 meant: freemangordon: the name in file */starhash-enabler/debian/control is "XB-Maemo-Upgrade-Description:" | 18:58 |
| Pali | XB mean include only in binary deb package (not to source dsc file) | 18:58 |
| DocScrutinizer05 | ...and it's supposed to override "Description:" tag in same file, for updates only | 18:58 |
| *** Martix has joined #maemo-ssu | 18:59 | |
| DocScrutinizer05 | mybe i'm completely wrong | 19:00 |
| DocScrutinizer05 | you could install starhash-enabler=0.1 and then upgrade to 0.2 to see the effects | 19:00 |
| DocScrutinizer05 | in HAM | 19:01 |
| freemangordon | Pali: so in actual debian/control XB_ is stripped? | 19:04 |
| freemangordon | *XB- | 19:04 |
| Pali | yes | 19:04 |
| Pali | XB- is prefix only for *debian*/control | 19:04 |
| Pali | in dsc or deb package is Xsomething- stripped | 19:05 |
| freemangordon | however - doc, you should really see how it looks like | 19:05 |
| DocScrutinizer05 | as long as firsttime-installers see "Community SSU package, includes latest fixes from Gitorious and the community" (or whatever) I'm stopping any rant | 19:07 |
| freemangordon | DocScrutinizer05: TBH I don't like ^^^ either | 19:07 |
| freemangordon | as it means absolutaly nothing :D | 19:07 |
| freemangordon | *absolutely | 19:08 |
| DocScrutinizer05 | it means this is a package that comes with more than just what's mentioned in changelog | 19:08 |
| freemangordon | well, lets check on gitorious, what is the first-time-installer description | 19:08 |
| freemangordon | hmm, how is it called? | 19:09 |
| DocScrutinizer05 | I don't give a fsck about gitorious, I'm talking about lusers looking at HAM, after they already almost gave up on that shit ever finishing (or even starting) to do anything | 19:09 |
| freemangordon | sure | 19:10 |
| freemangordon | DocScrutinizer05: they see that message. At least I saw it yesterday | 19:11 |
| DocScrutinizer05 | when those lusers see a maemo-cssu-system-upgrade then that tells them "install me to fix break in last version's foobar" then the average luser says "DUH! I never even *had* foobar, so why install this upgrade?" | 19:11 |
| freemangordon | XB-maemo-upgrade-description text appears on the different tab | 19:12 |
| freemangordon | you should explicitly click on it | 19:12 |
| freemangordon | are you fine wuth that? | 19:12 |
| freemangordon | *with | 19:12 |
| DocScrutinizer05 | yes | 19:12 |
| freemangordon | ok, good :) | 19:13 |
| DocScrutinizer05 | yesterday on T7.2 upgrade I actually missed the *changelog* that says what will actually come with this particular upgrade | 19:13 |
| freemangordon | DocScrutinizer05: Pali pushed the changes after it was out | 19:14 |
| freemangordon | or you mean on TMO? | 19:14 |
| DocScrutinizer05 | but I don't want the changelog to *replace* the description I've seen yesterday, for fisttime installers | 19:14 |
| freemangordon | it don't | 19:14 |
| DocScrutinizer05 | I'm talking about HAM, freemangordon | 19:14 |
| freemangordon | *doesn't | 19:14 |
| freemangordon | me too | 19:15 |
| freemangordon | DocScrutinizer05: lets leave it as it is and continue the discussion (if needed) after the next T | 19:16 |
| DocScrutinizer05 | and for all I ever knew, XB-maemo-upgrade-description *replaces* Description: | 19:16 |
| freemangordon | it does not, at least for system upgrades, I saw it with my eyes | 19:17 |
| DocScrutinizer05 | well, strange then | 19:17 |
| DocScrutinizer05 | for starhash-enabler it did | 19:17 |
| freemangordon | which PR was that? 1.2? | 19:18 |
| DocScrutinizer05 | 1.0.1? | 19:18 |
| freemangordon | well, seems HAM had evolved since | 19:18 |
| DocScrutinizer05 | quite possible | 19:18 |
| DocScrutinizer05 | I admitted that above | 19:18 |
| freemangordon | however, it looks pretty nice, you'll see | 19:19 |
| DocScrutinizer05 | I even seem to recall I "raised a bug2 against it | 19:19 |
| DocScrutinizer05 | so maybe they actually fixed it | 19:19 |
| freemangordon | check it, it might got resolved :P | 19:19 |
| DocScrutinizer05 | something like "get changelog on separate tab PLZKTNX" | 19:20 |
| *** NIN101 has quit IRC | 19:21 | |
| DocScrutinizer05 | I probably never noticed, since ^^^ "i guess I was the first to ever use XB-maemo-upgrade-description" | 19:22 |
| DocScrutinizer05 | and the only one, more or less | 19:22 |
| DocScrutinizer05 | I'll shut up now | 19:23 |
| DocScrutinizer05 | wait and see next upgrade :-) | 19:23 |
| DocScrutinizer05 | sorry for the noise, if I was wrong | 19:24 |
| Pali | freemangordon, I have updated all packages from cssu-thumb & devel but still getting reboot loop | 19:34 |
| *** NIN101 has joined #maemo-ssu | 19:34 | |
| Pali | how to fix that zlib1g reboot loop problem? | 19:34 |
| freemangordon | Pali: afaik you should not use zlib and libxml2 from devel | 19:35 |
| freemangordon | use -thumb version | 19:35 |
| kerio | D: | 19:35 |
| freemangordon | *versions | 19:35 |
| kerio | oh | 19:35 |
| Pali | libxml2 2.6.32.dfsg-5maemo4+0m5+0cssu0-thumb0 | 19:35 |
| Pali | zlib1g 1:1.2.7.dfsg-13maemo1+0cssu1+thumb0 | 19:36 |
| Pali | still reboot loops | 19:36 |
| freemangordon | 2.6.32.dfsg-5maemo4+0m5+0cssu1+thumb0 | 19:36 |
| freemangordon | libxml2 ^^^ | 19:36 |
| kerio | Pali: wait, -thumb0? | 19:36 |
| freemangordon | zlib1g 1:1.2.7.dfsg-13maemo1+0cssu1+thumb0 | 19:37 |
| freemangordon | those are the correct versions | 19:37 |
| Pali | http://pastebin.com/U7fwJ0yZ | 19:37 |
| freemangordon | apt-get update? | 19:37 |
| Pali | I have re-run apt-get update and upgrade before | 19:37 |
| freemangordon | NFC | 19:37 |
| kerio | Pali: sudo apt-get install libxml2=2.6.32.dfsg-5maemo4+0m5+0cssu1+thumb0 | 19:38 |
| kerio | freemangordon botched the versions again :s | 19:38 |
| Pali | now I see that | 19:38 |
| Pali | 0cssu0-thumb0 | 19:38 |
| Pali | 0cssu1+thumb0 | 19:38 |
| freemangordon | http://maemo.merlin1991.at/cssu/community-thumb/pool/free/libx/libxml2/ | 19:38 |
| kerio | yeah, apt-get is weird | 19:38 |
| kerio | freemangordon: well, even the non-thumby one should work now, right? | 19:38 |
| freemangordon | kerio: no | 19:39 |
| freemangordon | afaik | 19:39 |
| kerio | still the dependency issue? | 19:39 |
| kerio | oh well | 19:39 |
| freemangordon | Pali: where did you get that from? | 19:39 |
| Pali | I have no idea | 19:39 |
| Pali | I only used apt-get | 19:39 |
| Pali | freemangordon, we need some tool which list packages with higher version in status | 19:40 |
| kerio | Pali: the issue is just -thumb | 19:40 |
| kerio | instead of +thumb | 19:40 |
| kerio | right now, only xserver-xomap, xserver-xorg-core and fennec have that "problem" | 19:41 |
| freemangordon | Pali: WTF? | 19:41 |
| freemangordon | https://gitorious.org/community-ssu/libxml2/commit/cf3eba1ba26ad5d323f28df38d4c94ca9a95bb94 | 19:41 |
| kerio | and it's not a problem in fennec's case, because the next version would be 18 anyway | 19:41 |
| freemangordon | 3 months ago | 19:41 |
| kerio | Pali: dpkg -l | grep -- -thumb | 19:41 |
| kerio | (and ignore hildon-thumbnail) | 19:41 |
| freemangordon | yep, very good idea | 19:41 |
| Pali | we have similar script in cssu disabler | 19:42 |
| Pali | which find packages which are not in any repo | 19:42 |
| freemangordon | Pali: check for similar please | 19:42 |
| kerio | Pali: and what do you do with those? | 19:42 |
| freemangordon | Pali: you don't upgrade with HAM, do you :P | 19:43 |
| kerio | anyway, just pin the whole repo at 1001 priority | 19:43 |
| kerio | so it'll "downgrade" to its latest version | 19:43 |
| kerio | willingly, i mean | 19:43 |
| freemangordon | kerio: no | 19:44 |
| freemangordon | it will get higher priority that -T then | 19:44 |
| freemangordon | aiui | 19:44 |
| kerio | freemangordon: just to fix the versions i mean | 19:44 |
| kerio | then put it back at normal | 19:44 |
| kerio | i actually have community-thumb at priority 600 | 19:44 |
| kerio | so i won't actually go through cssu-testing | 19:44 |
| freemangordon | aah, ok | 19:44 |
| DocScrutinizer05 | btw I witnessed one user who probably fubared his system by "uninstalling" CSSU | 19:53 |
| DocScrutinizer05 | took like two days on #maemo to do the 1st-level-helpdesk for that user, and sorry I can't recall any details anymore, just that my conclusion been that CSSU uninstall is flawed by concept, never can work under certain conditions and most likely should get nuked | 20:02 |
| DocScrutinizer05 | similarly that set of catalogs in HAM eventually MUST vanish, since lusers can't be stopped to enable e.g. CSSU-devel catalog, once it's available under catalogs | 20:05 |
| kerio | DocScrutinizer05: +1, cssu-devel is dangerous, and we don't really need the domain information for it, people who are supposed to use it are knowledgeable enough to use apt | 20:07 |
| kerio | to test HAM deployment there's cssu-future | 20:07 |
| DocScrutinizer05 | I also don't get the rationale to have CSSU-T repo in CSSU-S catalog list | 20:09 |
| DocScrutinizer05 | this has to go away | 20:09 |
| DocScrutinizer05 | catalogs that are there will inevitably get enabled | 20:09 |
| DocScrutinizer05 | that's joe average user's nature | 20:10 |
| kerio | i'm not sure if you can add the domain information without the catalogs though | 20:10 |
| DocScrutinizer05 | if there's no way to manage catalogs in a sane way via HAM catalogs GUI, cssu must ship enabler for users to run, or exploit red-pill-mode in HAM or whatever | 20:12 |
| DocScrutinizer05 | no way CSSU-T or even CSSU-S shall ever ship e.g. CSSU-devel catalog preconfigured. Same applies for CSSU-T catalog in CSSU-S | 20:13 |
| merlin1991 | hm we discussed this quite some time ago | 20:14 |
| DocScrutinizer05 | no normal user of CSSU-S will ever want to get CSSU-T or CSSU-devel catalog enabled | 20:14 |
| merlin1991 | what was the problem with putting each catalog into a own package? | 20:14 |
| DocScrutinizer05 | for those who know what they're doing, we could ship enabler binary to run | 20:15 |
| DocScrutinizer05 | if that's even needed, since those experts should be able to edit config files | 20:15 |
| *** MrPingu has joined #maemo-ssu | 20:17 | |
| *** freemangordon has quit IRC | 20:48 | |
| *** freemangordon has joined #maemo-ssu | 21:06 | |
| *** dafox has joined #maemo-ssu | 21:23 | |
| *** nox- has joined #maemo-ssu | 21:24 | |
| dafox | hi all. I just wanted to thank you for making -thumb possible, and for helping me set up -thumb. It seems to be working great :) | 21:38 |
| freemangordon | luf: https://gitorious.org/community-ssu/libhildonfm/commit/f40a6ed07054380b34eaac995e8d19df1c8657f2 | 21:41 |
| DocScrutinizer05 | dafox: for all we know -thumb is working fine, just a tad cumbersome when you have to reflash or anything | 21:47 |
| DocScrutinizer05 | since there's no way back from a thumb system to a stock one | 21:47 |
| dafox | I think (hope) that -thumb is the future for the cssu, and that there will be no need to go back :) | 21:58 |
| DocScrutinizer05 | dafox: well, it's not about 2downgrading", it's about stock resources missing support for thimb, resulting in a broken system when user does not follow exactly the right sequence of things to do in a reflash for example | 22:00 |
| dafox | I'm sorry, I don't understand? Are you saying that with -thumb it is possible to have a non-working phone, even after re-flash? | 22:01 |
| DocScrutinizer05 | for instance installing uBoot to a cssu-thumb system already renders your system bootlooping due to uBoot coming with stock kernel | 22:01 |
| kerio | dafox: to run thumb2 binaries, you need a particular patch in the kernel you're running | 22:02 |
| DocScrutinizer05 | doing a reflash to stock and then restoring apps etc again renders system bootlooping, since you need a kernel supporting thumb | 22:02 |
| kerio | if that patch is lacking, shit will not fly | 22:02 |
| dafox | yes, I understood that, but I thought that reflashing reverted everything to stock (includign the kernel) | 22:03 |
| kerio | dafox: reflashing will give you a working system | 22:03 |
| kerio | it's not your system though | 22:03 |
| DocScrutinizer05 | dafox: that's exactly the problem | 22:03 |
| kerio | basically, you have to be very wary of what kernel you flash | 22:04 |
| DocScrutinizer05 | dafox: the only working sequence of things to do to reflash is: reflash Pr1.3 fiasco, insatll most recent powerkernel, only _then_ restore your system from your backup | 22:05 |
| kerio | DocScrutinizer05: actually, not really :) | 22:06 |
| DocScrutinizer05 | maybe I forgot ath | 22:06 |
| kerio | kernel-cssu has a different uname than kernel-power | 22:06 |
| DocScrutinizer05 | sth | 22:06 |
| DocScrutinizer05 | mhm | 22:06 |
| kerio | so it won't load the modules, even if they're pretty much the same | 22:06 |
| freemangordon | kerio: which modules? | 22:07 |
| DocScrutinizer05 | my point is it's not exactly foolproven | 22:07 |
| kerio | freemangordon: dem modules | 22:07 |
| freemangordon | is it backupmenu we talk about? | 22:07 |
| kerio | yep | 22:07 |
| dafox | aha, ok. So I've never had to restore a backup before. I always back up both optfs and rootfs, so does that then not include the kernel? | 22:07 |
| kerio | there's no easy way to install kernel-cssu on pr1.3 | 22:07 |
| freemangordon | kerio: what? | 22:08 |
| kerio | dafox: the kernel is in a separate partition | 22:08 |
| DocScrutinizer05 | dafox: nope | 22:08 |
| kerio | freemangordon: say you accidentally the whole system | 22:08 |
| dafox | ok, so simply restoring from backup without flashing works ok | 22:08 |
| kerio | you have BM backups made when you had kernel-cssu | 22:08 |
| kerio | to restore them with backupmenu you have to reflash pr1.3 | 22:08 |
| freemangordon | kerio: enable cssu-thumb repo in HAM, and you're fine | 22:08 |
| kerio | install backupmenu | 22:08 |
| freemangordon | cssu-thumb will appear | 22:09 |
| kerio | install kernel-cssu from the cssu-thumb repo | 22:09 |
| kerio | then restore | 22:09 |
| dafox | wasn't there an option in the flasher to flash the kernel only? Or some way to flash just a kernel? | 22:09 |
| freemangordon | and? i'd rather say this is a problem of BM, not -thumb | 22:09 |
| kerio | freemangordon: indeed | 22:09 |
| DocScrutinizer05 | dafox: yes | 22:09 |
| freemangordon | dafox: yes, there is | 22:09 |
| kerio | but it's also a problem of cssu-thumb users :) | 22:09 |
| kerio | dafox: yep, you'd have to install backupmenu on a fresh system to load the backup | 22:10 |
| kerio | then flash the appropriate kernel | 22:10 |
| kerio | then you can boot again | 22:10 |
| freemangordon | kerio: so, we conclude that -thumb users are screwed, because some third-party app does not support kernel restore? meh | 22:10 |
| kerio | my, how defensive | 22:11 |
| DocScrutinizer05 | freemangordon: we didn't conclude anything like that | 22:11 |
| dafox | suppose that you mess this up somehow (-thumb restored userspace, non -thumb kernel), could you still flash the system? | 22:11 |
| kerio | dafox: yep | 22:11 |
| kerio | there's very little that you can do on the software side to really ruin a N900 | 22:11 |
| DocScrutinizer05 | indeed | 22:12 |
| freemangordon | dafox: as a general rule - nothing can stop you from flashing but broken USB port :) | 22:12 |
| dafox | ok, I'm not really seeing the problem then... besides having a momentarily unbootable n900 | 22:12 |
| kerio | even blowing up CAL leaves you with an almost completely working n900 | 22:12 |
| kerio | you're basically missing bluetooth at that point | 22:12 |
| kerio | and toying with bluetooth-sysinfo would fix that, too | 22:12 |
| freemangordon | kerio: wifi MAC is not in CAL? | 22:12 |
| DocScrutinizer05 | dafox: it's just cumbersome, not a "problem" | 22:12 |
| kerio | freemangordon: iirc sambo7 (the nandtester) had working wifi and phone | 22:12 |
| kerio | freemangordon: is there a way to rebuild a CAL? | 22:13 |
| freemangordon | hmm, good. i've always tought wifi MAC is in CAL | 22:13 |
| freemangordon | kerio: sure | 22:13 |
| DocScrutinizer05 | [2013-01-08 20:47:31] <DocScrutinizer05> dafox: for all we know -thumb is working fine, just a tad cumbersome when you have to reflash or anything | 22:13 |
| dafox | but doesn't the only way to 'break' this already include flashing a new kernel? Flashing = cumbersome? | 22:13 |
| freemangordon | kerio: acrually it rebuilds by itself | 22:13 |
| kerio | freemangordon: ...really? :o | 22:14 |
| kerio | even after writing random data to its partition? | 22:14 |
| kerio | hm, maybe you need a flash_eraseall on it | 22:14 |
| DocScrutinizer05 | dafox: you still don't get it. The flashing resets your kernel, and that *is* your problem | 22:14 |
| freemangordon | kerio: CAL has 2 system and one user part | 22:14 |
| kerio | dafox: you need to unreset it from your pc's side | 22:15 |
| freemangordon | those 2 are used as some kind of RAID1 | 22:15 |
| kerio | aka download the kernel-cssu deb, unpack it, grab the .fiasco and flash it | 22:15 |
| freemangordon | if both become unusable, libcal writes to the first | 22:15 |
| dafox | DocScrutinizer05: we're talking about the sequence flash pr1.3 -> install backupmenu -> restore breaking boot, right? | 22:15 |
| kerio | neat | 22:15 |
| DocScrutinizer05 | dafox: basically yes | 22:16 |
| kerio | dafox: yep, you have to do flash pr1.3 -> install backupmenu -> restore -> flash kernel-cssu | 22:16 |
| freemangordon | kerio: you can install it from the device, no need of a PC | 22:16 |
| kerio | or that^ | 22:16 |
| freemangordon | and anyway you need a PC to flash it | 22:16 |
| kerio | freemangordon: it's actually faster to flash it from a PC though, if you have to restore a full backup anyway | 22:17 |
| kerio | dafox: or, if you haven't changed the kernel, boot from rescueOS -> untar | 22:17 |
| dafox | DocScrutinizer05: ok, but what I was thinking is that why do I need to reflash to restore? Can't one just boot backup menu and restore, without reflashing pr1.3? | 22:17 |
| kerio | dafox: if you can still run it, sure | 22:17 |
| kerio | then you avoid all problems | 22:17 |
| freemangordon | dafox: :nod: | 22:17 |
| DocScrutinizer05 | dafox: usually yes | 22:17 |
| DocScrutinizer05 | dafox: unless you entered a bootloop or sth | 22:17 |
| freemangordon | isn;t BM part of preinit script? | 22:18 |
| dafox | I think backupmenu is like grub, in that it runs before the kernels boots | 22:18 |
| dafox | no? | 22:18 |
| DocScrutinizer05 | there are however situations where you need a reflash, and then it becomes cumbersome to restore your complete thumb system | 22:18 |
| kerio | dafox: nope, bootmenu runs after the kernel is loaded | 22:19 |
| kerio | DocScrutinizer05: we really need robbiethe1st to port backupmenu to rescueos :) | 22:19 |
| *** kolp_ has joined #maemo-ssu | 22:19 | |
| DocScrutinizer05 | :nod: | 22:19 |
| DocScrutinizer05 | and include kernel backup | 22:19 |
| dafox | recently I noticed backup menu offering to boot from a different partition or something, so that does something weird to replace the running kernel then? | 22:20 |
| *** kolp has quit IRC | 22:20 | |
| MrPingu | Basically it's tarring the whole partition, no? | 22:20 |
| kerio | MrPingu: the whole partitions | 22:20 |
| DocScrutinizer05 | MrPingu: nope, it tars the whole fs | 22:20 |
| kerio | ..er, yes | 22:20 |
| DocScrutinizer05 | BM basically doesn't know about partitions | 22:21 |
| kerio | on restore, it cleans the ubifs partition from all files - possibly because nobody really understands ubi and ubifs | 22:21 |
| DocScrutinizer05 | it's no disk image | 22:21 |
| kerio | then does mkfs.ext3 on the homefs/optfs | 22:21 |
| kerio | and untars there | 22:21 |
| dafox | DocScrutinizer05: ok, I understand now, so you have two steps involving the flasher, instead of just one | 22:21 |
| kerio | ...and of course also untars the rootfs image on the rootfs | 22:21 |
| DocScrutinizer05 | yep | 22:21 |
| MrPingu | Isn't that easy to write just an script for and need to be issued by user, in rescueOS | 22:22 |
| kerio | dafox: or you have to install kernel-cssu in your newly-flashed system | 22:22 |
| DocScrutinizer05 | though one of them can be done on-board with N900 flasher | 22:22 |
| kerio | MrPingu: backupmenu has such a cute interface, though :c | 22:22 |
| DocScrutinizer05 | or what kerio said ^^^ | 22:22 |
| dafox | ok, this is all very good to know :) | 22:22 |
| MrPingu | kerio: you beat me | 22:23 |
| DocScrutinizer05 | o/ | 22:24 |
| DocScrutinizer05 | bbl, need to have a shower and then coordinate migration a tad further | 22:24 |
| DocScrutinizer05 | ivgalvez back and well! \o/ | 22:24 |
| kerio | yay :D | 22:25 |
| freemangordon | good | 22:25 |
| DocScrutinizer05 | yep | 22:25 |
| DocScrutinizer05 | I should probably post it on #maemo as well | 22:25 |
| kerio | dafox: the N900 can be ridiculously restorable | 22:27 |
| kerio | you've got the usb flashing, which pretty much works every time, at worst you have to coldflash the bootloader first | 22:28 |
| kerio | i've got uboot as my kernel, so at worst i can load something on a uSD and boot from there | 22:28 |
| kerio | i've got rescueOS installed, which gives me a clean system with a shell and some restore tools | 22:28 |
| kerio | after the maemo kernel is loaded, i've got bootmenu, which gives me usb mass storage and usb networking | 22:29 |
| kerio | and the bootmenu recovery console, which gives me an on-device shell too | 22:29 |
| kerio | then there's backupmenu, which has all the nifty backup restore stuff | 22:29 |
| *** NIN101 has quit IRC | 22:33 | |
| dafox | heh, n900 <3 :D (although I've only installed backup menu) | 22:57 |
| *** mickname_ is now known as mickname | 22:57 | |
| *** lizardo has quit IRC | 23:00 | |
| kerio | dafox: what do you mean? :o | 23:06 |
| kerio | when have you got your n900? | 23:06 |
| dafox | a bit more then a year ago (christmas 2011) | 23:07 |
| kerio | oh, i misunderstood you | 23:08 |
| dafox | I'm pretty happy with it, my old phone really was old (sony-ericsson p800) | 23:09 |
| kerio | bet on WizardNumberNext being a gentoo user | 23:16 |
| MrPingu | why? | 23:16 |
| kerio | dammit! | 23:17 |
| freemangordon | kerio: fun, isn;t it? :D:D:D | 23:17 |
| dafox | nice way to categorize people | 23:20 |
| kerio | bet on dafox being a gentoo user | 23:25 |
| dafox | guily as charged :p | 23:25 |
| MrPingu | What am I? :P | 23:25 |
| dafox | *guilty | 23:25 |
| *** kolp_ has quit IRC | 23:31 | |
| *** kolp has joined #maemo-ssu | 23:35 | |
| *** xxxxxxx has joined #maemo-ssu | 23:38 | |
| *** xxxxxxx has quit IRC | 23:39 | |
| *** ekze has joined #maemo-ssu | 23:39 | |
| *** MrPingu has quit IRC | 23:41 | |
| *** MrPingu has joined #maemo-ssu | 23:42 | |
| *** arcean has quit IRC | 23:46 | |
| *** arcean has joined #maemo-ssu | 23:46 | |
Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!