[00:33:17] *** Quits: Horwitz (~mich1@p200300EC9BC29E00022268FFFE64E7C4.dip0.t-ipconnect.de) (Ping timeout: 276 seconds)
[00:45:37] *** Joins: Horwitz (~mich1@p200300EC9BC31500022268FFFE64E7C4.dip0.t-ipconnect.de)
[00:45:37] *** ChanServ sets mode: +o Horwitz
[05:07:08] *** Joins: xenos1984 (~xenos1984@2001:bb8:2002:200:6651:6ff:fe53:a120)
[05:07:08] *** ChanServ sets mode: +v xenos1984
[12:58:22] *** Quits: snooky (~snooky@dmant.ovh) (Ping timeout: 245 seconds)
[14:32:51] *** Joins: jandegr (55eadb93@85.234.219.147)
[14:32:51] *** ChanServ sets mode: +v jandegr
[14:35:13] *** Quits: xenos1984 (~xenos1984@2001:bb8:2002:200:6651:6ff:fe53:a120) (Quit: Leaving.)
[14:54:49] *** Joins: xenos1984 (~xenos1984@0947-5f85-07cb-d895-d480-87c4-07d0-2001.dyn.estpak.ee)
[14:54:49] *** ChanServ sets mode: +v xenos1984
[15:10:39] <kazer> hi there
[15:42:08] <Navit> <jkoan> hi kazer
[15:43:46] *** Joins: snooky (~snooky@dmant.ovh)
[17:11:46] <ilovekiruna> hi kazer
[17:11:48] <ilovekiruna> how jkoan
[17:11:59] <ilovekiruna> kazer, how is the process for translations?
[17:16:39] <Navit> <jkoan> Ilovekiruna: on launchpad download the po files, the  we should have a script for the rest. Upload the resulting branches and create pull requests for them, we will merge then when ci passed
[17:16:50] <Navit> <jkoan> Really ugly process...
[17:18:57] *** Joins: gernot2 (5ed826ca@dslb-094-216-038-202.094.216.pools.vodafone-ip.de)
[17:18:57] <gernot2> hi
[17:20:13] <gernot2> Is there are al download link for nightly builds? look like the old links are 404
[17:20:43] <Navit> <jkoan> Download.navit-project.org
[17:21:05] <Navit> <jkoan> Those are all the last trunk builds
[17:22:35] <gernot2> Thank you. that is what i search.
[17:22:42] <ilovekiruna> thank you jkoan
[17:23:44] <ilovekiruna> jkoan: then one question, arent po files  anyway text files? Then why do we need launchpad in the whole process?
[17:24:38] <gernot2> its to track sync source changes
[17:25:04] <ilovekiruna> gernot2: do you mean me?
[17:25:13] <Navit> <jkoan> For users who are not technical affinity
[17:25:17] <gernot2> yes
[17:26:07] <kazer> ilovekiruna: https://wiki.navit-project.org/index.php/Translations
[17:26:20] <kazer> if you're looking for another PR, we could also revisit that page and move it to readthedocs :)
[17:26:29] <kazer> https://navit.readthedocs.io/en/latest/index.html 
[17:27:31] <kazer> ilovekiruna: the benefit of using launchpad is that 1) there is some sanity checks built it and 2) we have access to hundreds of translators. people who are not involved in navit but are involved in launchpad contributed to the translations
[17:27:47] <kazer> now, it's far from being perfect for sure
[17:30:28] <ilovekiruna> ok :-)
[17:37:14] <gernot2> just testing speed issues...
[17:38:37] <gernot2> Found that a map (bayen.bin) sample for polygon with holes render in 1 second.
[17:39:10] <gernot2> The same Region actual fetched drom planet extractor render in 3 seconds.
[17:39:40] <gernot2> polygon with holes work in both maps.
[17:41:52] <gernot2> actual nightly build tomtom
[17:53:15] <kazer> gernot2: polygon with holes, you mean https://github.com/navit-gps/navit/pull/889 ?
[17:53:23] <kazer> that's a huge difference!
[17:56:54] <gernot2> its not so big. Both maps contain working holes.
[17:57:56] <gernot2> I just make a leean SD-card with both maps. So i can switch between. 
[17:58:21] <kazer> well a 3x speed improvement sounds huge to me no?
[17:58:28] <kazer> especially for low power devices
[17:58:38] <gernot2> yes.
[17:59:23] <gernot2> now i compare 2 maps with the same size...
[18:00:06] <gernot2> booting
[18:02:16] <gernot2> @kazer second try the same result
[18:05:38] <gernot2> I try to trace this down a bit. Also layout "car sinple" is affected 
[18:10:03] <gernot2> can anyone compare the rendering speed of a old or the test map to a actual planet extractor map.
[18:11:06] <gernot2> i have speed differences of factor 3 to 4
[18:12:00] <ilovekiruna> kazer: which of my two suggested fosdem topics would you prefer?
[18:12:49] <ilovekiruna> did we ever get sth out from the guy who wanted to make a rudimentary speech interface?
[18:20:20] <gernot2> Ok... got the speed back....need 16000000 MB cache instead of 6000000 MB
[18:24:43] <gernot2> that is hard on a 64M Device and near impossible on a 32M Device. 
[18:25:42] <gernot2> The question is what the Difference between the test map and the extractor map is. 
[18:27:17] <Navit> <jkoan> The extracted map has a lot of empty files
[18:27:50] <gernot2> us it possible to remove this files?
[18:28:10] <gernot2> a "local clean" ?
[18:35:10] <Navit> <jkoan> Yes, extract(unzip) the bin, remove all empty files and repack(zip) again
[18:35:30] <gernot2> i try...
[18:38:55] <gernot2> real very much empty files
[18:40:03] <Navit> <jkoan> Yes, around 200-300MB of Metadata/Header inside the zip
[18:40:29] <gernot2> find . -size 0 -exec rm {} \; still running
[18:43:46] <Navit> <jkoan> Yes, I remember my last attempt, already though it crashed ^^
[18:44:26] <gernot2> that is not good
[18:47:07] <gernot2> I see there is a index file
[18:52:10] <Navit> <jkoan> Yes, but I didn't understood that file. All others are a sector at a zoom level
[18:55:35] *** Quits: jandegr (55eadb93@85.234.219.147) (Ping timeout: 260 seconds)
[19:10:09] <kazer> ilovekiruna | kazer: which of my two suggested fosdem topics would you prefer? 
[19:10:26] <kazer> i personally like the stf thing, but maybe it's because i'm more involved in it :)
[19:10:40] <kazer> i also think that it's more ontopic : testing navit itself
[19:10:57] <kazer> the assistant could be a separate feature, interfacing with navit via dbus
[19:11:34] <kazer> jkoan> The extracted map has a lot of empty files 
[19:11:53] <kazer> but this shouldn't affect rendering.. unless we're io bound there?
[19:17:23] <gernot2> a test with a other "weak" device would be good 
[19:23:50] <kazer> i have aerostitch's wince device
[19:24:07] <kazer> i also have a tomtom Ease but gps signal isn't working
[19:24:55] <gernot2> I test on tomtom
[19:26:16] <kazer> ok i'll test it on wince then
[19:27:02] <gernot2> btw i read that all TT use a program that translates gps to nmea like gltt. May be your device use a similar  thing for a other chipset
[19:30:56] <gernot2> yes good idea
[19:46:28] <kazer> yeah i looked into that but wasn't able to get it to work. looks like this device uses a newer hardware id. hopefully we can figure it out
[19:53:37] <ilovekiruna> jkoan: coul you have a look on my additions to German?
[19:56:44] <ilovekiruna> kazer: do you have an idea how log it takes for po files to be created?
[19:57:08] <kazer> couple of minutes at most. did you request a single po or a bundle?
[19:57:17] <ilovekiruna> a single
[19:57:17] <kazer> i usually just get a bundle, it takes the same amount of time
[19:57:19] <kazer> ok
[19:57:22] <ilovekiruna> added a few things for German
[19:57:26] <ilovekiruna> not much left there
[20:02:40] <kazer> cool. sometimes it's also worth importing updates from another language
[20:02:57] <kazer> you can see when a translation was last updated in the launchpad ui, and compare with what we have in the repository
[20:12:31] <ilovekiruna> i think I will make one PR for German and one for Swedish for hacktoberfest
[20:13:30] <ilovekiruna> will try to work on those in the coming days
[20:17:31] <gernot2> a map is unusable after removing the empty files. Just tried it.
[20:21:19] <gernot2> i have to go...
[20:21:23] <gernot2> by
[20:21:46] *** Quits: gernot2 (5ed826ca@dslb-094-216-038-202.094.216.pools.vodafone-ip.de) (Quit: gernot2)
[20:28:17] <ilovekiruna> kazer: we also need to check the automatic synchronization in launchpad
[20:28:24] <ilovekiruna> once I downloaded my translations
[20:28:33] <ilovekiruna> it seems broken for almost one year
[20:33:10] <ilovekiruna> jkoan: I think both of us can finish German tomorrow
[20:34:31] <kazer> ilovekiruna: synchronization was never fully automated, because there isn't really a way (to my knowledge) to generate the po and get a link that you can use. The link has to be sent via emails
[20:35:01] <kazer> so usually i download a bundle and import it via a script, but so far without writing code to poll a mailbox we can't really do that 100% automatically
[20:35:04] <ilovekiruna> synchronization is back from github to launchpad to my knowledge
[20:35:19] <kazer> ha! you mean the template
[20:35:21] <kazer> good to know
[20:35:31] <kazer> would you mind opening an issue so that we don't forget to fix it?
[20:35:36] <ilovekiruna> sure
[20:35:55] <kazer> thanks! also if you're looking for some PRs, we could move more content from the wiki to readthedocs
[20:37:51] <ilovekiruna> thanks for the tip
[20:38:01] <ilovekiruna> you find the issue under #893
[20:47:15] <kazer> thanks!
[20:50:14] <ilovekiruna> am just a little worried that if we activate it now, it would overwrite my changes
[21:02:28] <kazer> no, this sync should only push the translation template, not the translation files
[21:02:45] <kazer> did you push your changes directly from a .po file or did you edit the translation on launchpad?
[21:03:03] <ilovekiruna> i edit on launchpad
[21:20:43] <kazer> perfect
[21:20:48] <kazer> then you're safe
[21:21:10] <ilovekiruna> despite i have to say, launchpad doesnt seem to be a good platform
[21:21:16] <ilovekiruna> got quite a few timeout errors
[21:22:09] <ilovekiruna> kazer: can you explain me what this message meant: Select a search radius from screen center in miles
[21:22:20] <ilovekiruna> what is screen center in this context
[21:35:17] *** Quits: xenos1984 (~xenos1984@0947-5f85-07cb-d895-d480-87c4-07d0-2001.dyn.estpak.ee) (Quit: Leaving.)
[21:52:32] <kazer> > launchpad doesnt seem to be a good platform 
[21:52:52] <kazer> it has plenty of issues for sure. i did evaluate a few other options like transiflex, but none was really better
[21:53:01] <kazer> happy to revisit that if you can find a better alternative though
[21:53:07] <kazer> > Select a search radius from screen center in miles 
[21:53:32] <kazer> is used to change the radius around the current center of the map (instead of your position). it's used for POI search for example
[21:53:52] <ilovekiruna> aha
[21:53:55] <ilovekiruna> now it makes sense
[21:53:59] <kazer> :)
[22:10:41] <ilovekiruna> i think with jkoans approval we can close German soon