[00:31:51] *** Quits: Horwitz (~mich1@p200300EC9BC31500022268FFFE64E7C4.dip0.t-ipconnect.de) (Ping timeout: 264 seconds) [00:43:58] *** Joins: Horwitz (~mich1@p200300EC9BC28600022268FFFE64E7C4.dip0.t-ipconnect.de) [00:43:58] *** ChanServ sets mode: +o Horwitz [04:35:45] *** Joins: extorr (~extor@unaffiliated/extor) [04:38:47] *** Quits: extor (~extor@unaffiliated/extor) (Ping timeout: 245 seconds) [05:02:49] *** Joins: xenos1984 (~xenos1984@2001:bb8:2002:200:6651:6ff:fe53:a120) [05:02:49] *** ChanServ sets mode: +v xenos1984 [07:40:25] ilovekiruna: I changed another one, please review it ;) [12:57:29] *** Joins: jandegr (d5dbb11d@213.219.177.29.adsl.dyn.edpnet.net) [12:57:29] *** ChanServ sets mode: +v jandegr [14:42:06] *** Quits: xenos1984 (~xenos1984@2001:bb8:2002:200:6651:6ff:fe53:a120) (Quit: Leaving.) [15:04:27] *** Joins: xenos1984 (~xenos1984@cc7f-3a39-0136-1937-d480-87c4-07d0-2001.dyn.estpak.ee) [15:04:27] *** ChanServ sets mode: +v xenos1984 [15:12:10] hi there [15:12:34] Hi Kazer [15:12:42] 'sup jkoan? [15:13:19] Working to migrate some python code from python 2.6 to 3.5 [15:16:43] good stuff [15:20:17] And you? :) [15:21:50] next thing i'd like to finish is hooking up circleci/appium with stf to get these automated tests [15:22:05] i've also been doing some tests/updates on buildroot [15:26:25] I started my slave node for openstf yesterday. Master will be inside Docker as already discussed. My idea was to use openwrt for the pi since it already has good packages and is really small, I will see if it will work out :) [15:29:07] cool, i'm interested in this master/slave experiment, keep me posted :) [15:29:12] have you seen the appium tests? [15:51:11] But you know what I mean with the master slave concept (with openvpn)? [15:51:32] I read about it, saw the repo but didn't looked into it [15:57:37] yeah i see what you mean for the master slave. it would be nice if it works, because then my farm could become a slave as well, and all our devices would be accessible from one master, for Jan's use or for circleci [15:58:08] ok it might be worth having a quick look at appium maybe, to see how it behaves on your farm and for me to get feedback about it :) [16:02:55] Probably I find time on the weekend :) [16:20:28] ok great [16:28:55] in short, all you should need would be to call https://github.com/navit-gps/appium/blob/master/tests/test_first.py with two arguments, first one is a description of your target device (free string) and second one is the address of your remote adb device [16:29:06] python tests/test_first.py "J3-5.1.1-sdk22" 192.168.1.207:7401 [16:41:53] *** Joins: gernot2 (508abf78@p508ABF78.dip0.t-ipconnect.de) [16:41:53] hi [16:47:06] hi gernot2 [16:47:51] just translate a map with current maptool... [16:53:20] file size difference is 406 MB without and 422 MB with multipolygon [16:56:36] no speed problem with the selfmade binfile. Multipoligons with hole work well [16:59:44] that's a nice win i like the new rendering [17:48:13] is there a issue tracker for planetextractor? I think the maps need now more time/memory to render. The problem is not the multipolygon support. Was there an other change except multipolygon. I remember something with 64 bit. [17:49:05] A map converted with actual maptool work well. [17:49:22] there isn't a dedicated issue tracker for planetextractor, but you could raise an issue on navit's bug tracker [17:49:57] ok. I write some details. [17:55:09] thanks [18:36:52] hi gernot2, hi kazer [18:36:56] hi jkoan [18:37:07] hey ilovekiruna [18:37:19] jkoan: i will approve the last one. Again I see one needs a proper understanding of the english term first :-( [18:41:43] ilovekiruna: note that we can add comments in the code that become visible in the translation template to give context [18:42:29] the last suggestion by jkoan was about location [18:42:44] i interpreted it as geographical location, but should have been file path [18:43:40] kazer: I guess you will see more questions like this in the near future, when I will work more on Swedish [18:44:06] ilovekiruna: absolutely, not a problem [18:47:43] ilovekiruna: here's an example of comment from the code that will be visible in the .po : https://github.com/navit-gps/navit/blob/07006013135b619029f26df579a0ac6acb86cd4a/navit/navigation.c#L868-L869 [18:48:15] so if you want you could add something similar to for the search radius thing [18:48:27] kazer: actually here I would suggest even a lot more clutter :-p [18:48:48] I would claim this comment should be in every case of the switch statement [18:53:21] valid point [18:53:26] *** Quits: jandegr (d5dbb11d@213.219.177.29.adsl.dyn.edpnet.net) (Remote host closed the connection) [18:53:57] btw: one other question in the .po.in files [18:54:06] are the authors manually added? [18:54:25] or should it be generated automatically [19:01:28] it's added by the script that imports the translations. It regenerates the full header, including the dates, etc [19:01:54] script on our side or launchpad? [19:02:20] https://github.com/navit-gps/navit/blob/trunk/scripts/import_translation.sh [19:02:41] so you can just call that script with the url you get from launchpad [19:02:50] I c [19:04:52] for my curiousity [19:05:07] why do we clone for the import the repository compltey new? [19:11:47] because i usually run that in a temp folder. the goal is to import against trunk, with no other possible changes [19:14:23] and it seems to break on my machine [19:14:26] no clue what happens [19:14:38] the bash is closed when I run the script [19:18:32] i think the problem is that it uses git:// which requires ssh keys [19:20:06] ha! well it tries to push automatically at the end so yeah :) unless you usually use https authentication? [19:20:27] generally i actually do [19:22:49] makes sense then [19:24:42] but for the sake of the script I added a key [19:24:57] i think a key isnt helping much in terms of security [19:25:23] even if the key is encrypted, if a secret is known everything is gone [19:27:18] well you could just change the pull url [20:11:16] by [20:11:22] *** Quits: gernot2 (508abf78@p508ABF78.dip0.t-ipconnect.de) (Quit: gernot2) [21:23:35] kazer: would you prepare a first draft of the presentation, or how would you suggest doing it? [21:39:55] that's a good question [21:40:13] maybe this will help us get started :) [21:40:16] https://circleci.com/gh/navit-gps/appium/8#artifacts/containers/0 [21:40:22] https://8-211422384-gh.circle-artifacts.com/0/home/circleci/project/results/XIAOMI-Mi%20A2-sdk28-9-arm64-v8a_2019-10-02_21-38-50.html [21:41:02] we could for coordinating use your owncloud installation again [21:41:08] sure [21:42:01] how do you take the screenshots? [21:42:13] if possible i would even prefer vidos ;-) [21:42:30] https://github.com/navit-gps/appium/blob/master/tests/test_first.py#L28 [21:42:34] video is also working :) [21:43:02] wow [21:43:17] i remember when we last discussed it it seemd tricky on opentsf [21:43:23] openstf* [21:45:27] *** Quits: xenos1984 (~xenos1984@cc7f-3a39-0136-1937-d480-87c4-07d0-2001.dyn.estpak.ee) (Quit: Leaving.) [21:46:44] yeah it has quirks. the one thing that isn't working for me now is file push. i wanted to use that to push a map to the device, but it's tricky with python as you have to base64_encode it to a string, and then stf will crash because of memory issues [21:48:07] https://www.youtube.com/watch?v=6uIbeElmQYs [21:48:35] very boring as the first 45s are just appium getting the device ready :) but we'll be able to cut that out automatically [21:48:37] when i did intial tests, I downloaded via gui commands [21:49:35] how do you reset a device after the test ran? [21:49:54] appium removes the app automatically [21:50:10] > I downloaded via gui commands [21:50:13] yeah adb push works [21:50:43] btw : thank you so much for doing the initial work that made this possible [21:51:12] you are welcome [21:51:14] didnt do much [22:28:00] what exactly is this? [22:28:01] @strings/position_popup_view [22:42:48] where is that? [22:43:12] interestingly it was also not complete in UK English [22:43:32] Android resource: @strings/position_popup_view