Time Nick Message 00:16 emptty ok, the bug is fixed, thanks 01:17 us_0gb Do you know how helpful it is to have all the minetest images listed with their original license as opposed to just blanketly licensing them under CC-BY-SA? Very helpful. I'd like to thank whoever decided to do that. 06:18 celeron55 us_0gb: i guess i am mainly to blame or thank about any licensing 8) i know how much it sucks when people license things more restrictively than needed 06:21 us_0gb celeron55: Thanks! I've been thoroughly contemplating the licensing of screenshots, and have come to the conclusion that CC-BY-SA passes from the textures to the screenshots and isn't compatible with my GPL site. However, because the images are listed by their original licenses, I don't have to replace any that are under a more liberal license that is compatible with both GPL and CC-BY-SA. 07:45 kahrl ShadowNinja: yes, #920 is ready 07:45 ShadowBot https://github.com/minetest/minetest/issues/920 -- Add formspec table by kahrl 09:15 daswort which curl package do i need? 09:15 daswort p libcurl4-gnutls-dev - Entwicklungsdateien und Dokumentation für libcurl (GnuTLS-Variante) 09:15 daswort p libcurl4-nss-dev - development files and documentation for libcurl (NSS flavour) 09:15 daswort p libcurl4-openssl-dev 09:22 kahrl daswort: shouldn't matter which one 09:32 Megaf Hi everyone! 09:34 VanessaE hi 11:34 xyz ShadowNinja: no 12:19 kahrl john_minetest: just delete them from the database 12:21 kahrl no, there is no API for that 12:21 VanessaE being able to place a node that causes the game (etc) to highlight the whole mapblock as scheduled for delete would be useful. 12:22 VanessaE highlight as in, say, everything in that mapblock loses its light values or something 12:22 VanessaE (something that can be transmitted to the client with a minimal amount of fuss) 12:23 VanessaE no, there needs to be a highlight first, because you need to see where the edges are 12:23 VanessaE could use the thing worldedit does now (the selected area is enclosed in a large checkerboard box) 12:24 VanessaE but that doesn't solve the problem of actually deleting the mapblock from the database. 12:24 VanessaE feel free to code it :) 12:49 sapier VanessaE did your server crash again with 1090 or are the exceptions fixed by now? 12:50 VanessaE I have not seen any crashes, nop 12:50 VanessaE no* 12:50 sapier btw what time is it at your location? :-) 12:50 VanessaE 7:50am 12:51 sapier then good morning 12:51 VanessaE good morning. 13:17 kahrl so I'm trying to fix #1086 and realized we don't have a str_ends_with 13:17 ShadowBot https://github.com/minetest/minetest/issues/1086 -- Problems with safeWriteToFile() 13:18 kahrl but if I implement that based on str_starts_with it would take std::wstring, while the paths I would be calling this on are std::string 13:19 kahrl and doing a narrow_to_wide on each might be wasteful 13:19 Exio4 why are the paths std::string? 13:19 kahrl no idea 13:21 kahrl honestly if I rewrote minetest I'd use utf8 std::string everywhere and std::wstring only when calling OS functions, where absolutely needed 13:22 sapier I'd use std::string too but I guess that doesn't help right now ;-) 13:23 Exio4 use char*! 13:23 kahrl how about a VARIANT BSTR 13:24 sapier I don't even know waht VARIANT BSTR is ;-) 13:24 kahrl lucky 13:25 sapier what is it 13:25 sapier ? 13:25 kahrl http://msdn.microsoft.com/en-us/library/windows/desktop/ms221627%28v=vs.85%29.aspx 13:26 celeron55 kahrl: you're free to modify minetest to do so 13:26 celeron55 probably everyone agrees that using wstring anywhere else in irrlicht's interfaces is bad 13:26 celeron55 +tha 13:26 celeron55 +n 13:26 Kray using wstring in general is bad 13:26 sapier reading the first sentences I already have the "I don't ever wanna use this" feeling ;-) 13:28 kahrl celeron55: I could modify it but I can't test every setup to see what I've broken 13:28 celeron55 do you imply you can't assume narrow_to_wide and wide_to_narrow works? 13:29 kahrl we've had so many charset problems in the past that I don't really know about that 13:29 sapier something completely different, if vanessaE doesn't report an error till 8pm utc I'm gonna merge the protocol implementation fixes. If anyone has reason not to do so please tell 13:30 VanessaE personally I think you should wait another day 13:30 VanessaE it's only 1:30pm UTC 13:31 VanessaE I've barely been on 1090 for a day 13:31 kahrl e.g. when compiling with or without gettext breaks display of international characters in some parts of minetest, wtf? 13:31 VanessaE (as supplied in your pull I mean) 13:31 celeron55 kahrl: this fixes it for me: http://paste.ubuntu.com/6704413/ 13:31 celeron55 feel free to merge 8) 13:31 VanessaE celeron55: did you try that with the irc mod? :P 13:31 kahrl ah, that was why 13:32 celeron55 VanessaE: no 13:32 sapier you know the one merging it will be responsible for the things beeing broken after it ? ;-) *just kidding* 13:32 celeron55 i'm responsible for it working on my computer! but there are a few other computers 13:33 celeron55 at least my neighbor has one 13:33 kahrl what? when I try to download that patch as raw text it sends me to a launchpad login site 13:33 sapier yes for i18n it's like walking a dense minefield it's almost impossible not to break something else on fixing something 13:33 celeron55 kahrl: yeah it sucks 13:34 kahrl c&p to the rescue! 13:34 celeron55 select it and then cat > foo.patch\n^D 13:36 VanessaE celeron55: is it wrong that I've always c&p -> text editor -> save ? heh 13:36 celeron55 VanessaE: yes, it's bloated 13:37 VanessaE force of habit :) 13:37 Exio4 VanessaE: that is why you still use pastebin.ubuntu.com 13:38 VanessaE Exio4: actually no. I do that for a different reason entirely. 13:38 kahrl anyway, I'm going to merge that patch 13:38 VanessaE because my usual pastebin choice insists on wrapping the text, and has a limit on the number of pastes per day without a login 13:39 kahrl who should I set as author? 13:39 celeron55 i made it i guess 13:39 kahrl alright 13:44 kahrl well now the patch doesn't apply because the pastebin converted tabs to spaces 13:44 kahrl I'll do it by hand I guess 13:45 VanessaE actually I get spaces when viewed as raw too 13:45 VanessaE (I am logged in) 13:45 VanessaE afair, ubuntu's pastebin doesn't do that. 13:45 VanessaE weird. 13:50 kahrl lol, I set the date wrong 13:50 kahrl eh whatever 13:59 xyz not as bad as it could be https://gitorious.org/opensuse 13:59 kahrl haha 14:05 VanessaE hm, not entirely fixed. 14:05 VanessaE [01-09 09:05] ° <-- test 14:05 VanessaE renders as a blank char when it comes back through the minetest client. 14:06 VanessaE make that no char at all. 14:06 VanessaE that might be irc mod's fault though. 14:06 sapier did anyone check it on cyrillic or chinese systems? 14:07 VanessaE RECV: :VanessaE!~vanessa@24-176-135-68.dhcp.sffl.va.charter.com PRIVMSG #minetes 14:07 VanessaE t :ab <-- test 14:07 VanessaE looks like it was logged correctly though...sorta 14:09 VanessaE U+00B0 is the correct code for the symbol I typed, at least. 14:09 xyz you sure there's that char in font we use? 14:10 VanessaE idk, but i'm using freetype, DejaVu Sans font, so most likely yes 14:10 VanessaE it's just the "degrees" sign 14:11 xyz sapier: just checked it, works fine 14:12 sapier ok guess you checked on linux? 14:12 sapier actually just merge it if there's an issue someone will complain ... guess about may or june 14:13 xyz yes it's merged already you're a bit slow 14:14 sapier :-) i don't look at commit history twice a minute ;-P 14:19 VanessaE ok, confirmed 14:19 VanessaE that fixed the problem with the IRC mod 14:21 VanessaE at least half of it. 14:22 VanessaE server -> same server and server -> IRC works. IRC -> server and server -> other server via IRC deletes the foreign chars. 14:24 VanessaE sapier: someone experiencing a brief spike of what may be sign-on lag. they are using an old client (one of those tablet ports) 14:25 VanessaE or at least, suddenly the map stopped loading for them 14:25 VanessaE and they're on Nostalgia (== absolutely no heavy code, ultra-lightweight, probably even lighter than Pilzte--- nodetopia) 14:25 sapier could be the other old client bug stopping map loading completely too 14:25 VanessaE probably 14:25 sapier or the default 1024 is to big for your server too 14:26 VanessaE just keeping you up to date 14:26 sapier how many servers do you run there? 14:26 VanessaE five. 14:26 VanessaE and right now they're all pretty quiet. 14:26 VanessaE http://digitalaudioconcepts.com/vanessa/hobbies/minetest/stats.html 14:26 sapier so you should be save if you have more then 20mbit upstream 14:26 VanessaE I have 100Mbps upstream. 14:27 sapier ok then no need to change anything about the 1024 14:28 VanessaE john_minetest: it compiles so fast there's almost no time for it to grab a sample :) 14:28 sapier guess those spikes are compiling? *g* 14:29 VanessaE that last big spike was, yeah 14:29 VanessaE the two before that, backup scripts and testing stuff 14:29 VanessaE the rest, various 14:29 sapier compiling doesn't show up in disk activity? interesting 14:30 VanessaE sure it does, but the machine has 8 GB RAM so I guess a lot of that is cached. 14:30 sapier that's what I meant seems like all the interim data is cached only 14:31 sapier active objects would be even more interesting 14:31 VanessaE john_minetest: that's next on the list as soon as I learn a simple way to do so 14:31 troller john_minetest, always 100% 14:31 sapier realized about 2 seconds of lag with 1100 mobs active 14:32 troller VanessaE servers have average 0 players, so they needed ~0 kbps upstream ;) 14:32 sapier was a bug there never should've been 1100 mobs 14:33 VanessaE troller: actually, my servers have been getting a bit busy lately, especially nostalgia, thanks to those android clients 14:33 VanessaE sapier: the same bug that caused 300+ deer to show up in one area of my server? :) 14:33 sapier very likely 14:34 sapier spawning did consider incompletely spawned mobs as player spawned ... and player spawned mobs don't count for population check 14:34 VanessaE -> #minetest 14:49 Megaf__ you cpu might be fast but you disk and ram are not 14:49 Megaf__ your* 14:54 xyz wat? ram is fast 15:04 xyz another string-related thing: why do we use char* at places? in functions like getPlayer 15:05 xyz and there's a lot of C-style stuff which can be rewritten 15:05 sapier If I had to bet I'd bet on "historic reasons" 15:13 sapier I guess it's more about code safety 15:15 sapier you can't forget to free a passed std::string but have to be very carefull on char* .... of course if you allocate the string by new and pass pointers you've got same problem 15:15 xyz john_minetest: it depends 15:40 celeron55 minetest originally didn't use the C++ standard library at all; there are probably remains of that still 18:24 PilzAdam whats the status on #1090 ? 18:24 ShadowBot https://github.com/minetest/minetest/issues/1090 -- Network fixes udp 1 by sapier 18:25 celeron55 sapier is waiting on enough testing results 18:39 PilzAdam hmm... could it be that it keeps downloading media when I press Esc to cancel it? 18:39 sapier PilzAdam: by now it's on track, VanessaE wants some more time to give her opinion, but as far as I can se it's fine 18:40 sapier there's a timeframe where you can't chancel startup, no idea why, but I doubt this is related to networking protocol 18:40 sapier -h 18:41 PilzAdam no, it returns to the menu, but its still downloading stuff 18:42 sapier I see no way how this could be possible, but of course server will continue to send if no disconnect is sent 18:42 sapier as celeron didn't want to wait for disco to be approved this is possible 18:43 sapier in worst case server will stop after peer timeout 18:43 PilzAdam well, the server sent so much that my connection was locked up 18:43 sapier wow :-) 18:43 sapier should only be a couple of seconds till timeout is reached + network internal queueing 18:44 sapier but you're right there is still a open issue 18:44 PilzAdam is max_packets_per_iteration for server only? 18:45 sapier no it's client too but most likely it wont have any effect on client 18:45 sapier it's only relevant if you have more then one peer 19:49 Calinou tool times need fixes: most wood/stone tools are too slow 20:16 xyz https://github.com/minetest/minetest/blob/master/src/main.cpp#L39 why? 20:29 celeron55 because http://sourceforge.net/p/irrlicht/code/HEAD/tree/tags/release-1.8/examples/03.CustomSceneNode/main.cpp#l28 20:31 xyz no zlibwapi.lib on page from your link 20:32 celeron55 well apparently i utilized that kind of linking on some first windows builds 20:34 celeron55 you know, cmake came in only after like half a year of development 22:26 Ritchie john_minetest: skins are blue in preview formspec or if somebody uses them? 22:39 Ritchie john_minetest: what version of minetestserver and client do you have? maybe you have something bad with your skin images on server