IRC logs for #openttd on OFTC at 2020-02-07
⏴ go to previous day
00:06:03 <DorpsGek_III> [OpenTTD/OpenTTD] LordAro commented on issue #7980: Station coverage area stays visible sometimes even when turned off https://git.io/JvZHn
01:24:58 *** snail_UES_ has joined #openttd
01:42:49 *** WormnestAndroid has quit IRC
01:44:14 *** WormnestAndroid has joined #openttd
01:48:24 *** WormnestAndroid has joined #openttd
01:55:59 *** WormnestAndroid has quit IRC
02:01:14 *** WormnestAndroid has joined #openttd
04:47:39 *** heffer_ has joined #openttd
06:42:01 *** sla_ro|master has joined #openttd
07:34:56 *** andythenorth has joined #openttd
08:52:09 *** mrtmol[m] has left #openttd
09:13:57 *** andythenorth has joined #openttd
09:24:16 <DorpsGek_III> [OpenTTD/OpenTTD] nielsmh requested changes for pull request #7981: Close #7735: Implement protocol handling https://git.io/JvnJR
09:26:20 *** andythenorth has joined #openttd
09:51:37 *** andythenorth has joined #openttd
09:57:25 *** el3ktr4 has joined #openttd
09:58:15 <DorpsGek_III> [OpenTTD/OpenTTD] LordAro commented on pull request #7981: Close #7735: Implement protocol handling https://git.io/JvnUm
10:16:31 *** el3ktr4 has joined #openttd
10:39:09 *** Eddi|zuHause2 has joined #openttd
11:05:17 *** Eddi|zuHause2 is now known as Eddi|zuHause
11:06:07 <Eddi|zuHause> i don't understand cities:skylines... i had 2 types of each school, and now one type is just gone, all the buildings i had of it disappeared and i can't build it anymore
11:27:08 <Eddi|zuHause> i'm pretty sure they were original vanilla and later updated vanilla versions
11:28:01 <Eddi|zuHause> i've no clue when they were changed
11:28:18 <Eddi|zuHause> they might have been of some building sets?
11:36:50 <Eddi|zuHause> hm weird, now i seem to be able to build them again...
11:48:45 *** Sheogorath has joined #openttd
12:05:45 <Eddi|zuHause> also all of my police stations and some fire stations seem to have disappeared
12:31:17 *** jjavaholic has joined #openttd
12:47:08 *** jjavaholic has joined #openttd
13:51:43 *** WormnestAndroid has quit IRC
13:51:57 *** WormnestAndroid has joined #openttd
14:05:03 *** gelignite has joined #openttd
14:36:20 <DorpsGek_III> [OpenTTD/OpenTTD] nielsmh commented on pull request #7981: Close #7735: Implement protocol handling https://git.io/Jvntx
15:40:22 *** DorpsGek_III has joined #openttd
15:42:31 <planetmaker> the protocol handling sounds like a very nice idea :)
15:42:43 *** Hirundo has joined #openttd
15:45:30 <SpComb> as long as you're careful with untrusted / potentially malicious URLs
15:45:53 <SpComb> like, a URL handler that leads to local code execution is a bad idea
15:46:13 *** XeryusTC has joined #openttd
15:46:36 *** ericnoan has joined #openttd
15:48:23 <SpComb> heh yeah, there was a format string vuln on unknown query params
15:49:02 <SpComb> usually you use those to leak stuff, but some printf implementations also allow you to write stuff
15:50:20 <DorpsGek_III> [OpenTTD/OpenTTD] SpComb commented on pull request #7981: Close #7735: Implement protocol handling https://git.io/JvnmV
16:01:05 <spnda> When are the there going to be any precompiled binaries on bundles.openttdcoop.org for NML? They haven't been there since the NRT update, just wondering
16:01:24 <nielsm> when someone gets bundle making for windows working again, I think
16:01:53 <nielsm> also they'd be released on openttd website going forward afaiu
16:02:51 <nielsm> time to park a laptop in front of my desktop keyboard again, to work on that fluidsynth thing...
16:09:25 <planetmaker> glx implemented some github action for building it... dunno which OS. But where do they end up?
16:10:11 <planetmaker> or does my memory mix up something?
16:10:33 <andythenorth> glx tested the build on GH
16:10:43 <spnda> andythenorth: Well, kinda want to build it myself then
16:10:43 <andythenorth> I'm not sure if that generates artefacts we can distribute yet
16:11:02 <spnda> Just I have no idea about python, especially pip
16:11:04 <planetmaker> spnda, not difficult. It's python after all... so in principle it does not even need building
16:11:20 <planetmaker> spnda, did you look at my latest posting in the NML thread at tt-f?
16:11:27 <planetmaker> did you try that approach?
16:11:51 <planetmaker> the czech guy has been asking me the same... trying to teach him how to get python running
16:12:10 <planetmaker> same issue @ andythenorth ?
16:12:19 <andythenorth> oops paste fail sorry
16:13:39 <spnda> well, pip doesn't really like my python installation for some reason. Also it's trying to use Python3 instead of Python2...
16:14:04 <planetmaker> python2 won't get you anything with nml
16:14:30 <planetmaker> you need python 3.5+
16:14:44 <planetmaker> and ply and pillow modules installed
16:14:54 <spnda> ok whatever, thought it was one of those stupid python2 only softwares
16:14:55 <andythenorth> 3.8 is substantially faster in my experience, although pypy is fastest
16:15:09 <andythenorth> (3.8 faster than 3.5-3.7)
16:16:35 <spnda> I've got 3.7 I'm fine with that
16:20:27 <spnda> makes me use a very old version of pip, but I can install pyinstaller so something good atleast
16:58:17 <nielsm> wtf, I don't understand why my linux build fails to load graphics all of a sudden
16:58:50 <nielsm> grf debug says it does load the baseset without errors, but then it tried to use fallback sprites and complains about the terrain generator sprites and all kinds of stuff
17:00:48 *** Wormnest has joined #openttd
17:01:26 <nielsm> happens with both original graphics and opengfx
17:04:11 <LordAro> spnda: i find `python -m pip` works more reliably than `pip` anyway
17:07:28 *** el3ktr4 has joined #openttd
17:10:13 <nielsm> uh yeah something weird is going on here
17:10:21 <nielsm> # time (make clean && make -j4)
17:11:56 <nielsm> looks like I have ccache installed and don't remember installing it, and it's probably making mistakes
17:28:40 <nielsm> okay this might be pretty serious
17:28:46 <nielsm> the error changes between runs of the same binary
17:28:57 <nielsm> uninstalling ccache did not affect it
18:04:02 <nielsm> I started a memtest half an hour ago, hasn't found anything yet tho
18:07:06 <andythenorth> oh dear I've angered AWS billing monitor
18:07:09 <andythenorth> might have to pay :P
18:11:59 <spnda> Where is station tile rendering being handled? (Not for GUI purposes, but a placed station)
18:54:07 <nielsm> yeah memtest passed...
19:00:13 <nielsm> built again after reboot, still behaves strangely
19:00:41 <nielsm> now it occasionally loads the game then crashes during something else, and gives errors about other things
19:01:06 <nielsm> time to wipe the machine?
19:04:16 <nielsm> it's like file i/o randomly fails/gives garbage
19:09:34 <nielsm> okay it seems to have to do with my code so ignore everything I said
19:13:14 *** andythenorth has left #openttd
19:13:19 *** andythenorth has joined #openttd
19:24:04 *** Progman has joined #openttd
19:33:19 <andythenorth> or do they look identical?
19:35:07 <spnda> Close to identical, would still say they're different though
19:40:03 <andythenorth> is frigging hard drawing steam engines
19:40:11 <andythenorth> they all look identical 100% all of them
19:44:47 *** WormnestAndroid has quit IRC
19:44:52 *** WormnestAndroid has joined #openttd
19:45:19 <DorpsGek_III> [OpenTTD/OpenTTD] nielsmh opened pull request #7983: Fix #7884: [Fluidsynth] Use recommended method of setting sample rate https://git.io/Jvncz
19:45:32 *** frosch123 has joined #openttd
19:45:50 <nielsm> wrong ticket number in commit name
19:46:28 <DorpsGek_III> [OpenTTD/OpenTTD] nielsmh updated pull request #7983: Fix #7884: [Fluidsynth] Use recommended method of setting sample rate https://git.io/Jvncz
19:50:05 <DorpsGek_III> [OpenTTD/OpenTTD] LordAro approved pull request #7983: Fix #7885: [Fluidsynth] Use recommended method of setting sample rate https://git.io/JvncP
19:50:07 <nielsm> and to explain the crashes I was getting: I didn't read the console output well enough and didn't see the message about trying to lock the same mutex twice in the same thread
19:50:20 <nielsm> that error caused an abort
19:50:34 <nielsm> which caused other threads to fail at random spots
19:50:55 <nielsm> causing spurious error messages
19:52:41 <nielsm> there's also a thing where the call to set synth gain returns failure often or maybe every time, causing console spam
19:52:51 <nielsm> I'm not sure if that's related to the fluidsynth version I'm linking against
19:53:06 <nielsm> I don't remember which version it is ;)
19:54:28 *** tokai|noir has joined #openttd
19:54:28 *** ChanServ sets mode: +v tokai|noir
19:55:11 <nielsm> hmm, on second thought, I'm not sure if this is the right way to do the render routine, if it should just do a regular blocking mutex lock
19:56:21 <nielsm> on one hand it's a soft realtime routine, on the other hand you risk music playback popping because you were setting volume or querying IsSongPlaying exactly the wrong time
19:57:04 <nielsm> maybe those should not lock the mutex?
21:00:51 <DorpsGek_III> [OpenTTD/OpenTTD] nielsmh commented on pull request #7983: Fix #7885: [Fluidsynth] Use recommended method of setting sample rate https://git.io/JvnWw
21:01:14 <DorpsGek_III> [OpenTTD/OpenTTD] nielsmh merged pull request #7983: Fix #7885: [Fluidsynth] Use recommended method of setting sample rate https://git.io/Jvncz
21:32:02 *** Wolf01 is now known as Guest15821
21:36:54 <DorpsGek_III> [OpenTTD/OpenTTD] nielsmh opened pull request #7984: Fix #7525: Move autorenew setting to Basic category https://git.io/JvnlT
22:02:38 <DorpsGek_III> [OpenTTD/OpenTTD] LordAro approved pull request #7984: Fix #7525: Move autorenew setting to Basic category https://git.io/JvnlR
22:05:37 <DorpsGek_III> [OpenTTD/OpenTTD] nielsmh merged pull request #7984: Fix #7525: Move autorenew setting to Basic category https://git.io/JvnlT
22:06:45 <DorpsGek_III> [OpenTTD/OpenTTD] nielsmh updated pull request #7729: Add #7525: Allow autoreplace with same model vehicle https://git.io/JeeIF
22:22:58 *** Smedles has joined #openttd
22:26:33 *** Smedles has joined #openttd
22:36:51 <Wolf01> peter1138: humble choice
22:39:40 <DorpsGek_III> [OpenTTD/OpenTTD] frosch123 commented on pull request #7729: Add: Allow autoreplace with same model vehicle https://git.io/Jvn8K
22:40:24 *** Smedles has joined #openttd
22:42:40 <frosch123> LordAro: nielsm: you may want to branch a few days before RC1
22:42:54 <frosch123> branches involves updating various things
22:44:05 <LordAro> branch saturday evening/sunday morning, release sunday evening? :p
22:45:23 <LordAro> bumping the "major" version is just like doing a release, but without the release, right?
22:47:13 <peter1138> Wolf01, can't choose until payment, so...
22:47:16 <glx> oh nice we'll finally be able to merge #7270
22:47:53 <LordAro> i should probably review that...
22:50:41 <Wolf01> peter1138, payment should be today
22:57:14 *** peter1138 has joined #openttd
22:57:14 *** ChanServ sets mode: +o peter1138
22:58:30 <DorpsGek_III> [OpenTTD/OpenTTD] LordAro approved pull request #7979: Fix #7592: Limit road vehicle caching near destination https://git.io/Jvn4f
22:59:20 <glx> hmm there's the remote kick to fix too
22:59:53 <nielsm> yeah any kick or ban that would affect the host should be rejected in some way
23:00:43 <LordAro> kicks that affect the host are already ignored, might as well make the behaviour the same for rcon
23:00:53 <LordAro> because for whatever reason, rcon != host
23:01:48 <glx> by id is checked, by ip is not
23:02:14 <DorpsGek_III> [OpenTTD/OpenTTD] nielsmh merged pull request #7979: Fix #7592: Limit road vehicle caching near destination https://git.io/JvZSH
23:02:15 <DorpsGek_III> [OpenTTD/OpenTTD] nielsmh closed issue #7592: Road vehicles don't balance between multiple loading bays https://git.io/fjlL0
23:03:29 <LordAro> and also consider/work out if a CVE is worth filing
23:04:46 <frosch123> no, it only affects the user triggering it, not others
23:05:18 <LordAro> that's what i said, but TB & milek7 disagreed
23:05:24 <LordAro> given the crash is a use-after-free
23:05:36 <milek7> it requires rcon password anyway
23:05:59 <LordAro> which 99% of the time is the server owner, but i guess not necessarily always
23:06:29 <nielsm> is there any protection against bruteforcing rcon passwords?
23:07:49 <_dp_> what's the status of #7912? would be nice to have in 1.10
23:07:53 <frosch123> oh, the server crashes... i thought the rcon client crashes
23:08:21 <frosch123> _dp_: it would be nice if someone finished nrt
23:09:48 <nielsm> I suppose it's good enough to merge as-is, question is mostly if it should have more features or not before leaving draft
23:10:27 <_dp_> nielsm, it's useful enough as it is imo
23:11:09 <_dp_> nielsm, there is also sharing thing I want to build on top of it but that probably not happening for 1.10
23:11:40 <nielsm> I'll just rebase and do a test build, then mark as ready
23:12:23 <nielsm> I guess the storybook buttons aren't happening for 1.10
23:12:35 <glx> hmm server is protected from kick by IP it seems
23:13:51 <DorpsGek_III> [OpenTTD/OpenTTD] nielsmh updated pull request #7912: Feature: Disallow industry production changes from GS https://git.io/Jej1Z
23:13:59 <milek7> the issue is in kicking the client issuing rcon command
23:15:02 <milek7> it frees NetworkGameSocketHandler and it crashes later in ReceivePackets()
23:15:42 <nielsm> _dp_, maybe a flag that allows decreasing but not closing?
23:15:50 <nielsm> or is that getting too specific?
23:15:54 <frosch123> gread, gh is broken, i can't comment on diffs
23:16:20 <_dp_> nielsm, idk, have no use for that
23:16:36 <_dp_> nielsm, all I want is to stop all production changes rly
23:16:37 <nielsm> yeah I can't imagine why that would be useful either
23:17:12 <_dp_> as I'm sick tired fixing my rebuild-after closing thing that constantly bugs out xD
23:17:15 <DorpsGek_III> [OpenTTD/OpenTTD] frosch123 commented on pull request #7912: Feature: Disallow industry production changes from GS https://git.io/Jvn4u
23:19:43 <DorpsGek_III> [OpenTTD/OpenTTD] nielsmh commented on pull request #7912: Feature: Disallow industry production changes from GS https://git.io/Jvn4w
23:21:06 <DorpsGek_III> [OpenTTD/OpenTTD] frosch123 commented on pull request #7912: Feature: Disallow industry production changes from GS https://git.io/Jvn45
23:23:06 <DorpsGek_III> [OpenTTD/OpenTTD] nielsmh commented on pull request #7912: Feature: Disallow industry production changes from GS https://git.io/Jvn4j
23:23:11 <DorpsGek_III> [OpenTTD/OpenTTD] frosch123 commented on pull request #7912: Feature: Disallow industry production changes from GS https://git.io/JvnBe
23:31:17 <DorpsGek_III> [OpenTTD/OpenTTD] glx22 opened pull request #7985: Fix #7976: Don't kick the client doing the rcon https://git.io/JvnBt
23:31:20 <DorpsGek_III> [OpenTTD/OpenTTD] nielsmh updated pull request #7912: Feature: Disallow industry production changes from GS https://git.io/Jej1Z
23:39:08 <_dp_> actually the way it is not it won't rly stop stuff from closing
23:40:56 <nielsm> hmm yeah actually, I'm not sure this protects secondary industries from the closure if no deliveries within 5 years check
23:41:22 <_dp_> nielsm, I don't even think it protects primaries :p
23:41:30 <_dp_> if they're already on lowest production
23:42:05 <_dp_> new_prod == old_prod == 0
23:42:45 <_dp_> ah, INDCTL_NO_PRODUCTION_CHANGE protects from closing
23:43:43 <_dp_> it has newgrf issue though
23:43:59 <_dp_> so replacing it with no close flag seems like a good solution to me
23:45:34 <nielsm> should newgrf be able to query these flags? (not set them)
23:45:41 <nielsm> to at least know if it's under the effect
23:46:00 <nielsm> so nobody has to be surprised when the request to change production or close was ignored
23:46:59 <frosch123> all newgrf that are still in development (i.e. firs) don't use either of those results
23:47:36 <frosch123> hmm, or was "no closure" a setting in firs?
23:47:50 <nielsm> firs does have some settings for closure rules
23:50:10 <andythenorth> it's a parameter for secondary industry
23:50:26 <andythenorth> other industries don't close
23:50:30 <nielsm> then how about changing the first flag to a full-on stasis flag, when set the industry never does anything
23:50:36 <_dp_> I can't even imagine why would grf want to know this stuff
23:50:45 <nielsm> no cargo acceptance, no production, no callbacks, no sounds or animations
23:50:54 <frosch123> _dp_: ever played with ecs?
23:51:35 <_dp_> but I mostly want it for nogrf servers so don't rly think much about grf interaction
23:52:10 <frosch123> do you remember the drama about daylength and ecs?
23:52:41 <_dp_> I'm not paying any attention to daylength so no xD
23:53:29 <frosch123> ecs replicates the ottd mechanics internally and then submits differential updates to ottd via the callbacks. so any little behaviour change in ottd makes ecs essentially desync with ottd
23:53:44 <frosch123> so it broke badly when industries produced more than 8 or 9 times a month
23:54:02 <frosch123> and ecs even cared whether there are 8 or 9 in which month
23:54:28 <_dp_> well, I don't think anyone is going to update ecs over this anyway xD
23:54:39 <frosch123> anyway, the "more modern" industry sets do not use the production change callbacks for changing production
23:54:53 <frosch123> they just use the monthly callback for cyclic updates
23:56:37 <DorpsGek_III> [OpenTTD/OpenTTD] nielsmh updated pull request #7912: Feature: Disallow industry production changes from GS https://git.io/Jej1Z
23:57:01 *** andythenorth has left #openttd
continue to next day ⏵
