IRC logs for #openttd on OFTC at 2021-03-16
⏴ go to previous day
00:01:31 <DorpsGek> [OpenTTD/OpenTTD] James103 commented on issue #8869: Setting resolution really small in openttd.cfg crashes the game https://git.io/JqF5J
00:02:24 <TrueBrain> he is a bit slow today :P
00:10:52 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on issue #8869: Setting resolution really small in openttd.cfg crashes the game https://git.io/JqF5J
00:13:49 <TrueBrain> x exec speed : 2.33/sec (zzzz...)
00:13:57 <TrueBrain> I love how afl tells you shit is going wayyyyy to slow :D
00:27:30 <DorpsGek> [OpenTTD/OpenTTD] James103 commented on issue #8869: Setting resolution really small in openttd.cfg crashes the game https://git.io/JqF5J
00:28:32 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
00:29:21 *** WormnestAndroid has joined #openttd
00:30:58 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on issue #8869: Setting resolution really small in openttd.cfg crashes the game https://git.io/JqF5J
00:31:29 <TrueBrain> removed my local NewGRFs .. "exec speed : 6.04/sec (zzzz...)"
00:32:08 <glx> hehe grf scan at each start
00:32:31 <TrueBrain> exec speed : 9.16/sec (zzzz...)
00:32:50 <TrueBrain> Can't gain much more without not being sure about the results
00:37:44 <TrueBrain> I should feed it a dictionary, so it can try all kind of weird values, instead of this random binary flipping of bits :P
00:41:44 <milek7> Total run time : 1 hours, 28 minutes
00:41:44 <milek7> Crashes found : 66 locally unique
00:42:11 <milek7> certainly it crashes a lot :P
00:47:57 <TrueBrain> We established months ago that fuzzing save games is not interesting atm, as there are almost no safeguards in place
00:48:14 <TrueBrain> You even pointed that out :p
00:48:58 <DorpsGek> [OpenTTD/OpenTTD] James103 commented on issue #8627: Inflation problem (since 1.11.0 beta1) - price and cost levels, therefore the difficulty level varies depending on the game start date https://git.io/JtBaj
00:49:46 <TrueBrain> You just liked trying it again didn't you? Can't blame you :D
00:49:59 <milek7> I'm somewhat curious if this is remotely-exploitable :P
00:50:21 <TrueBrain> Rb once fuzzed the whole network protocol, that was funny
00:50:29 <TrueBrain> he just send every possible packet to a server
00:51:00 <TrueBrain> I wonder if we can make a fuzzer that does user interactions of some sorts
00:51:03 <TrueBrain> like clicking random buttons
00:51:06 <TrueBrain> not sure how that would look
00:52:01 <TrueBrain> random squirrel-scripts should be possible :P
00:52:06 <TrueBrain> just learn him the language, which is relative easy
00:55:07 <TrueBrain> I wonder if we can make AFL work on DoCommands .. might be interesting too
00:55:44 <TrueBrain> well, checking if a single DoCommand on an empty map works is easy
00:55:54 <TrueBrain> for all of them, with all possible p1/p2s
00:56:28 <TrueBrain> anyway, I will let the config run overnight, just to see if it can find more odd things :P Really surprised it found 1 ..
00:58:20 <DorpsGek> [OpenTTD/OpenTTD] Bohaska approved pull request #8776: Change: Don't decrease vehicle reliability when stopped https://git.io/JqbUB
02:10:14 *** jellyknight has joined #openttd
02:17:37 *** gelignite has quit IRC (Ping timeout: 480 seconds)
02:38:57 *** Wormnest has quit IRC (Quit: Leaving)
03:52:42 *** D-HUND has quit IRC (Ping timeout: 480 seconds)
05:00:55 *** jellyknight has quit IRC (Quit: Stay safe!)
05:11:18 *** WormnestAndroid has quit IRC (Remote host closed the connection)
05:12:59 *** WormnestAndroid has joined #openttd
05:39:19 *** didac has quit IRC (Remote host closed the connection)
05:46:57 *** snail_UES_ has quit IRC (Quit: snail_UES_)
05:59:41 *** Wuzzy has quit IRC (Remote host closed the connection)
06:34:52 *** sla_ro|master has joined #openttd
06:51:21 *** andythenorth has joined #openttd
07:10:10 *** andythenorth has quit IRC (Quit: andythenorth)
07:15:14 *** Flygon has quit IRC (Ping timeout: 480 seconds)
07:18:19 *** andythenorth has joined #openttd
07:24:07 *** Flygon_ has joined #openttd
07:29:13 *** Flygon has quit IRC (Ping timeout: 480 seconds)
07:32:43 *** Flygon_ has quit IRC (Ping timeout: 480 seconds)
07:43:23 <DorpsGek> [OpenTTD/OpenTTD] nielsmh commented on issue #8869: Setting resolution really small in openttd.cfg crashes the game https://git.io/JqF5J
07:50:51 <peter1138> Hmm, content download crashes.
07:51:25 <peter1138> There's a bug report about it already.
07:52:25 <peter1138> Inside gz_open() for me. Hmm.
07:53:59 <peter1138> Is this one of those "nothing has changed here for years" ones>
07:54:38 <peter1138> Ah, no PR#8868 was fairly recent.
08:06:35 <peter1138> But I imagine unrelated.
08:13:48 <peter1138> Either something about my build, or something about my vcpkg build on zlib :/
08:15:05 <LordAro> whatever it is, it appears to have happened to 3 people now
08:15:20 <LordAro> lufkily, appears to be development environment only
08:15:23 <andythenorth> write a game, run it anywhere
08:15:27 <andythenorth> give or take the vulns
08:15:38 <andythenorth> and the platform-dependent performance
08:16:19 <peter1138> state->start = LSEEK(state->fd, 0, SEEK_CUR);
08:16:29 <peter1138> is where it fails in gzlib.c
08:16:31 <andythenorth> Make Flash Great Again
08:17:01 <LordAro> something funny going on with file descriptors
08:17:57 <Wolf01> Wow lot of talking this morning
08:18:39 <Wolf01> No, the loud noises are my neighbours cutting bricks for the garden path
08:19:53 <andythenorth> goes it throw out cargo subtypes?
08:20:08 * andythenorth has spent too long discovering which trainsets FIRS broke
08:20:44 <andythenorth> grumpy lolz ensued
08:20:54 <Wolf01> I just can't understand how you can break a trainset with an industry set
08:21:27 <andythenorth> triviiallaliailalaily
08:23:00 <andythenorth> trains use cargo subtypes to change features
08:23:11 <andythenorth> if the cargo is missing, the train is missing
08:23:19 <andythenorth> result: broken train grf
08:24:09 <andythenorth> newgrf community is very big on using subtypes
08:24:21 <andythenorth> new authors are told it all works, and are given code to copy from
08:24:51 <Wolf01> Trainset should account for that, some fallback or whatever
08:24:55 <andythenorth> but the bug reports come to me, or JGR (because they assume it's a JGRPP issue)
08:25:11 <andythenorth> off-topic: the amount of crap JGRPP wades through
08:25:37 <andythenorth> endless support issues that are "grf" or "you are literally holding it wrong" or "that's just the way the base game works"
08:25:50 *** HerzogDeXtEr has joined #openttd
08:26:14 <andythenorth> hmm is Twitter today?
08:27:47 <peter1138> /* Need to open the file with fopen() to support non-ASCII on Windows. */
08:27:48 <Wolf01> The main problem is that a user like me won't be able to tell if the problem (missing train) is because of poor grf choices or an actual bug
08:28:00 <peter1138> Hmm, is there a non-ASCII content download I can try?
08:32:09 <peter1138> Web Blocking: Wed Blocking is a scenario...
08:32:13 <peter1138> So, er, web or wed?
08:32:43 <Wolf01> I would suggest some grf preset, but I'm too much biased toward my play style to do it, also I don't even know if I get broken features because I don't know how every piece of the puzzle should work
08:33:23 <andythenorth> I didn't know I'd broken RUKTS for a long time
08:34:07 <andythenorth> Every time I tried it, I assumed it just contained 5 multiple units as a design choice
08:34:22 <andythenorth> Nope, the rest of the set was missing :P
08:34:57 <peter1138> gzFile fin = gzopen(GetFullFilename(ci, true).c_str(), "rb");
08:35:22 <peter1138> Well, that works for me, I don't see any FILE handle that doesn't get closed.
08:35:50 <peter1138> But opening with fd should work of course.
08:36:14 <andythenorth> the whole train is selected, but the livery doesn't apply
08:36:46 <Wolf01> I would also say that multiple times I started games with poorly configured grfs, for example missing all vehicles because you need to enable them via parameters and I thought they will become available in future... grf sets need a manual sometimes
08:40:26 * andythenorth must now go write a powerpoint
08:40:39 <andythenorth> deadline is like 30 minutes from now
08:41:28 <Wolf01> At least is not 30 minutes in the past
08:56:47 <orudge> Looks like GitHub did catch up with all the queued actions
08:58:36 *** andythenorth has quit IRC (Quit: andythenorth)
08:58:49 <orudge> TrueBrain: is there something else that needs doing to enable nightly builds for OpenMSX, other than adding the workflows into the OpenMSX repository? :)
09:10:02 <LordAro> website stuff & workflow stuff, i would think?
09:34:57 <TrueBrain> orudge: what LordAro says, and S3 IAM user has to be set up (and the secrets needs to be set)
09:35:53 <LordAro> i imagine the the GHA outage last night had something to do with the massive Azure outage
09:36:14 <TrueBrain> possibly; but in general there were a lot of outages
09:37:27 <TrueBrain> owh, I completely forgot I wanted to rename the "max height" setting in New Game
09:37:32 <TrueBrain> into "Height Limit"
09:38:37 <TrueBrain> seems "resolution" is the only config setting that can crash OpenTTD
09:38:55 <TrueBrain> it bitflipped everything in a default configuration, and nothing else broke, it seems :P
09:39:16 <LordAro> well, nothing crashed :p
09:45:34 <TrueBrain> to make it more efficient, I should learn it a dict to try, instead of random bitflips and hope it rolls a valid word etc .. but that takes effort :P
09:46:08 <DorpsGek> [OpenTTD/workflows] orudge opened pull request #14: Add: [CDN] openmsx-nightlies is now also published on the CDN https://git.io/JqARE
09:47:18 <peter1138> 45 minute video meeting with my manager. Mostly waffle.
09:47:24 <peter1138> Sadly, not waffles.
09:48:51 <TrueBrain> I so do not miss those
09:49:44 <DorpsGek> [OpenTTD/workflows] TrueBrain approved pull request #14: Add: [CDN] openmsx-nightlies is now also published on the CDN https://git.io/JqA0B
09:56:05 <DorpsGek> [OpenTTD/workflows] orudge merged pull request #14: Add: [CDN] openmsx-nightlies is now also published on the CDN https://git.io/JqARE
09:57:17 <Timberwolf> I bumped up against the "this is how everybody is told to do it" thing with NewGRF once or twice.
09:58:14 <Timberwolf> Seems with roadtypes a few people define roadtype labels they don't actually use, and/or write their compatibility tables into the fallback list.
09:58:59 <Timberwolf> (Is that what the bit in NML is called where you say, "if DIRT is available use DIRT, otherwise use OFFR, otherwise if OFFR isn't available use ROAD"?)
09:59:48 <Timberwolf> So you get a situation where someone has no stagecoaches because they're using a roadtype which defines DIRT as a label, but never introduces a roadtype for it.
10:08:30 <Timberwolf> Or even better, tries to set up compatibility so dirt road vehicles can ford waterways, but if no dirt roads are available then anything defined for DIRT runs on water :)
10:08:45 <Timberwolf> (The roadtype waterways, not "real" water)
10:10:57 <TrueBrain> lol, fuzzing DoCommandP results in a lot of crashes :D
10:11:06 <TrueBrain> it is suppose to be network-safe :P
10:11:31 <spnda> i think im just about finished porting osie to C++17 and CMake... waste of time but idk
10:12:35 <TrueBrain> hmm ... MSU is spiking like a mofo
10:12:47 <TrueBrain> that might be me :D
10:12:58 <TrueBrain> sending 20k+ advertisements .. oops
10:13:18 <TrueBrain> but it shouldn't advertise ..
10:13:21 <TrueBrain> lets wait and see if it is me :P
10:18:23 <TrueBrain> well, /dev/null'd traffic on my end, I cannot be it anymore at least :)
10:19:39 <TrueBrain> orudge: am I right to assume you tested the "build" step in the OpenMSX publish workflow?
10:19:49 <orudge> TrueBrain: yes, on my own account - it worked
10:20:16 <TrueBrain> I expected nothing less :D
10:20:24 <TrueBrain> docs/ files are called readme.ptxt
10:20:32 <TrueBrain> but referenced as readme.txt in the bundle step
10:20:51 <orudge> It gets preprocessed by something or other
10:20:56 <TrueBrain> I love how many extensions our readmes have :D
10:21:07 <orudge> so you end up with a readme.txt
10:21:11 <TrueBrain> Secrets are installed, so it should be good to go from my end
10:21:44 <orudge> Great, we'll find out tonight :)
10:26:22 <TrueBrain> owh, it was me who was killing the MSU .. I forgot to give an empty configuration
10:34:40 <TrueBrain> hmm, it seems most crashes via DoCommandP happen because of an error message with an invalid string
10:43:04 <TrueBrain> I wonder if this means I can kick everyone from my own company :P
10:44:32 <TrueBrain> nah, too bad, it is only shown to the local client
10:44:38 <TrueBrain> so you can crash your own client; not that impressive :P
11:10:55 <DorpsGek> [OpenTTD/steam-data] TrueBrain opened pull request #3: Add: Python script to convert language files to what Steam expects https://git.io/JqAQd
11:11:35 <TrueBrain> out of the 7 crashes I have left over with DoCommandP, 5 are in CmdCompanyCtrl :P
11:11:44 <TrueBrain> it really does no validation, and only assumptions :D
11:13:00 <TrueBrain> by what I can tell they cannot be abused by a client, and it either only crashes the client, or the server had to initiate it (and crashes itself), all only with a modified client
11:13:04 <TrueBrain> but it is still fuzzing .. :D
11:15:14 <spnda> wasnt surprised that there was no webhook
11:15:31 <TrueBrain> there is; just no .dorpsgek.yml configuration
11:16:12 <spnda> but kinda what I was referring too, webhook wasnt hooked to that repository
11:16:47 <TrueBrain> well, we do not use webhooks; not sure what you mean?
11:17:38 <spnda> DorpsGek kinda is a webhook, no?
11:18:04 <TrueBrain> no; we use a GitHub Apps for it. And that is what I mean: if the repo reporting activity has no .dorpsgek.yml, DorpsGek is silent in this channel
11:18:12 <TrueBrain> it does rat it out to me, and that is why I can copy/paste it this easily
11:19:24 <TrueBrain> if you would be to add a .dorpsgek.yml in master, it will magically start working
11:19:54 <spnda> i mean that could be the second pr including the readme.txt fixup
11:20:06 <TrueBrain> Message: NOT_REACHED triggered at line 762 of OpenTTD/src/saveload/saveload.cpp <- fuzzing managed to hit that statement, funny :D
11:20:14 <TrueBrain> spnda: check any other repo how .dorpsgek.yml is introduced
11:21:04 <TrueBrain> ah, CmdChangeSetting triggers that
11:21:44 <TrueBrain> I think this means I can remote crash servers :P
11:22:23 <TrueBrain> will have to proof-of-concept that, ofc
11:24:48 <peter1138> Is this our predesigned security flaw so that we can have a security release to get 1.11 on Debian 10? ;-)
11:25:12 <TrueBrain> well, we learnt they just backport the fix and still don't update :P
11:26:23 <TrueBrain> it crashes inside the DC_EXEC block
11:26:33 <TrueBrain> so if I am lucky, it even distributes it to all clients first
11:27:08 <TrueBrain> sadly "p1" is optimized out, so I cannot see what setting it is setting via gdb :P
11:27:45 <TrueBrain> "locale.digit_group_separator"
11:29:25 <spnda> what the hell is that docs/osie.1 file?
11:39:38 <TrueBrain> ah, the command that can crash the game is marked as "server-only"; I assumed DoCommandP would handle that, but one layer higher does
11:39:42 <TrueBrain> so the server can only crash clients
11:53:19 <peter1138> Is it sleep time yet?
11:57:19 <TrueBrain> but lunchtime it is!
12:10:32 <DorpsGek> [OpenTTD/steam-data] LordAro commented on pull request #3: Add: Python script to convert language files to what Steam expects https://git.io/JqxTm
12:49:38 *** Gustavo6046 has quit IRC (Ping timeout: 480 seconds)
12:49:55 <spnda> seems my osie pr works. Tested on win and ubuntu 20.04
13:02:52 <DorpsGek> [OpenTTD/steam-data] TrueBrain updated pull request #3: Add: Python script to convert language files to what Steam expects https://git.io/JqAQd
13:07:44 *** Gustavo6046 has joined #openttd
13:25:58 <spnda> just wanna mention, its up for review
13:32:24 *** snail_UES_ has joined #openttd
13:32:53 *** andythenorth has joined #openttd
13:41:38 <andythenorth> ok time for lunch
13:41:40 <andythenorth> cheese on toast?
13:42:15 * LordAro had a cheese & ham sandwich
13:43:27 <andythenorth> cheese + ham toasty
13:43:59 <peter1138> spnda, do we not already use CMake? o_O
13:44:14 <peter1138> Oh. This is... not OpenTTD.
13:44:22 <LordAro> is osie even necessary anymore, now that bananas2 is a thing?
13:44:34 <peter1138> "merge 1 commit into OpenTTD:master" < annoying lack of information from Github...
13:44:49 <peter1138> OpenTTD is... the account, not the repo o_O
13:44:54 <spnda> LordAro: But isn't that just a web api.
13:45:19 <spnda> peter1138: Yes agreed, they should show the repo name, the account is really confusing
13:45:45 <spnda> and I mean, it definitely isn't necessary but keeping stuff running and up-to-date is still good practice imo
13:45:47 <TrueBrain> LordAro: osie extracts the metadata from OpenTTD screenshots, not?
13:46:02 <LordAro> oh, i'm misremembering tools
13:46:02 <spnda> shows you the companies, newgrfs and game version and a bit more
13:46:15 <LordAro> what was the bananas upload tool?
13:47:40 <spnda> TrueBrain: Btw, the BaNaNaS API exist yet? (If so, any docs?)
13:48:03 <TrueBrain> "The API is documented on SwaggerHub."
13:48:22 <spnda> swaggerhub? lmao what a name
13:49:31 <spnda> shouldn't that just be called bananas-cli
13:49:32 * LordAro wonders if anyone checked whether the opengl-generated screenshots also contain the metadata correctly
13:50:06 <TrueBrain> spnda: debating names is something I gave up long ago; we have several frontends, one is a CLI
13:52:29 <spnda> LordAro: Idk if anyone checked, but it definitely works
13:53:23 <andythenorth> when you want salad
13:53:31 <andythenorth> then you find you already made some
14:02:40 *** Compu has quit IRC (Remote host closed the connection)
14:17:14 <DorpsGek> [OpenTTD/OpenMSX] orudge closed issue #12: Big Man Boogie is a temporary replacement for Keep on rolling (bring it back, maybe?) https://git.io/JqFDu
14:28:16 *** Jan has quit IRC (Remote host closed the connection)
14:30:23 <Klaas> When I try to compile OpenTTD on my raspberry pi (following the steps on the site) I encounter an error: 'collect2: error: ld returned 1 exit status'. What could be the problem? I already have compiled an older version successfully some time ago
14:30:54 <LordAro> we'd need the complete output
14:31:04 <LordAro> that just says "there has been an error"
14:39:20 *** jgx has quit IRC (Remote host closed the connection)
14:42:33 *** Flygon has quit IRC (Quit: A toaster's basically a soldering iron designed to toast bread)
14:56:36 <Klaas> [ 1%] Built target regression_files [ 2%] Built target baseset_files -- Version string: 20210315-master-g5c73f93361 -- Generating rev.cpp -- Generating CPackProperties.cmake -- Generating Doxyfile -- Generating Doxyfile_AI -- Generating Doxyfile_GS [ 2%] Built target find_version [ 3%] Built target script_window [ 13%] Built target script_ai [ 13%] Built target script_ai_includes [ 23%] Built target script_game [ 2
14:56:50 *** Klaas has quit IRC (Quit: Page closed)
14:58:05 <Klaas> I get these errors in my log during compiling: openttd/src/industry_cmd.cpp:2218: undefined reference to `Pool<Industry, unsigned short, 64u, 64000u, (PoolType)1, false, true>::MAX_SIZE' Same for town
14:59:05 *** Wormnest has joined #openttd
15:00:57 <LordAro> Klaas: the complete output, please
15:01:03 <LordAro> put it on pastebin.com or somewhere similar
15:01:10 *** andythenorth has left #openttd
15:01:36 <LordAro> including the cmake output, if possible
15:04:43 <LordAro> i'm wondering if the compiler is too old, what is the version of gcc ?
15:06:24 <LordAro> that should be plenty new enough
15:09:10 <Klaas> I'm doing a fresh retry now. But I don't believe that will help, since I already tried on multiple devices
15:12:43 <TrueBrain> what you can try, and this is a long shot: open src/core/pool_type.hpp, around like 85 you find "static const size_t MAX_SIZE = Tmax_size"
15:12:50 <TrueBrain> change "const" into "constexpr", like: "static constexpr size_t MAX_SIZE = Tmax_size"
15:12:54 <TrueBrain> and give that a try
15:18:46 *** sla_ro|master has quit IRC ()
15:50:21 <_dp_> hmm, I spammed openttd server with commands and everything was fine but now it refuses to shutdown...
16:00:44 *** Klaas has quit IRC (Remote host closed the connection)
16:00:54 <TrueBrain> he left without telling if that helped :'(
16:12:57 *** Progman has joined #openttd
16:27:21 *** frosch123 has joined #openttd
16:38:56 <spnda> I just noticed that the NML-CAT2.0 : WATERCRAFT GRF has "42069" as its version
16:40:10 *** sla_ro|master has joined #openttd
16:43:09 <frosch123> can you compute the age of the author from that?
16:44:02 <frosch123> TrueBrain: gcc 8.3.0 is in debian stable, i use that all the time
16:44:57 <TrueBrain> frosch123: not sure why I wanted to know that :D
16:47:27 <LordAro> frosch123: i'm the one who asked about compiler age :p
16:47:37 <LordAro> and that's just because they didn't specify what distro they were running
16:47:44 <LordAro> and raspberry pi is a bit of a crapshoot
16:50:14 <TrueBrain> the reason I think that "const" -> "constexpr" might fix it, as we had a very similar issue with std::min and "static const" in classes
16:53:38 <LordAro> could be something weird about the arm compiler
16:54:08 <DorpsGek> [OpenTTD/workflows] frosch123 opened pull request #15: Add: sync actions for eints projects OpenGFX, OpenSFX, OpenMSX and steam-data. https://git.io/Jqpy5
16:54:40 <frosch123> i updated the openttd-github branch in eints
16:54:53 <frosch123> but my changes are somewhat incompatible
16:55:06 <frosch123> so we have to tag eints and the workflows together
16:55:14 <frosch123> otherwise the GHA will fail
16:57:43 <TrueBrain> I will look at it after dinner :)
16:58:10 <DorpsGek> [OpenTTD/steam-data] frosch123 approved pull request #3: Add: Python script to convert language files to what Steam expects https://git.io/Jqp98
17:04:44 *** PlMysza has joined #openttd
17:26:37 *** jottyfan has joined #openttd
17:32:41 <spnda> TrueBrain: Thing with the bananas-api. A ISO 8601 timestamp doesn't require a timezone, but the API is still like "REEEE NO TIMESTAMP"
17:33:26 <spnda> and eventhough my thingy is sending "1970-01-01T01:00:00.000"
17:47:55 <LordAro> spnda: you know where bugs go ;)
17:50:08 <spnda> huh there are 2 lines in the bananas api code specifically erroring out without timezone information...
17:50:35 <TrueBrain> so the solution is simple: add a timezone :)
17:52:13 <spnda> I don't want to change any output of a default function in a popular programming language. There's a reason for it doing it and it's best to keep to the iso spec :)
17:53:34 <LordAro> "If no UTC relation information is given with a time representation, the time is assumed to be in local time. "
17:53:39 <LordAro> this tends to cause confusion
17:54:04 <TrueBrain> and that is why timezone is enforced
17:54:32 <TrueBrain> it is one of those ... "lovely" issues when you talk about serialization protocols :D
17:55:00 <spnda> i'm sure google thought of something when they decided to not append the timezone (atleast local) to the toIso8601String() function, but ok
17:55:23 <TrueBrain> either way, if your popular programming language supports ISO 8601, it should be able to add timezones :)
17:55:53 <TrueBrain> in Python it is a bit of a disaster
17:55:57 <TrueBrain> but for completely different reasons
17:56:18 <TrueBrain> (and the reason the servers runs in UTC, and a Z is appended in some places)
17:56:32 <TrueBrain> so I hope your popular programming language is better than Python :D
17:57:07 <spnda> Personally, I'd prefer most languages over Python. Especially this, Dart
17:57:16 *** Gustavo6046 has joined #openttd
17:57:44 <TrueBrain> luckily enough we live in a world where you have the freedom to pick :)
17:58:14 <LordAro> suggests that it'll output the 'Z' if the utc() constructor is used
17:58:14 <spnda> Yes, prints "1970-01-01T01:00:00.000"
17:58:27 <spnda> when used as since.toUtc().toIso8601String()
17:58:44 <TrueBrain> you got to love when documentation disagrees with the implementation :D
17:59:11 <TrueBrain> very curious, even more as it is explicitly defined to add a Z if it is in UTC
17:59:26 <spnda> but yes, UTC has that Z thingy for some reason
18:00:46 <TrueBrain> well, "for some reason" is by spec :D It is not so much an accident :)
18:00:52 <TrueBrain> sadly, there is no ISO format that disallows local time
18:00:54 <TrueBrain> that would be perfect
18:01:07 <spnda> but tbh, this cli is coming together quite nicely
18:01:23 <TrueBrain> I am happy the API documentation works well enough :)
18:01:32 <TrueBrain> if you want to test it, please do test it against staging
18:01:35 <TrueBrain> you can go nuts on staging
18:01:49 <TrueBrain> (or better, test it against a local API, but that is not for everyone)
18:01:52 <spnda> would still love a little note there "only accepts dates with timezone info"
18:02:14 <TrueBrain> the API return value was clear enough that you immediately picked it up, not?
18:02:34 <spnda> would make it 2 seconds quicker to implement :)
18:02:41 *** Gustavo6046 has joined #openttd
18:03:03 <LordAro> TrueBrain: internet suggests RFC 3339 requires a timezone, but i'm not sure
18:03:46 <TrueBrain> that would be interesting :D
18:03:53 <TrueBrain> even more if marsmallow would support it
18:03:57 <TrueBrain> but that might be too much to ask :P
18:05:24 <TrueBrain> "If the time in UTC is known, but the offset to local time is unknown, this can be represented with an offset of "-00:00"."
18:05:26 <TrueBrain> you got to love hacks :D
18:05:48 <TrueBrain> "For example, the date format "10/11/1996" is completely unsuitable for global interchange because it is interpreted differently in different countries. "
18:07:40 <TrueBrain> LordAro: bah, RFC 3339 made the silly mistake to allow a space for "T"
18:07:48 <TrueBrain> so an ISO 8601 could fail parsing a valid RFC 3339
18:08:01 <spnda> TrueBrain: /package/{content-type}/{unique-id}/{upload-date}, does that get the nearest package from that date too?
18:08:07 <LordAro> doesn't ISO8601 also allow a space?
18:08:15 <TrueBrain> LordAro: it seems it does not
18:08:37 <TrueBrain> an API should not do fuzzy matching
18:08:42 <TrueBrain> an entry either exists or it doesn't :)
18:08:54 <TrueBrain> there are search endpoints for fuzzy matching
18:09:16 <TrueBrain> do I dare to ask why you would want that? :P
18:10:01 <DorpsGek> [OpenTTD/workflows] TrueBrain approved pull request #15: Add: sync actions for eints projects OpenGFX, OpenSFX, OpenMSX and steam-data. https://git.io/JqhtA
18:10:50 <TrueBrain> frosch123: looks good; very nice :D
18:10:57 <milek7> iso8601 is hideously complex
18:11:35 <spnda> TrueBrain: Idk, was really just wondering. But also thought that maybe one could pass 1970-00-00 in and it would get the first ever version
18:11:50 <TrueBrain> there are other endpoints for that :)
18:12:04 <TrueBrain> well, "first" is not really interesting, I guess
18:12:08 <TrueBrain> "last" is more interesting
18:13:03 <frosch123> TrueBrain: merge PR, tag eints?
18:13:13 <TrueBrain> frosch123: go for it :)
18:13:36 <TrueBrain> you have a window of ~30 minutes
18:13:39 <TrueBrain> so that should be plenty :P
18:13:52 <TrueBrain> (we are in between 2 eints runs, to find the CET / CEST timezone :P)
18:14:02 <TrueBrain> so silly GHA doesn't allow timezones for the schedules :D
18:14:26 <TrueBrain> spnda: if you are asking if that is implemented already, no, it is not :)
18:16:52 <frosch123> oh, workflows seem to use the eints branch, so the tag does not even matter
18:17:07 <DorpsGek> [OpenTTD/workflows] frosch123 merged pull request #15: Add: sync actions for eints projects OpenGFX, OpenSFX, OpenMSX and steam-data. https://git.io/Jqpy5
18:17:38 <TrueBrain> frosch123: ah, yes, makes sense I guess .. doing a checkout of "latest tag" is not something git knows how to do with a single command :P
18:17:51 <TrueBrain> owh well, tagging it makes it "official" :D
18:18:40 <frosch123> yeah, but since i already updated the branch, GHA would have failed today, if we delayed workflows for some reason
18:18:56 <TrueBrain> well, a good thing we did this before the next run :D
18:19:00 <TrueBrain> sometimes you win by luck ;)
18:19:19 <frosch123> yep, let's see what it commits today
18:19:25 <frosch123> and hope it does not commit secrets :p
18:19:51 <TrueBrain> well, I have to rotate some secrets anyway, so ... :P
18:20:06 <TrueBrain> ugh, I have to clean up all the capsule images for Steam and put them in the repo
18:20:11 <TrueBrain> I really do not want to :P
18:20:32 <frosch123> sounds like i do not want to know how many there are
18:21:04 <TrueBrain> well, 12, they added a new one
18:21:10 <TrueBrain> all different formats
18:21:13 <TrueBrain> all different requirements
18:21:40 <TrueBrain> it is like they invested in making sure UI people will have a job once you list yourself on Steam
18:32:13 *** gelignite has joined #openttd
18:32:56 <TrueBrain> title screen voting is going well by the looks
18:33:20 <frosch123> thanks for the reminder :)
18:33:41 <TrueBrain> was it subtle enough? :D
18:34:01 <TrueBrain> I still love that I managed to make animated gifs :D Not sure it helps, but damn, it looks nice :D
18:34:47 <frosch123> in the past years i always downloaded the saves. the screenshots were mostly not useful
18:34:59 <frosch123> let's try the gifs :)
18:35:21 <TrueBrain> when watching them I sometimes forget I cannot click the menu :P
18:35:35 <TrueBrain> next time, specially build Emscripten version? :D
18:35:54 <frosch123> sometimes they have eastereggs. crashing vehicles and such
18:36:34 <TrueBrain> I really do not know how people make these title games .. there is so much to see and look at
18:37:43 <frosch123> ok, good, there are only upvotes. i was worried people would also downvote
18:38:17 <TrueBrain> I do have a downvote button, but it seems most people do not :P
18:38:18 <spnda> I didn't even see a downvote button on github discussions.. hmm
18:38:39 <TrueBrain> I think if you are an Owner or Member you also are allowed to downvote
18:38:41 <TrueBrain> otherwise you are not
18:38:44 <TrueBrain> which is really odd
18:51:15 <frosch123> oh damn, i failed to understand the voting mechanism
18:51:22 <frosch123> i thought the emohis were the votes
18:51:32 <TrueBrain> more people seem to have that issue, yes
18:51:42 <TrueBrain> not sure it matters
18:51:54 <LordAro> maybe bold the bit about voting?
18:52:07 <TrueBrain> you really think people will read? :D
18:52:28 <TrueBrain> but feel free; you should also be able to edit it :)
18:53:14 <frosch123> the votes and the likes have slightly different numbers
18:53:41 <spnda> ok now this is a slight issue. I changed my GitHub name and i'm still shown as "PNDA" on bananas
18:54:27 <frosch123> don't worry, access is managed via your numeric id
18:54:41 <frosch123> the name will only update for new packages
18:54:57 <frosch123> everyone who migrated from old openttd accounts is also listed with their openttd account
18:56:12 <frosch123> spnda: it's not "obviously" :) we discussed whether it is necessary. and "being correct" won only slightly over "noone is so insane to change their name, when it is in every git repository" :p
18:56:45 <spnda> I would always go by uuid of a user, will never change
18:57:00 <frosch123> try @mention then :p
18:57:09 *** andythenorth has joined #openttd
18:59:11 <TrueBrain> export to staging worked ..
18:59:20 <frosch123> i am looking at it :p
19:01:15 <frosch123> let's say, it's not fast :p
19:01:38 <TrueBrain> memory-wise, it is not noticeable that there are more projects
19:01:49 <TrueBrain> you do see based on the CPU when people are translating
19:01:54 <TrueBrain> but that has always been the case
19:02:20 <DorpsGek> - Update: Translations from eints (by translators)
19:02:49 <TrueBrain> for the other repos DorpsGek won't report the commits btw
19:02:59 <TrueBrain> as you might have noticed yesterday
19:03:01 <TrueBrain> not sure we want to :P
19:03:30 <LordAro> i think we'll cope without them
19:03:35 <frosch123> yes, i figured people would complain about 5 notices in a row
19:03:44 <TrueBrain> people complain either way :P
19:03:55 <TrueBrain> okay, OpenSFX takes longer than I would expect
19:04:12 <TrueBrain> I know OpenTTD takes a long time, but, lot of strings etc .. sure
19:04:15 <TrueBrain> but why the others are this slow?
19:04:44 <TrueBrain> owh, sorry, nevermind
19:04:49 <TrueBrain> it was uploading them back to OpenTTD
19:05:10 <TrueBrain> so the step it is at now should be quicker, I hope :D
19:05:17 <andythenorth> pls provide lolz
19:05:44 <TrueBrain> frosch123: aawwwhhh
19:05:55 <andythenorth> game requires that link
19:06:03 <TrueBrain> it doesn't like an empty push, I guess?
19:06:07 <frosch123> branch protection :p
19:06:19 <frosch123> i made dorpsgek admin
19:06:25 <frosch123> but i guess there are more settings
19:06:33 <TrueBrain> yeah, you have to unselect the "including administrators"
19:06:35 <TrueBrain> for branch protection
19:08:22 <TrueBrain> I guess retriggering is fine
19:08:25 <frosch123> wait for nightly, then start again?
19:08:32 <TrueBrain> worst case, a new translator commit for OpenTTD
19:08:49 <TrueBrain> nah, just restart now, what is the worst that can happen :)
19:09:49 <frosch123> restarted the movie
19:10:08 <TrueBrain> checkouts are speedy :D
19:10:12 <TrueBrain> GitHub did that nicely
19:10:20 <TrueBrain> I mean, 7s for OpenTTD to checkout
19:10:22 <TrueBrain> that is .. impressive :D
19:10:39 <frosch123> i recall there was a issue. gha does a sparse checkout, then our script does a full pull
19:10:58 <TrueBrain> btw, even if "master" changes during a nightly, it doesn't matter; sources are checked out in 1 place, and send as zips to all the other targets
19:13:21 <TrueBrain> frosch123: for pull requests, we pull in more commits till we are at "master" :)
19:13:29 <TrueBrain> that was the biggest change in checkout times we made :)
19:13:49 <TrueBrain> so we shouldn't be doing full clones anywhere anymore, if I am not mistaken :D
19:14:40 <DorpsGek> - Update: Translations from eints (by translators)
19:14:44 <TrueBrain> okay, that is funny :D
19:14:53 <TrueBrain> what are the odds someone is translating exactly now :D
19:15:12 <frosch123> and what lag are they experiencing while the script runs?
19:16:05 <TrueBrain> frosch123: no clue :)
19:17:03 <frosch123> hmm, there are actually some issues with the script as it runs now
19:17:17 <frosch123> it does not expect the checkout to be up-to-date already
19:18:27 <spnda> its gonna be fun getting my cli to be able to upload files right now... cba figuring out how to do that github auth thingy
19:19:59 <frosch123> copy it from bananas-frontend-cli?
19:19:59 <TrueBrain> OpenGFX and OpenSFX look good
19:20:25 *** virtualrandomnumber has joined #openttd
19:20:40 <frosch123> yay, adding new translations worked as well
19:20:45 *** virtualrandomnumber has quit IRC ()
19:21:01 <TrueBrain> OpenMSX is taking its sweet time ..
19:21:21 <TrueBrain> there we go, looks good too
19:21:37 <spnda> frosch123: Not using Python, so eh, but eventually yes
19:22:09 <frosch123> Wuzzy: why "ersatzset"?
19:22:29 <Wuzzy> because the English is "replacement set"
19:22:36 <TrueBrain> nice work frosch123 , looks good :D
19:22:48 <TrueBrain> and you didn't even need my help for the workflow :P
19:23:10 <frosch123> i did, you pointed to the lines
19:23:17 <spnda> TrueBrain: Is Authentication using the auth http header with a Bearer access token?
19:23:17 <Wuzzy> if you think the "Ersatz" / "replacement" is stupid, fix OpenGFX/OpenSFX/OpenMSX first befor touching translations, thanks
19:23:25 <DorpsGek> [OpenTTD/steam-data] TrueBrain merged pull request #3: Add: Python script to convert language files to what Steam expects https://git.io/JqAQd
19:23:56 <TrueBrain> spnda: authentication is an OAuth2 flow via GitHub
19:24:18 <spnda> TrueBrain: Yes but after you've authenticated
19:24:54 <spnda> idk i don't know much about oauth2
19:25:55 <Wuzzy> frosch123: maybe its better to remove any mentining of the word "replacement set" in all 3 OpenX basesets. they are perfeectly valid in their own right
19:26:47 <Wuzzy> unfortunately, if we do that, we have to reset ALL translations of the basesets
19:27:14 <Wuzzy> hopefully no longer a big deal now that they are on eints
19:27:35 <frosch123> i don't understand why eints works
19:28:02 <frosch123> it always skips the upload
19:29:03 <spnda> oh wait I can easily integrate the GitHub OAuth2 stuff. Dart literally has a built-in library for oauth2 applications
19:30:23 <TrueBrain> w00p, 2 more languages translated for Steam Store Page
19:30:28 <frosch123> ah, it only does the file-changed-check in the other case. all good :p
19:30:33 <TrueBrain> spnda: I wouldn't know; what ever the server returns, so a cookie I would guess
19:31:04 <frosch123> this script has too much devzone history in it :p
19:31:12 <TrueBrain> frosch123: clean it up! :D
19:31:24 <frosch123> i guess it reached legacy code level :) it works, don't touch it
19:31:28 <TrueBrain> pretty sure the pull it does, or what-ever causes it to pull tags, is not needed anymore? :D
19:31:50 <TrueBrain> did you notice the lack of emails on info@
19:31:53 <TrueBrain> would it really have worked?
19:32:39 <frosch123> there was enough noise from other gh notifications, i wouldn't notice if info@ was missing
19:33:03 <frosch123> oh, more bananas migrations
20:04:19 <DorpsGek> [OpenTTD/BaNaNaS] frosch123 opened pull request #86: Change: add 'dandan' as co-author to all Japan sets. https://git.io/JqhX3
20:17:04 <TrueBrain> frosch123: unusual PR
20:18:02 <DorpsGek> [OpenTTD/BaNaNaS] TrueBrain approved pull request #86: Change: add 'dandan' as co-author to all Japan sets. https://git.io/JqhDm
20:18:07 <TrueBrain> well, knowing you, you validated it etc
20:18:10 <frosch123> they shared access on devzone
20:18:33 <DorpsGek> [OpenTTD/BaNaNaS] TrueBrain merged pull request #86: Change: add 'dandan' as co-author to all Japan sets. https://git.io/JqhX3
20:18:46 <frosch123> i just don't know how you get involved with both japanese and swiss newgrf :)
20:19:23 <TrueBrain> being very good at your work, I assume :P
20:29:26 *** jottyfan has joined #openttd
20:43:52 <michi_cc> Hmm, I tried to approve osie#1, but apparently I'm not good enough :)
20:44:22 <andythenorth> oops why did that highlight TB?
20:44:29 * andythenorth should go to sleep
20:45:17 <TrueBrain> Love you too andythenorth :p
20:46:24 <andythenorth> so releasing grfs is too much faff
20:47:20 <andythenorth> I have to build, do bananas, then build grf.farm website, log in to AWS, upload docs blah blah
20:47:40 <andythenorth> should I write an AWS sync client, or just download one of the mac gui sync apps?
20:47:45 <LordAro> andythenorth: you should be careful not to highlight TrueBrain needlessly
20:47:57 <andythenorth> LordAro what could go wrong?
20:49:43 <spnda> I guess the dorpsgek file instantly worked lol
20:50:02 <glx> yeah if it works it's immediate
20:52:36 <TrueBrain> spnda: if I build things, I build them good :p no half-assed shit ;)
20:53:28 <spnda> isn't that *over-engineering*
20:54:43 <TrueBrain> And now complaints it works too well? How silly ....
20:55:02 *** jottyfan has quit IRC (Quit: jottyfan)
20:55:22 <TrueBrain> This will be a long evening :D
21:17:48 <spnda> ah yes, its a firefox issue
21:22:32 <andythenorth> oof it will be livestream time soon
21:22:32 <LordAro> sounds like the sort of thing that's getting ignored as part of the shell
21:22:41 <LordAro> & forks the process, for instance
21:22:46 <LordAro> need to quote it properly
21:22:49 <andythenorth> this is a very busy work time of year :|
21:29:51 <frosch123> do we have a list of guests?
21:31:22 <TrueBrain> Didn't you delegate it andythenorth?
21:31:32 <TrueBrain> Sounded like a good move :p
21:31:33 <andythenorth> I would have to check
21:31:44 <andythenorth> delegating only works when understood
21:33:32 <andythenorth> 2TallTyler and Timberwolf are both up for being in it
21:33:43 <andythenorth> I have NFI about technical matters
21:33:55 <andythenorth> but I do have a ring light, so eh, the game is on
21:34:17 <Timberwolf> Heh. I have a camera with one built in.
21:34:34 <Timberwolf> I don't think it's quite powerful enough to offset bad room lighting.
21:34:49 <andythenorth> what are we talking about?
21:34:52 <Timberwolf> And having a powerful light on the thing you're supposed to be looking at isn't the most useful feature.
21:34:59 <andythenorth> was frosch123 writing a poem about newgrf and GS?
21:35:54 <LordAro> i thought frosch123 was doing something with hovering buildings
21:35:59 <andythenorth> Timberwolf are we talking about designing mods? making mods? The mod spec?
21:36:02 <andythenorth> or all the mistakes we made?
21:36:26 <Timberwolf> Heh, I've no idea.
21:37:10 <frosch123> LordAro: did you make that up? or should i grep my logs what that was about? sounds interesting
21:37:17 <Timberwolf> The problem I have is I try stuff on a completely uninformed basis and half the time it accidentally works out.
21:37:39 <spnda> wait is there a stream tonight?
21:37:57 <andythenorth> it takes weeks to remember which button to press
21:38:04 *** jottyfan has joined #openttd
21:38:06 <LordAro> frosch123: i'm sure someone mentioned something about that involving turning buildings into disaster vehicles
21:38:13 <andythenorth> maybe Weds 31st, or maybe the weekend near then
21:38:57 <andythenorth> LordAro the idea was towns that move around
21:39:22 <andythenorth> traditionally this is the point where someone says "but GS"
21:39:31 <andythenorth> then we discuss why almost nobody makes GS :)
21:40:04 <andythenorth> towns could migrate to the seaside for July - August
21:41:48 <TrueBrain> andythenorth: pick a date, tell me who is on it, and I will make sure the stream works technically.
21:42:10 <Timberwolf> The midweeks are slightly better for me, but anything should work.
21:42:30 <andythenorth> I am favouring April 1
21:42:44 <andythenorth> I have a shitload of work to finish Mar 31st
21:43:09 <TrueBrain> 1st of April in the evening I am a bit bust with other OpenTTD related matters .. might be hard to multitask those :D
21:43:56 <andythenorth> bank holiday in the UK, traditionally people travel away
21:43:57 <frosch123> some ottd release was made on the 25k party
21:44:02 *** Progman has quit IRC (Remote host closed the connection)
21:44:02 <LordAro> and not mention release at all
21:44:47 <TrueBrain> Well, if we prepare everything the 31th, we could do it the 1st
21:44:50 <andythenorth> I like April 2nd
21:44:52 *** jottyfan has quit IRC (Quit: jottyfan)
21:45:33 <andythenorth> gives me a break after some work stuff
21:45:48 <TrueBrain> Just Steam scares me a bit :p
21:45:59 <TrueBrain> Especially stability of service :D
21:46:22 <Timberwolf> "Good evening, we are coming to you live from after The Event."
21:46:29 <Timberwolf> "Remember, REMAIN INDOORS."
21:46:38 <andythenorth> livestream a high traffic event
21:46:56 <andythenorth> nothing like a bit of attention to make sysadmin interesting
21:46:58 <TrueBrain> Show http traffic Live :p
21:47:01 * andythenorth been there, done that
21:47:15 <andythenorth> 'the white house is on the phone again, when will the site be up?'
21:48:30 <TrueBrain> I say we let frosch123 pick the date :p
21:48:57 <frosch123> i am fine with all 3 mentioned dates
21:49:24 <frosch123> if you need a tiebeaker, we go for the 2nd :)
21:50:01 <Timberwolf> I'm fine with all of those.
21:50:08 <andythenorth> I have put 2nd in my diary
21:50:13 <andythenorth> so that's OFFICIAL
21:50:17 <LordAro> Timberwolf: Do not think about The Event
21:50:21 <andythenorth> we should get a shared calendar :P
21:51:43 <TrueBrain> I read 3 people here .. any other guest?
21:52:23 <andythenorth> I am at 4 so far, but we might find more
21:52:45 <TrueBrain> That is not helpful andythenorth .. looking for names :p
21:53:24 <andythenorth> I have decided it's only people beginning with T
21:53:38 <andythenorth> Tyler, Timberwolf...Truebrain
21:53:59 <TrueBrain> I am not :p I just make sure you can :D
21:54:11 <TrueBrain> So frosch, Andy, Tyler and timber?
21:54:17 <andythenorth> ok, you break my alliteration :(
21:54:36 <TrueBrain> What are we calling it .. "graphics for the weak"?
21:54:39 <andythenorth> who else has views on mod spec, modding, modding stupidity
21:54:50 <andythenorth> glx said no before?
21:54:57 <frosch123> TrueBrain: "disagreenig"
21:55:08 *** Samu has quit IRC (Quit: Leaving)
21:55:10 <TrueBrain> 4 people is plenty, to be clear. 5 is max, I think :)
21:55:20 <glx> andythenorth: it's still no
21:55:22 <TrueBrain> frosch123: haha, yes!
21:55:50 <glx> it's already hard for me to watch streams when 480p is not available ;)
21:55:52 *** HerzogDeXtEr has quit IRC (Read error: Connection reset by peer)
21:56:35 <andythenorth> what about dorpsgek?
21:56:55 <TrueBrain> It can make an appearance, sure :l
21:57:16 <andythenorth> get it to say 'new graph'
21:57:26 *** jottyfan has joined #openttd
21:57:29 <TrueBrain> Going to make it sing
21:57:30 <andythenorth> as those are the official pronunciations now
21:57:56 <glx> too bad DorpsGek no longer "say" things ;)
21:58:04 <Timberwolf> This could be turned into a north-south thing. "groof" vs. "graaaf"
21:58:13 <TrueBrain> Okay, that is all settled. We need to test everything beforehand, but I will get back to that :)
21:58:19 <andythenorth> anything but G R F
22:00:29 <TrueBrain> andythenorth: do you tell Tyler? :p
22:00:57 <andythenorth> timezone delay though
22:01:20 <TrueBrain> Can be fun with upcoming summertime :p
22:02:32 <frosch123> what timezone is tyler? i remember australian
22:02:39 <frosch123> which measn 2000 CEST is a bit early :p
22:03:01 <andythenorth> tyler is US, about 5 hours from me
22:03:18 <frosch123> oh, so east coast. then it's okay
22:03:53 <spnda> I wish github gave more info other than just 404-ing on invalid authorize requests....
22:04:24 * andythenorth wonders what happened to pikka
22:04:30 <andythenorth> absent from Twitter etc
22:06:24 <frosch123> Last active:14 Mar 2021 14:12 <- reading forums
22:08:47 <frosch123> who build the website at this time?
22:10:08 <frosch123> did that job run for 12 hours?
22:11:11 <frosch123> info@ just got a mail that the website deployment failed 12 hours ago, or so
22:12:18 <TrueBrain> More like 30 minutes ago, lol
22:12:36 <TrueBrain> One of the Open**X overlapped with another
22:12:51 <TrueBrain> Schedule was a bad boy
22:12:55 <TrueBrain> Guess we need to put them even further apart
22:13:43 <TrueBrain> Something to fix tomorrow
22:13:49 <frosch123> ah, so baseset nightlies build at this time
22:14:03 <frosch123> i didn't know they had nightlies
22:14:14 *** Kitrana has joined #openttd
22:14:15 *** jottyfan has quit IRC (Quit: jottyfan)
22:14:26 <TrueBrain> orudge and glx have been doing that
22:14:33 <TrueBrain> OpenMSX happened today
22:14:47 <TrueBrain> Eints is causing them all to build :D
22:15:36 <frosch123> i forward the blame to the translators translating things :p
22:15:45 <DorpsGek> [OpenTTD/OpenTTD] JGRennison opened issue #8871: SDL2/Linux: OpenGL: Window sometimes filled with unitialised pixels at startup https://git.io/Jqjv9
22:16:15 <TrueBrain> Owh, yeah, JGR has a good point, happens on all OSes
22:16:17 <glx> hmm maybe only one of them should rebuild website on eints trigger
22:16:25 <frosch123> yes, i also have that, but i did not assume it was fixable
22:16:35 <TrueBrain> At least, also macOS and win32 reported it
22:16:57 <TrueBrain> glx: no, that is far more complex to build :)
22:17:11 <TrueBrain> Just moving them more apart should be fine :)
22:17:53 <andythenorth> I tried to screenshot on macos, but it's not reliable
22:18:07 <andythenorth> can't repro currently, but usually shows as mixed red/black pixels
22:18:49 <TrueBrain> Maybe michi_cc has some magic up his sleeve :D
22:20:38 <spnda> wait stupid me... I have to auth with the BaNaNaS GitHub application's client identifier don't I? but where would I find such thing
22:20:59 <spnda> ah stupid me again... in the url
22:26:20 *** frosch123 has quit IRC (Quit: be yourself, except: if you have the opportunity to be a unicorn, then be a unicorn)
22:27:32 *** andythenorth has quit IRC (Quit: andythenorth)
22:29:01 *** didac has quit IRC (Remote host closed the connection)
22:30:59 <spnda> TrueBrain: I think that the bananas API is kinda useless if I'm not mistaken. From what I understand I need to get authenticated with the official BaNaNaS github app. Though if I do that, I cannot specify a redirect uri and it will always redirect to the bananas website, making it impossible to catch the oauth2 credentials.
22:37:57 <spnda> hmm seems like it actually just doesn't like my redirect_uri as a query
22:43:37 <_dp_> on spectating client some wagons didn't attach
22:43:52 <_dp_> nothing fancy in the command log, just buying a train
23:15:09 <LordAro> there's a 1.11 change i could suggest is related, but nothing 1.10.3
23:18:29 <_dp_> is there any way to check ids of the stuff in the game btw?
23:36:16 <_dp_> ok, I kinda know why it didn't attach, that wagon has different id
23:36:26 <_dp_> but why the fuck didn't it desync right there and then
23:38:12 <TrueBrain> Remember the patch in JGRPP that makes desyncs happen a lot earlier? :D
23:38:25 <TrueBrain> We should cherrypick that really
23:38:32 *** nielsm has quit IRC (Ping timeout: 480 seconds)
23:39:07 <_dp_> well, yeah, but it should've desynced in vanilla anyway if I understand it right
23:39:15 <_dp_> command passed on server but failed on client
23:39:21 <TrueBrain> Not sure it would have caught this one, but it validates a lot more than only checking random :D
23:40:18 <TrueBrain> But it does look like a fun bug :D
23:40:31 <TrueBrain> Question is ... can you reproduce it?
23:42:17 <_dp_> lol, if I could I wouldn't have been asking :p
23:46:54 *** jottyfan has joined #openttd
23:52:49 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
23:55:00 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
23:55:15 *** WormnestAndroid has joined #openttd
23:55:39 *** spnda has quit IRC (Quit: Page closed)
23:59:19 *** sla_ro|master has quit IRC ()
continue to next day ⏵
