Config
Log for #openttd on 1st May 2020:
Times are UTC Toggle Colours
00:05:45  *** Progman has quit IRC
00:24:06  *** Wormnest_ has joined #openttd
00:28:11  *** arikover has quit IRC
01:49:12  *** gelignite__ has joined #openttd
01:51:15  *** spnda has quit IRC
01:55:49  *** adikt has quit IRC
01:56:22  *** gelignite_ has quit IRC
02:05:40  *** glx has quit IRC
02:22:05  *** debdog has joined #openttd
02:25:31  *** D-HUND has quit IRC
02:29:00  *** gelignite__ has quit IRC
02:35:32  *** Flygon has joined #openttd
03:31:28  *** Gustavo6046 is now known as Guest23785
03:31:32  *** Gustavo6046 has joined #openttd
03:31:58  *** Guest23785 has quit IRC
03:38:41  *** Wormnest_ has quit IRC
03:55:44  *** Wormnest_ has joined #openttd
05:35:33  *** Wormnest_ has quit IRC
05:59:30  *** andythenorth has joined #openttd
06:05:36  *** nielsm has quit IRC
06:36:33  <andythenorth> yo
06:52:18  *** Progman has joined #openttd
07:15:17  *** sla_ro|master has joined #openttd
08:15:00  *** Wolf01 has joined #openttd
08:46:58  *** Samu has joined #openttd
08:49:27  <Samu> hi
08:51:56  <Samu> error: https://dev.openttdcoop.org/projects
08:52:00  <Samu> did it die
08:57:00  <andythenorth> yes
08:57:41  <andythenorth> usually someone from coop will restart it, usually the same day
08:57:42  <LordAro> F
08:57:57  <andythenorth> or P or S
08:58:21  <andythenorth> but this is one of the reasons for devzone -> GitHub
09:04:53  *** gelignite has joined #openttd
09:09:56  <Samu> if I am to release my canal pathfinder ai library to bananas, how would others be able to import it?
09:10:12  <Samu> will it show up in online content?
09:12:25  <Samu> ah, it's on Check Online Content instead
09:12:58  <Samu> not shown if i click from inside AI/Game Script settings
09:14:46  <Samu> how do I force download Graph.Aystar 5? https://bananas.openttd.org/package/ai-library/4752412a
09:14:55  <Samu> i got 4 and 6, but not 5
09:15:17  <Samu> wanted to see if version 5 doesn't include the GetLength part
09:18:41  *** iSoSyS has joined #openttd
09:59:37  <Eddi|zuHause> so coop depends on FPS?
10:00:21  <Eddi|zuHause> you know i'm generally short on those :p
10:37:28  <DorpsGek_III> [OpenTTD/OpenTTD] James103 commented on issue #8093: Regular server desynchronisations from clients https://git.io/JfkmV
10:45:03  <DorpsGek_III> [OpenTTD/OpenTTD] James103 commented on issue #8099: UI elements (incl. news message) do not resize properly, causing graphical glitches https://git.io/JftoL
10:51:28  <DorpsGek_III> [OpenTTD/OpenTTD] James103 opened issue #8102: Installer does not say that Windows 8 and 10 are supported. https://git.io/JfO7I
10:52:13  <andythenorth> I should send James the issue count memo
10:52:34  <andythenorth> every issue kills a kitten
10:55:48  *** frosch123 has joined #openttd
11:00:48  <DorpsGek_III> [OpenTTD/OpenTTD] orudge commented on issue #8102: Installer does not say that Windows 8 and 10 are supported. https://git.io/JfO7I
11:02:01  *** ookfof[m] has quit IRC
11:02:01  *** fiddeldibu[m] has quit IRC
11:02:01  *** fjodor[m] has quit IRC
11:02:01  *** tonyfinn[m] has quit IRC
11:02:01  *** khavik[m] has quit IRC
11:02:02  *** grag[m] has quit IRC
11:02:02  *** paulus[m] has quit IRC
11:02:02  *** osvaldo[m] has quit IRC
11:02:02  *** iarp[m] has quit IRC
11:02:02  *** ist5shreawf[m] has quit IRC
11:02:02  *** udo[m] has quit IRC
11:02:02  *** cawal[m] has quit IRC
11:02:02  *** pothyurf[m] has quit IRC
11:02:02  *** jeeg[m] has quit IRC
11:02:02  *** dude[m]1 has quit IRC
11:02:02  *** yur3shmukcik[m] has quit IRC
11:02:02  *** natmac[m] has quit IRC
11:02:02  *** dag[m] has quit IRC
11:02:02  *** jact[m] has quit IRC
11:02:02  *** ciet[m] has quit IRC
11:02:02  *** ad5twoknebor[m] has quit IRC
11:02:02  *** hylshols7qui[m] has quit IRC
11:02:02  *** yoltid[m] has quit IRC
11:02:02  *** nartir[m] has quit IRC
11:02:02  *** labs[m] has quit IRC
11:02:02  *** buggeas40d[m] has quit IRC
11:02:02  *** elliot[m] has quit IRC
11:02:02  *** freu[m] has quit IRC
11:02:02  *** vanessa[m] has quit IRC
11:02:02  *** natalie[m] has quit IRC
11:02:02  *** menelaos[m] has quit IRC
11:02:03  *** karl[m]1 has quit IRC
11:02:03  *** albert[m] has quit IRC
11:02:03  *** blim[m] has quit IRC
11:02:03  *** patricia[m] has quit IRC
11:02:03  *** josef[m]1 has quit IRC
11:02:03  *** joey[m] has quit IRC
11:02:03  *** einar[m] has quit IRC
11:02:03  *** cyberjunkie[m] has quit IRC
11:02:03  *** UncleCJ[m] has quit IRC
11:02:03  *** julie[m] has quit IRC
11:02:03  *** Aileen[m] has quit IRC
11:02:03  *** Heiki[m] has quit IRC
11:02:03  *** leward[m] has quit IRC
11:02:03  *** Corns[m] has quit IRC
11:02:04  *** johanna[m] has quit IRC
11:02:11  *** josef[m] has quit IRC
11:02:11  *** dekeract[m] has quit IRC
11:02:11  *** olmvnec[m] has quit IRC
11:02:11  *** lapav[m] has quit IRC
11:02:11  *** nolep[m] has quit IRC
11:02:11  *** ircer[m] has quit IRC
11:02:11  *** twom[m] has quit IRC
11:02:11  *** glothit7ok[m] has quit IRC
11:02:11  *** karoline[m] has quit IRC
11:02:11  *** robert[m]1 has quit IRC
11:02:11  *** pina[m] has quit IRC
11:02:11  *** amal[m] has quit IRC
11:02:11  *** magdalena[m] has quit IRC
11:02:11  *** igor[m] has quit IRC
11:02:12  *** arron[m] has quit IRC
11:02:12  *** BekJut[m] has quit IRC
11:02:55  <LordAro> i still think at least 70% of these matix users are bots
11:03:01  <LordAro> matrix*
11:10:21  *** _dp_ has quit IRC
11:10:30  *** buggeas40d[m] has joined #openttd
11:19:21  <frosch123> TrueBrain: what do we still run on digital ocean?
11:30:04  *** Heiki[m] has joined #openttd
11:30:05  * Heiki[m] is not a bot
11:36:13  * andythenorth considering building RL state machines (model train signals)
11:36:34  <andythenorth> might have to learn about diodes and relays
11:37:45  <andythenorth> the logic isn't hard to state
11:37:57  <andythenorth> but checking 'vars' involves a lot of wires :(
11:39:05  <frosch123> but relays are like diesel engines, while diodes are electric. how can you mix them?
11:39:58  <andythenorth> I don't know yet :)
11:40:19  <andythenorth> I think I need diodes to prevent one var over-writing another
11:41:32  <andythenorth> I might not need the relays, but they're an abstraction layer - they reduce the number of wires I have to solder to the physical switch in the control panel
11:42:11  <Eddi|zuHause> andythenorth: i think you should go digital if you have that kind of thoughts
11:42:17  <andythenorth> sophisticated railway modellers actually just write software to control these things :P
11:42:28  <andythenorth> but that's just replicating the hobby I already have
11:42:31  <andythenorth> and my day job
11:42:45  <andythenorth> code does not smell like solder
11:43:07  <Eddi|zuHause> you can still solder a raspi to your layout :p
11:43:42  <andythenorth> arduino seems to be the weapon of choice
11:43:51  <Eddi|zuHause> like, you can actually put it into a signal box :p
11:44:46  <andythenorth> I was trying to keep it all rigorously analogue
11:44:50  <andythenorth> where is the boundary? :P
11:45:01  * andythenorth googles 'solid state electronics'
11:45:27  <andythenorth> hmm transistors are 'just' tiny relays
11:45:31  <andythenorth> and diodes are diodes
11:45:38  <andythenorth> https://en.wikipedia.org/wiki/Solid-state_electronics
11:46:00  <Eddi|zuHause> analogue means you can't have any real signal logic, you can press only buttons
11:46:32  <Eddi|zuHause> my signals have a relais built into them
11:46:52  <Eddi|zuHause> so if you switch them red/green it can control the power state of the track
11:47:04  <andythenorth> my kids want that
11:47:22  <andythenorth> I've also found 3DPT or 4DPT switches can achieve the same result
11:47:26  <andythenorth> but longer wires needed :P
11:47:43  <andythenorth> it's quite interesting trying different things
11:47:43  <Eddi|zuHause> there's about a 30-40cm piece of track that is controlled by that signal, the rest is always powered
11:47:58  <andythenorth> yeah that's exactly what my kids are proposing
11:48:08  <andythenorth> I also have to refactor the wiring colour schema :P
11:48:25  <Eddi|zuHause> pro tip for wiring longer distances: ethernet cables
11:48:45  <andythenorth> a 120cm x 360cm oval turns out to have a lot of complexity
11:49:24  <Eddi|zuHause> you have 8 wires in a cable
11:49:27  <TrueBrain> frosch123 : the old new CDN. On the agenda to demolish this month.
11:50:40  <andythenorth> there's also train detector signals, which I'm considering :P
11:51:02  <andythenorth> and I found a push-button sound circuit board which uses MP3s loaded over USB
11:51:23  <Eddi|zuHause> if you want sound, really go digital
11:51:24  <andythenorth> these trains are all analogue, in the UK adding DCC and sound is 200% price
11:51:56  <andythenorth> I am stubbornly resisting DCC
11:52:04  <andythenorth> 2 wire analogue is more interesting
11:52:19  <andythenorth> I have a box full of DCC chips stripped out of ebay purchases
11:52:21  <Eddi|zuHause> my trains are all 30 years old
11:52:32  <andythenorth> I send you DCC chips :P
11:52:40  <Eddi|zuHause> no digital stuff in this...
11:53:08  <Eddi|zuHause> but you can actually power the electric engines from an overhead wire
11:53:47  <andythenorth> my kids also want that
11:54:01  <andythenorth> and they are planning live power pylons also for some reason
11:54:12  <andythenorth> that won't be happening
11:54:50  *** arikover has joined #openttd
11:56:14  <Eddi|zuHause> my engines have a screw on the top that lets you turn it 90° to switch from rail-mode to overhead-mode
11:56:48  <Eddi|zuHause> as in, it operates an actual physical switch
11:58:04  <andythenorth> are they 2 rail or 3 rail?
12:03:41  <Wolf01> <andythenorth> but checking 'vars' involves a lot of wires :( <- make a can bus and attach everything to 1 wire :P
12:04:12  <andythenorth> that's a bit DCC ish :)
12:11:28  <Wolf01> Nah, you can make an analog multiplexer
12:17:04  <andythenorth> o_O
12:21:25  <Wolf01> A bit of transistors, some op-amps, a load of resistances... we made a simple one at school with 3 channels sensing the current on a wire
12:24:42  <Flygon> Scrolling up.
12:24:50  <Eddi|zuHause> andythenorth: my trains are 2-rail
12:25:03  <Flygon> That chat is really making me want to properly learn Electronics so I can design that enhanced SN76489.
12:25:30  <Flygon> (I whipped up something accumulator based, but I don't have the skills to implement it. :D)
12:25:59  *** fiddeldibu[m] has joined #openttd
12:25:59  *** josef[m] has joined #openttd
12:25:59  *** ist5shreawf[m] has joined #openttd
12:26:00  *** johanna[m] has joined #openttd
12:26:00  *** ad5twoknebor[m] has joined #openttd
12:26:01  *** julie[m] has joined #openttd
12:26:01  *** robert[m]1 has joined #openttd
12:26:01  *** iarp[m] has joined #openttd
12:26:01  *** hylshols7qui[m] has joined #openttd
12:26:01  *** cawal[m] has joined #openttd
12:26:01  *** nolep[m] has joined #openttd
12:26:01  *** albert[m] has joined #openttd
12:26:01  *** jeeg[m] has joined #openttd
12:26:01  *** karoline[m] has joined #openttd
12:26:02  *** leward[m] has joined #openttd
12:26:02  *** dag[m] has joined #openttd
12:26:02  *** glothit7ok[m] has joined #openttd
12:26:02  *** ircer[m] has joined #openttd
12:26:03  *** joey[m] has joined #openttd
12:26:03  *** yoltid[m] has joined #openttd
12:26:04  *** UncleCJ[m] has joined #openttd
12:26:04  *** ookfof[m] has joined #openttd
12:26:05  *** arron[m] has joined #openttd
12:26:05  *** igor[m] has joined #openttd
12:26:05  *** natalie[m]1 has joined #openttd
12:26:06  *** Aileen[m] has joined #openttd
12:26:07  *** einar[m] has joined #openttd
12:26:07  *** udo[m] has joined #openttd
12:26:07  *** nartir[m] has joined #openttd
12:26:07  *** menelaos[m] has joined #openttd
12:26:08  *** BekJut[m] has joined #openttd
12:26:09  *** Corns[m] has joined #openttd
12:26:09  *** jact[m] has joined #openttd
12:26:09  *** magdalena[m] has joined #openttd
12:26:09  *** amal[m] has joined #openttd
12:26:09  *** natmac[m] has joined #openttd
12:26:09  *** blim[m] has joined #openttd
12:26:09  *** patricia[m] has joined #openttd
12:26:09  *** dude[m] has joined #openttd
12:26:09  *** elliot[m] has joined #openttd
12:26:09  *** fjodor[m] has joined #openttd
12:26:09  *** ciet[m] has joined #openttd
12:26:09  *** olmvnec[m] has joined #openttd
12:26:09  *** labs[m] has joined #openttd
12:26:09  *** pina[m] has joined #openttd
12:26:10  *** josef[m]1 has joined #openttd
12:26:10  *** lapav[m] has joined #openttd
12:26:10  *** pothyurf[m] has joined #openttd
12:26:10  *** tonyfinn[m] has joined #openttd
12:26:10  *** grag[m] has joined #openttd
12:26:10  *** paulus[m] has joined #openttd
12:26:10  *** dekeract[m] has joined #openttd
12:26:10  *** yur3shmukcik[m] has joined #openttd
12:26:10  *** karl[m] has joined #openttd
12:26:10  *** twom[m] has joined #openttd
12:26:11  *** khavik[m] has joined #openttd
12:26:11  *** freu[m] has joined #openttd
12:26:11  *** osvaldo[m] has joined #openttd
12:26:12  *** cyberjunkie[m] has joined #openttd
12:26:12  *** vanessa[m] has joined #openttd
12:26:25  <Wolf01> Flygon, nice :)
12:28:20  <Flygon> tl;dr - imagine a really really ghetto SID.
12:29:12  <Flygon> 4-bit volume precision, logarithmic. So the Saw/Triangle are actually non-linear, like the AY "Buzzer" instruments.
12:29:25  <Flygon> 16 Duty Cycles. No ADSR envelopes.
12:29:49  <Flygon> Honestly, it was an exercise in figuring out how the SN76489 could've been not-crappy but also not-RIP-silicon-budget.
12:30:28  <Eddi|zuHause> i've not understood a single sentence you just said
12:30:43  <Flygon> Soundchip use to go Beep, now it goes Brrrr.
12:31:14  <Eddi|zuHause> so, you invented a door bell? :p
12:31:23  <andythenorth> ooof
12:31:26  <andythenorth> I was eating Eddi|zuHause
12:31:38  <Flygon> Sure, I've invented the Door Bell. :P
12:31:42  <andythenorth> well played
12:31:58  * Flygon actually laughing really hard IRL
12:49:27  <Samu> how do I download version 5 ? https://bananas.openttd.org/package/ai-library/4752412a/2009-03-29T20:06:10+00:00
12:51:04  *** supermop_Home has joined #openttd
12:52:04  <andythenorth> so 'alkenes' or 'olefins'? :P
12:52:41  <supermop_Home> olefins sounds very british
12:52:54  *** arikover has quit IRC
12:53:01  <andythenorth> they seem to be interchangeable
12:53:09  *** arikover has joined #openttd
12:53:22  <supermop_Home> but idk if many americans ever utter the word 'alkenes' outside of high school chemistry class
12:53:36  *** glx has joined #openttd
12:53:36  *** ChanServ sets mode: +v glx
12:54:01  <andythenorth> I didn't know it existed until a few weeks ago
12:54:13  <andythenorth> chemistry is largely mysterious to me
12:54:40  <supermop_Home> tbh I've never heard my father-in-law, a chemical engineer who specialized in industrial production of polypropylene, say either word
12:55:13  <supermop_Home> but i know you weirdos say 'polyolefin'?
12:56:23  <andythenorth> well usually it would be propene (propylene) and ethene (ethylene)
12:56:32  <andythenorth> as commodity basic chemicals
12:56:41  <andythenorth> but as they both are cracked from naptha in the same cracker
12:56:47  <andythenorth> and both go to a polymer plant...
12:57:16  <supermop_Home> i mean we usually use the 'ene suffix names and similar for most organics here
12:57:17  <andythenorth> there are alternative cracking chains which split them, but they seem to be less standard
12:57:39  <Eddi|zuHause> andythenorth: so i think i've heard "olefines" more before
12:58:09  <supermop_Home> but wouldn't really say the generic al-xxx type name outside of an academic or theory type setting
12:58:46  <supermop_Home> Eddi|zuHause i think its us / Europe divide then?
12:59:33  <Eddi|zuHause> dunno, they started calling our local chemistry complex that, when the americans (DOW) took over
12:59:42  <andythenorth> I don't really like composite cargos, but chemical industry requires some level of generic grouping
13:00:03  *** iSoSyS has quit IRC
13:00:24  <supermop_Home> Eddi|zuHause also what industry say vs us lay people who are trying to remember our high school text books and CRC manuals might differ
13:00:58  <Eddi|zuHause> i think when we design an industry set, we can orient ourselves on the industry terminology
13:01:01  <andythenorth> there are about 33 listed here https://www.essentialchemicalindustry.org/chemicals.html
13:01:14  <andythenorth> they're not all relevant, and some other cargos are needed
13:01:39  <andythenorth> some of the more niche polymer feedstocks aren't worth covering
13:02:14  <Eddi|zuHause> i think that list is too specific for game abstraction terms
13:02:36  <andythenorth> the organics are definitely
13:02:46  <supermop_Home> i just never encounter olefin in my life except when talking to british designers who refer to polypropylene chairs and bins from muji as 'polyolefin'
13:02:55  <andythenorth> my rough guide has been to look at how many million tonnes are produced
13:03:11  <andythenorth> a secondary consideration is if the industry appearance or transport method is 'interesting'
13:03:52  <supermop_Home> hoppers of pellets
13:03:56  <andythenorth> taken too literally everything will travel by either insulated tank car, or pipeline
13:04:01  <supermop_Home> more fun than pipes
13:04:03  <andythenorth> yes pellets
13:04:04  <andythenorth> etc
13:04:05  <supermop_Home> so far
13:04:20  <andythenorth> as we don't have pipelines, that's a thing
13:04:26  <andythenorth> we do have PIPE :P
13:04:43  <supermop_Home> bucket ropeway of pellets - pipe even the solids
13:05:31  <supermop_Home> torpedo cars of molten plastic
13:07:40  <andythenorth> transport tubes
13:08:08  <andythenorth> current https://firs-test-1.s3.eu-west-2.amazonaws.com/firs-bltc/docs/html/economies.html#better_living_through_chemistry
13:08:10  <andythenorth> quite broken
13:08:44  <andythenorth> pleasingly fat middle though
13:12:30  *** nielsm has joined #openttd
13:21:23  *** supermop_Home has quit IRC
13:28:17  *** supermop_Home has joined #openttd
13:28:22  <supermop_Home> ugh
13:29:40  *** arikover has quit IRC
13:42:47  *** arikover has joined #openttd
13:46:16  *** arikover has quit IRC
13:50:40  <Samu> oh no, I found a ship pathfinder https://www.tt-forums.net/viewtopic.php?p=797775#p797775
13:50:46  <Samu> i've been working for nothing
13:51:20  <Samu> gonna test it
13:59:46  <DorpsGek_III> [OpenTTD/OpenTTD] glx22 commented on issue #8099: UI elements (incl. news message) do not resize properly, causing graphical glitches https://git.io/JftoL
14:14:54  <Samu> has some outdated code like AIAbstractList, it's now AIlist
14:14:58  <Samu> AIList*
14:25:12  *** supermop_Home has quit IRC
14:55:55  <DorpsGek_III> [OpenTTD/OpenTTD] glx22 commented on issue #8093: Regular server desynchronisations from clients https://git.io/JfkmV
15:02:26  *** Wormnest_ has joined #openttd
15:03:15  <Samu> well, the pathfinder is flawed
15:03:29  <Samu> already found it blocking old paths with locks
15:03:49  <Samu> looks like mine is "superior"
15:04:16  <Samu> it also fails building aqueducts due to other aqueducts planned
15:05:01  <Samu> but it's faster than mine :(
15:07:20  <Samu> https://i.imgur.com/0LbCcd5.png - connecting 5A to 5B, my pathfinder on the left, the one i found on the forums on the right
15:16:42  <andythenorth> hmm chemistry is so much more complicated than steel industry
15:17:23  <andythenorth> steel: burn 3 minerals together, pour out, burn again with oxygen + additives -> roll into shapes -> profit
15:18:34  <andythenorth> chem industry seems to vary highly by era, country and process
15:18:47  <andythenorth> sometimes an A is processed to produce B
15:18:55  <andythenorth> and sometimes B is processed to produce A
15:19:15  <andythenorth> and sometimes processing B into C produces more A
15:19:18  <andythenorth> very baffling
15:19:51  <andythenorth> I mean it's all just bonds and counting, but eh
15:28:54  <DorpsGek_III> [OpenTTD/OpenTTD] frosch123 opened pull request #8103: Fix #8093: Do not call OrderBackup::Restore in the command test run, and use DC_AUTOREPLACE for actions that shall be reverted. https://git.io/JfOjg
15:41:14  <LordAro> oh woops
15:41:21  <LordAro> TrueBrain_ii: ^ you broke the CI :p
15:41:33  <LordAro> indirectly
15:42:03  <glx> maybe we just need to fix yaml
15:42:10  <LordAro> yeah, that'd be the better solution
15:43:07  <LordAro> frosch123: https://github.com/OpenTTD/OpenTTD/blob/master/azure-pipelines/templates/ci-opengfx.yml might as well update it in the same PR
15:43:57  <frosch123> no idea what that means
15:44:08  <frosch123> anyway, i don't get this OrderBackup thing yet
15:44:44  <LordAro> url has moved to https://cdn.openttd.org/opengfx-releases/0.6.0/opengfx-0.6.0-all.zip
15:44:58  <frosch123> how does ogfx affect openttd?
15:45:08  <LordAro> it's required for regression tests
15:45:14  <LordAro> because Reasons(tm)
15:48:17  <glx> linux CF should be fine, it uses apt to get opengfx
15:48:18  <frosch123> oi, we have an entire folder for azure stuff
15:50:14  <frosch123> LordAro: if the old ogfx url is no longer valid, then also our installer does no longer work?
15:50:27  <LordAro> TB said he tested it...
15:51:35  <Eddi|zuHause> "it was tested" was always a sure thing that something breaks right after release :p
15:51:57  <glx> this one works
15:52:24  <glx> installer uses http://binaries.openttd.org/installer/opengfx-1.2.0.7z
15:52:25  <DorpsGek_III> [OpenTTD/OpenTTD] frosch123 updated pull request #8103: Fix #8093: Do not call OrderBackup::Restore in the command test run, and use DC_AUTOREPLACE for actions that shall be reverted. https://git.io/JfOjg
15:57:31  * glx fetches the PR to do some tests :)
15:58:48  <DorpsGek_III> [OpenTTD/OpenTTD] frosch123 updated pull request #8103: Fix #8093: Build+Refit changed game-state in command test run, and thus caused desyncs https://git.io/JfOjg
15:58:50  <frosch123> glx: i added another commit
15:59:43  <frosch123> not sure whether to put first and third into one
15:59:48  <frosch123> or whether to split the first into two
16:12:47  <glx> frosch123: seems to work
16:14:21  *** iSoSyS has joined #openttd
16:15:06  <glx> ah I triggered an assert
16:16:54  <nielsm> I wonder if it's be reasonable to make industry production changes merge into a single shared monthly newspaper in smooth economy
16:19:34  <glx> frosch123: test and exec may return different costs
16:21:23  <frosch123> that sounds odd
16:21:47  <glx> I think it happens when free wagons can be refited
16:22:03  <glx> I'll retry this exact case to be sure
16:23:52  <frosch123> ah, you built a wagon?
16:24:32  <glx> yep confirmed
16:25:36  *** Flygon has quit IRC
16:25:51  <glx> using UKRS I build a utility van ([mail], goods, valuable), then I build a DMU ([passenger], mail, valuable) with auto refit to valuables
16:28:14  <frosch123> yeah, it attaches the free wagons first, then refits them
16:28:27  <frosch123> i guess the intention is that the free wagons shall not be refitted?
16:28:55  <nielsm> quite sure only the one wagon you're building should refit
16:29:33  <glx> yes freewagons attached to the new engine should not be refited
16:30:28  <frosch123> p2 bit 25
16:30:42  <frosch123> do i just put 1 << 25 into the code? i can't remember
16:32:18  <glx> trying with the flag
16:32:29  <frosch123> it's wrong
16:32:33  <frosch123> 1 << 16 is better
16:32:37  <FLHerne> Eddi|zuHause: "it was tested" was always a sure thing that something breaks  <- because the things people bother to test at all are the risky ones? :P
16:33:47  <FLHerne> Similarly, code with a comment explaining how the implementation works is more likely to be wrong
16:34:40  <glx> comment is rarely updated when implementation changes ;)
16:35:02  <DorpsGek_III> [OpenTTD/OpenTTD] frosch123 updated pull request #8103: Fix #8093: Build+Refit changed game-state in command test run, and thus caused desyncs https://git.io/JfOjg
16:35:54  <TrueBrain> there is always this one place thse uses some deprecated URLs :D Happy we found that place :P
16:36:35  <glx> frosch123: 1 << 25 works for me
16:37:33  <frosch123> it should break for articulated and dual-headed engines
16:37:40  <TrueBrain> ah, I was about to say: I did redirect /extra, didn't I ? But opengfx got renamed ofc :D Guess apt and yum are also going to break :P
16:37:40  <glx> oh no only head of multihead is refitted
16:37:59  <glx> but it doesn't assert ;)
16:46:38  <DorpsGek_III> [OpenTTD/OpenTTD] BeWorld2018 opened issue #8104: SDL2 : Fullscreen to Window https://git.io/Jf3Jn
16:51:52  <DorpsGek_III> [OpenTTD/OpenTTD] glx22 commented on pull request #8103: Fix #8093: Build+Refit changed game-state in command test run, and thus caused desyncs https://git.io/Jf3JX
16:55:44  <DorpsGek_III> [OpenTTD/OpenTTD] glx22 commented on issue #8104: SDL2 : Fullscreen to Window https://git.io/Jf3Jn
17:25:01  *** andythenorth has quit IRC
17:27:09  <TrueBrain> https://github.com/OpenTTD/BaNaNaS/compare/6ac43cc788f9...da61bb20fde9 <- this is what we did it for frosch123  :) I am really happy people can finally make edits like this :)
17:27:19  <TrueBrain> (I am being serious, to be clear)
17:27:30  *** andythenorth has joined #openttd
17:27:44  <frosch123> oh, i thought you meant renaming stuff
17:27:46  <TrueBrain> and I said this while andythenorth was gone, because spies :)
17:27:52  <andythenorth> yes
17:28:11  <frosch123> anyway, when does ottd drop the spritefont? :p
17:28:24  <TrueBrain> when you finally make a patch for it?
17:30:04  <frosch123> the description was already in korean before
17:30:48  <TrueBrain> CloudFlare hasn't responded yet ... I hate waiting for repliesssssssss :D
17:31:05  <frosch123> it's friday
17:31:27  <frosch123> maybe their ceo is ottd fan :p
17:31:31  <TrueBrain> I sent it yesterday! :P
17:31:44  <TrueBrain> yeah, that is my scare .. that they clicked the link, started to play, and forgot to reply :P
17:31:49  *** andythenorth has quit IRC
17:31:56  <TrueBrain> nah, they are busy people :) They will get back to me sooner or later :)
17:32:19  <frosch123> how many employees do they have? 10?
17:32:27  <TrueBrain> a LOT more
17:32:37  <TrueBrain> https://www.cloudflare.com/people/
17:33:03  <frosch123> they put everyone on their website?
17:33:10  <frosch123> boy would that page burn in germany :)
17:33:16  <TrueBrain> that I do not know; but at least those people work their :P
17:33:33  <TrueBrain> it is only a first-name
17:33:37  <TrueBrain> it is not THAT bad
17:34:11  <frosch123> oh, Darth SriBear works there
17:36:58  <frosch123> TrueBrain: even on my company-internal sites, about 10% do not have a photo
17:40:04  <frosch123> there are 814 people listed on that page. wiki claims valve has only 360
17:40:38  *** spnda has joined #openttd
17:41:17  <frosch123> hmm i guess they have business clients instead of end-customers. so they probably actively sell stuff
17:43:40  <frosch123> also, user migrations are now 13:13, tt-forums vs email
17:43:49  <frosch123> i was hoping for more forum people
17:45:21  *** iSoSyS has quit IRC
17:45:46  <frosch123> no, lies, i did 3 on my own, so email is a head
18:02:23  <DorpsGek_III> [OpenTTD/OpenTTD] frosch123 opened pull request #8105: Fix: OpenGFX download link did no longer work. https://git.io/Jf3kP
18:02:29  <frosch123> let's see whether 7z is installed
18:03:20  *** andythenorth has joined #openttd
18:04:24  <frosch123> nope, missing at least on osx
18:05:15  <DorpsGek_III> [OpenTTD/DorpsGek] pyup-bot opened pull request #23: Scheduled monthly dependency update for May https://git.io/Jf3k5
18:09:03  *** spnda has quit IRC
18:11:39  *** andythenorth has quit IRC
18:12:20  <LordAro> ha
18:14:45  <nielsm> turns out, 7z is _not_ installed on the mac
18:14:55  <frosch123> osx-dependencies already contains xz
18:15:16  <frosch123> unless those are just libs
18:16:53  <DorpsGek_III> [OpenTTD/OpenTTD] TrueBrain commented on pull request #8105: Fix: OpenGFX download link did no longer work. https://git.io/Jf3Iq
18:17:22  <DorpsGek_III> [OpenTTD/OpenTTD] frosch123 updated pull request #8105: Fix: OpenGFX download link did no longer work. https://git.io/Jf3kP
18:17:54  <DorpsGek_III> [OpenTTD/DorpsGek] LordAro approved pull request #23: Scheduled monthly dependency update for May https://git.io/Jf3Is
18:17:55  <frosch123> aw, i really liked making it a unit-test for the installer
18:17:59  <TrueBrain> there really is no benefit to using "1.2.0" which is actually "0.6.0" .. this will go horribly wrong :P
18:18:17  <TrueBrain> it is REALYL the wrong place if that was your goal :P
18:18:23  *** andythenorth has joined #openttd
18:18:44  <frosch123> well, i also didn't like writing a fixed version into it
18:19:08  <LordAro> mm, there's no "latest" link for the releases, is there?
18:19:24  <TrueBrain> you can fetch the latest.yaml, if that is your itch
18:19:26  <LordAro> probably not one that wouldn't have the version in the filename, anyway
18:19:39  <TrueBrain> https://cdn.openttd.org/opengfx-releases/latest.yaml
18:19:50  <frosch123> oh dear, i thought xz and 7z were the same :p
18:20:01  <LordAro> oh dear.
18:20:04  <LordAro> they're really not
18:20:33  <TrueBrain> but who ever thought up naming the installer files "opengfx-1.2.0" to indicate OpenTTD client versions should retroactively be punished (sorry .. I am guessing Rb)
18:21:11  <frosch123> how would you have named them?
18:21:18  <frosch123> it seems to meet all goals
18:21:32  <TrueBrain> having a opengfx-0.6.0 and an opengfx-1.2.0 .. is .. just asking for trouble
18:21:33  <frosch123> giving installers the latest version that is compatible
18:21:50  <TrueBrain> so anything else would have worked :P
18:21:53  <TrueBrain> just .. about .. anything
18:22:04  <TrueBrain> opengfx-for-openttd-1.2.0+ might have been the best :P
18:23:48  <DorpsGek_III> [OpenTTD/nml] matthijskooijman opened issue #111: Versioned source tarball not available through github https://git.io/Jf3IB
18:23:53  <DorpsGek_III> [OpenTTD/DorpsGek] TrueBrain merged pull request #23: Scheduled monthly dependency update for May https://git.io/Jf3k5
18:24:15  <TrueBrain> we will see if LordAro is correct :D
18:24:39  <TrueBrain> I am surprised 'jq' is pretty common these days, but there is not a common tool for yaml yet?
18:25:18  <TrueBrain> pip install yq
18:25:19  <TrueBrain> lol
18:25:39  <LordAro> pronounced "yuck"
18:26:16  <TrueBrain> guess it is "good enough"
18:28:33  <DorpsGek_III> [OpenTTD/OpenTTD] frosch123 updated pull request #8105: Fix: OpenGFX download link did no longer work. https://git.io/Jf3kP
18:35:00  <DorpsGek_III> [OpenTTD/nml] matthijskooijman commented on issue #111: Versioned source tarball not available through github https://git.io/Jf3IB
18:55:35  <DorpsGek_III> [OpenTTD/nml] matthijskooijman commented on issue #111: Versioned source tarball not available through github https://git.io/Jf3IB
19:01:24  <DorpsGek_III> [OpenTTD/nml] LordAro commented on issue #111: Versioned source tarball not available through github https://git.io/Jf3IB
19:01:30  <LordAro> TrueBrain: ^ thoughts?
19:02:50  <TrueBrain> it is indeed normal to build the pip tarballs on the linux machine
19:02:52  <TrueBrain> why can't you?
19:03:25  <LordAro> because the action i'm using to build the manylinux-compatible wheels doesn't really allow for it
19:03:32  <LordAro> well it does, but it'd be a pain
19:03:36  <TrueBrain> ps: maybe split the workflows over the different OSes, instead in a single one with all of them "if" statements for OSes :P
19:03:46  <LordAro> mm, maybe
19:03:48  <TrueBrain> ah .. that is silly of that action
19:03:55  <TrueBrain> so pick the OSX indeed :P
19:04:04  <TrueBrain> or make a job specific for tarball
19:04:06  <TrueBrain> also pretty common
19:04:34  <andythenorth> it's Friday!
19:04:38  <TrueBrain> what is build for OSX btw? I only see it pip install something, and that is it
19:04:40  * andythenorth has lost sanity this week
19:05:17  <LordAro> TrueBrain: ! startsWith(...ubuntu..)
19:05:21  <TrueBrain> ugh
19:05:24  <TrueBrain> as said, hard to read :D
19:05:31  <LordAro> indeed
19:05:39  <LordAro> i'll see what i can do about splitting them up
19:05:48  <LordAro> though i'd have to duplicate the caching stuff, which is sad
19:06:09  <TrueBrain> caching stuff is per OS anyway
19:06:11  <DorpsGek_III> [OpenTTD/nml] matthijskooijman commented on issue #111: Versioned source tarball not available through github https://git.io/Jf3IB
19:10:38  <andythenorth> right what has OS X done wrong now?
19:12:24  *** andythenorth has quit IRC
19:16:42  *** andythenorth has joined #openttd
19:21:24  <DorpsGek_III> [OpenTTD/nml] matthijskooijman opened issue #112: setup.py overwrites nml/__version__.py https://git.io/Jf3Lj
19:21:34  <andythenorth> not sure what to do with nml now
19:21:43  <andythenorth> it's broken in prod
19:21:58  <andythenorth> I merged the fix, but now that's blocked on debate
19:22:01  <andythenorth> so can't release
19:22:26  <LordAro> it's no more broken that the current release is
19:22:31  <LordAro> i wouldn't say that's blocking
19:22:46  <andythenorth> https://github.com/OpenTTD/nml/pull/110#issuecomment-620169614
19:22:56  <andythenorth> author of PR thinks it shouldn't be there
19:23:08  * andythenorth confused emoji
19:23:21  <LordAro> well, revert it then
19:23:33  <andythenorth> that is one solution yes
19:24:08  <LordAro> clearly someone needs to nail down a) what NRT should do b) what NRT does c) what NML thinks NRT does and d) resolve the 3
19:24:15  <glx> main issue is version doesn't use __version__.py at all
19:25:14  <LordAro> glx: you can fix that one :)
19:25:32  <glx> there's dead code also (hg stuff ;) )
19:26:13  <andythenorth> currently of the 3 main NRT participants, peter has sad quit, Wolf has no further opinion, and I am confused
19:26:25  <andythenorth> good outcome, classic major OpenTTD change :)
19:27:50  <andythenorth> afaict the issue resolves to: are labels pooled or not?
19:29:27  <frosch123> andythenorth: the maximum limit of types is an arbirtrary number
19:29:37  <michi_cc> That depends on where you ask. OpenTTD and the specs published on some wiki disagree, the mythical somebody has to decide which is right.
19:29:48  <frosch123> it's shared accross all grfs, so what's the point in discussing the limit for a single grf?
19:30:22  <DorpsGek_III> [OpenTTD/nml] matthijskooijman commented on issue #111: Versioned source tarball not available through github https://git.io/Jf3IB
19:30:26  <frosch123> 64 types is hard for ottd to achieve, either 63, or 127 :p
19:30:29  <frosch123> and noone wants 127
19:31:06  <DorpsGek_III> [OpenTTD/nml] LordAro opened pull request #113: Rewrite release workflow https://git.io/Jf3tn
19:31:12  <frosch123> i don't know the memes that well, so i don't know whether 63 vs 64 is bike-shedding or yak-shaving
19:31:35  <frosch123> if you want to fix nrt in ottd, fix the bridge catenary, or the ai support
19:31:45  <frosch123> but don't bother with 63/64 :)
19:33:51  <andythenorth> I thought the 63 vs 64 was a non issue?
19:33:52  <DorpsGek_III> [OpenTTD/nml] glx22 commented on issue #111: Versioned source tarball not available through github https://git.io/Jf3IB
19:33:54  <andythenorth> it's 63
19:34:13  <andythenorth> I thought the issue was whether labels for Roadtype and Tramtype can collide or not?
19:34:51  <andythenorth> FLHerne hi hi
19:34:59  <frosch123> that's a different issue than 110
19:35:03  <andythenorth> yes
19:35:15  <andythenorth> I only have confusion currently, but I'm not sure the confusion is valid
19:35:41  <andythenorth> 63 is the actual spec and what's implemented, the wiki has 'alternative facts'
19:35:51  <andythenorth> I know why and how that happened
19:36:24  <frosch123> now ottd, wiki and nml are consistent, why do you care more?
19:37:16  <frosch123> there is nothing more to achieve for an arbitrary number, except consistency
19:37:37  <andythenorth> mostly because I don't want to shed any more contributors over this
19:37:51  <andythenorth> we 1 down already
19:38:07  <frosch123> it's over a year ago
19:38:15  <frosch123> everyone should have made their mind up by now
19:38:22  <DorpsGek_III> [OpenTTD/nml] matthijskooijman commented on pull request #113: Rewrite release workflow https://git.io/Jf3t0
19:40:23  <DorpsGek_III> [OpenTTD/nml] LordAro commented on pull request #113: Rewrite release workflow https://git.io/Jf3ta
19:40:41  <DorpsGek_III> [OpenTTD/nml] LordAro commented on pull request #113: Rewrite release workflow https://git.io/Jf3tV
19:42:05  <DorpsGek_III> [OpenTTD/nml] glx22 commented on pull request #113: Rewrite release workflow https://git.io/Jf3tK
19:43:40  <LordAro> glx: i'm happy to merge #108 into #113 if you like
19:44:02  <glx> makes sense to merge them
19:44:21  <LordAro> glx: can you make sure there's a correct GITHUB_TOKEN secret?
19:44:28  <DorpsGek_III> [OpenTTD/nml] matthijskooijman commented on pull request #113: Rewrite release workflow https://git.io/Jf3tP
19:44:38  <glx> it's generated by github action
19:44:54  <glx> as long as the run doesn't take longer than 1h it's valid
19:45:00  <LordAro> ah right
19:45:26  <DorpsGek_III> [OpenTTD/nml] matthijskooijman approved pull request #108: Codechange: [Actions] Upload standalone executable to GitHub release https://git.io/Jf3ty
19:46:09  <FLHerne> andythenorth: I think the spec'd behaviour as implemented by #110 is much nicer
19:46:51  <FLHerne> I just think it was a bit hasty to push it without discussing with Peter why he implemented it differently
19:47:02  <FLHerne> And how practical it would be to change that
19:47:11  <FLHerne> s/implemented by/assumed by/
19:47:32  <Samu> what happened to peter?
19:47:59  <glx> hmm just to test I ran "pip install nml", and I can't run nmlc
19:48:05  <glx> (on windows)
19:48:40  <andythenorth> FWIW, I think the solution for nml in 110 is the only viable one currently
19:48:44  <LordAro> did it put it somewhere not on PATH?
19:48:44  <FLHerne> andythenorth: In any case, #110 is no more wrong than the current behaviour
19:48:46  <andythenorth> and authors will just need to take care
19:49:02  <DorpsGek_III> [OpenTTD/nml] matthijskooijman commented on issue #111: Versioned source tarball not available through github https://git.io/Jf3IB
19:49:05  <FLHerne> andythenorth: It allows the user to be wrong, rather than preventing the user from being right
19:49:11  <glx> File "C:\Users\Loïc\AppData\Local\Programs\Python\Python37\Scripts\nmlc-script.py", line 1
19:49:11  <glx> SyntaxError: Non-UTF-8 code starting with '\xef' in file C:\Users\Loïc\AppData\Local\Programs\Python\Python37\Scripts\nmlc-script.py on line 1, but no encoding declared; see http://python.org/dev/peps/pep-0263/ for details
19:49:16  <glx> it's in path
19:49:22  <glx> it just doesn't work
19:49:27  <FLHerne> (where 'wrong' is "matches the limits in OTTD")
19:50:33  <TrueBrain> glx: your path is the issue :) This happens with some versions of pip
19:50:55  <TrueBrain> your have to change the shebang before it works
19:51:00  <FLHerne> Er, *doesn't* match
19:51:03  <TrueBrain> check other files in Scripts to see to what, at least, on linux :)
19:52:40  <glx> but with "pip install -e ." it works
19:53:53  <LordAro> glx: oh heavens
19:54:05  <LordAro> \xef, is that a UTF8 BOM?
19:55:20  <TrueBrain> glx: so compare the files :)
19:55:38  <TrueBrain> for me it happens that the shebang is wrong because I have a space in my path
19:55:40  <TrueBrain> and weird shit happens
19:55:45  <TrueBrain> pip install mostly goes wrong, not always
19:55:55  <TrueBrain> haven't pinned down when exactly it does one thing or the other :P
19:55:58  <TrueBrain> pip is cute
19:56:13  <LordAro> there's no nmlc-script.py in the repo, must be generated by something
19:56:29  <andythenorth> reddit has been busy :) https://i.redd.it/p128gy5mf4w41.png
19:56:31  <LordAro> nor does \xef appear anywhere in the source
19:56:35  <TrueBrain> everything in Script / bin folder is generated LordAro
19:57:09  <andythenorth> also this was really nice :) https://www.tt-forums.net/viewtopic.php?f=29&t=86978
19:58:00  <glx> oh \xef is ï IIRC
20:06:09  <blathijs> Hm, should have joined into the discussion here too :-)
20:06:43  <blathijs> Seems that CRLF line endings have another problem on Linux: The kernel does not like "python3\r" as an interpreter from the shebang :-p
20:06:53  * blathijs adds another workaround trying to get his .deb to build
20:07:28  <LordAro> blathijs: i've run into that one before now :)
20:08:43  <blathijs> Seems to work now, though. If it workaroudn all the newlines and the __version__.py overwriting, it builds (of course I now realize I should just have repacked the 0.5.0 tarball with fixed newlines and use that, rather than importing the broken one and working around it, but well).
20:10:34  <glx> oh I see with "-e ." there's a "-x" added at the end of the shebang line
20:10:53  <blathijs> Gotta run now. I'll be away most of tomorrow, but I'll try to check GH, just in case there is some progress and maybe a tarball to do a preflight run with? :-)
20:11:04  <DorpsGek_III> [OpenTTD/nml] LordAro updated pull request #113: Rewrite release workflow https://git.io/Jf3tn
20:11:31  <blathijs> Thanks all for jumping on this :-D
20:11:59  <DorpsGek_III> [OpenTTD/nml] LordAro commented on pull request #108: Codechange: [Actions] Upload standalone executable to GitHub release https://git.io/Jf3qX
20:11:59  <DorpsGek_III> [OpenTTD/nml] LordAro closed pull request #108: Codechange: [Actions] Upload standalone executable to GitHub release https://git.io/JfqJv
20:12:14  <LordAro> glx: ok, now it's completely untested :)
20:14:00  <glx> well #108 worked for me when I tried in my fork
20:15:29  <LordAro> i'm trying your method of a test commit to remove the twine calls
20:17:36  <glx> I can try it on my fork too
20:17:46  <LordAro> https://github.com/LordAro/nml/releases/tag/0.5.1-RC2
20:17:47  <LordAro> seems to work
20:17:54  <LordAro> though i think your version stuff might be broken
20:18:19  <glx> hmm it's supposed to use the tag
20:18:29  <LordAro> oh, it's named the source badly
20:18:33  <LordAro> that's easily fixed
20:20:42  <DorpsGek_III> [OpenTTD/nml] LordAro updated pull request #113: Rewrite release workflow https://git.io/Jf3tn
20:22:32  <LordAro> glx: `python nml/version_info.py` definitely seems to work when the tag is directly checked out
20:22:43  <LordAro> is this the same issue OTTD had - it's the commit that's checked out, not the tag?
20:23:04  <glx> ah maybe
20:24:55  <DorpsGek_III> [OpenTTD/nml] TrueBrain approved pull request #113: Rewrite release workflow https://git.io/Jf3mv
20:26:29  <TrueBrain> You have to fetch the tags separately if you need them
20:26:40  <TrueBrain> Default checkouts only checkout HEAD
20:27:17  <TrueBrain> See website workflows to see how; I made a helper for that
20:27:35  <TrueBrain> Or any other repo, for that matter
20:28:05  <TrueBrain> https://github.com/OpenTTD/bananas-api/blob/master/.github/workflows/publish.yml#L29
20:28:08  <TrueBrain> As example
20:28:50  <TrueBrain> I love checkouts with depth=0 .. it is quick :)
20:28:51  <LordAro> i'm looking at the docs - it looks like GITHUB_REF would be enough?
20:29:08  <LordAro> https://help.github.com/en/actions/reference/events-that-trigger-workflows#release-event-release
20:29:18  <TrueBrain> We use 'got desribe' in nml right?
20:29:28  <TrueBrain> So no, git needs more than a ref in that case
20:29:31  <LordAro> + https://github.com/actions/checkout#checkout-pull-request-head-commit-instead-of-merge-commit
20:29:32  <TrueBrain> See my URL
20:29:49  <glx> yes git describe is used now so action variables are no help :)
20:30:11  <LordAro> have you read the docs?
20:30:20  <LordAro> on the release event, GITHUB_REF is the tag of the release
20:30:31  <LordAro> why is that not enough?
20:30:58  <glx> nml version is determined via "git describe"
20:31:27  <glx> so internally it will be wrong too
20:31:28  <LordAro> yes, so use GITHUB_REF as the ref parameter of the actions/checkout stage
20:31:49  <TrueBrain> Have you read anything I wrote? (What a lame thing to say btw :p). It -might- work if you pull refs/tags/NNN, but how things normally work is that you get the hash-ref, not the tag-ref. Try it out
20:31:49  <LordAro> i'm not saying use it to determine the name of things
20:32:13  *** dP has joined #openttd
20:32:13  *** dP is now known as _dp_
20:32:18  <TrueBrain> Just know git describe is annoying ;)
20:33:11  <TrueBrain> So we have two conversations, great :p
20:34:41  <frosch123> TrueBrain: ah, right, i had this thing to discuss with you :)
20:34:53  <TrueBrain> Shall I run now? :p
20:35:39  <frosch123> :)
20:38:02  <TrueBrain> I like that he clearly forgot already :p
20:40:02  <LordAro> ok, as it turns out, TrueBrain is correct
20:40:08  <LordAro> surprise surprise
20:40:25  <LordAro> i still maintain that those docs suggest that you should get the actual tag
20:40:41  <TrueBrain> Was that sarcastic? :p
20:40:55  <LordAro> only the 2nd line :p
20:41:09  <TrueBrain> Don't get me wrong, I am totally fine with you finding these things out yourself :)
20:41:48  <TrueBrain> Own experience is best :) (and I mean this honestly)
20:41:59  <LordAro> it wasn't my intention for "have you read the docs?" to be quite so confrontational though, so that's my bad
20:42:00  <glx> the doc has https://github.com/actions/checkout#fetch-all-tags ;)
20:42:15  * TrueBrain hugs LordAro
20:42:28  <glx> but yeah actions can be a pain for "simple" things
20:42:54  <TrueBrain> glx : that is what openttd-actions does ;)
20:43:04  <TrueBrain> I wrapped it in a nice action
20:43:28  <glx> yeah and we needed workarounds for commit-checker too
20:43:49  <LordAro> oh i see, and just fetching the tags is enough for git describe to work out what the commit is actually pointing to?
20:44:03  <LordAro> i'd have thought you'd have to check the tag out explicitly
20:44:08  <TrueBrain> It needs only the first tag in the tree
20:44:44  <LordAro> we (nml) don't need the dispatch stuff, do we?
20:44:44  <glx> hmm but a merge commit can be wrong in our case
20:44:49  <TrueBrain> But I cannot remember if that was possible in git
20:45:20  <glx> tag will be correct, but I use == 0 to determine if it's a release or not ie exact tag
20:45:31  <TrueBrain> LordAro : unlikely. Only if you want to rebuild.. don't think that is useful
20:46:25  <glx> so I think we need to use both HEAD stuff, and openttd action to get the tags
20:46:34  <DorpsGek_III> [OpenTTD/nml] LordAro dismissed a review for pull request #113: Rewrite release workflow https://git.io/Jf3mv
20:46:34  <DorpsGek_III> [OpenTTD/nml] LordAro updated pull request #113: Rewrite release workflow https://git.io/Jf3tn
20:46:43  <TrueBrain> It is what we settled in with 'website'
20:46:55  <TrueBrain> All I remember that was a bitch to figure out :p
20:47:00  <TrueBrain> It is a blur otherwise
20:47:29  <DorpsGek_III> [OpenTTD/nml] TrueBrain approved pull request #113: Rewrite release workflow https://git.io/Jf3m2
20:48:39  <LordAro> fake-0.5.1-RC4 coming up...
20:48:49  <TrueBrain> :D
20:50:31  <glx> hmm still wrong it seems
20:50:38  <TrueBrain> :(
20:51:03  <LordAro> :/
20:51:12  <TrueBrain> What is wrong exactly?
20:52:45  <LordAro> checking out the tags didn't seem to help
20:52:59  <LordAro> still getting a version of 0.5.1.post0.dev20200501+0
20:53:25  <TrueBrain> What is that +0?
20:54:09  <LordAro> invented for pypi compatibility, just the nth commit of the day
20:54:12  <LordAro> or something like that
20:54:18  <TrueBrain> Anyway, clearly it did pick up the tag
20:54:24  <glx> theorically changeset
20:54:37  <LordAro> hang on
20:54:39  <glx> but it should not enter in this section
20:54:42  <LordAro> $ python nml/version_info.py
20:54:42  <LordAro> 0;master;0.5.1;False;False;2020-05-01;0.5.1.post0.dev20200501+0
20:55:04  <glx> yes it doesn't detect the "release" state
20:55:05  <LordAro> is what i'm getting locally (after having checked out refs/tags/0.5.1-RC4 and fetching the tags)
20:55:28  <TrueBrain> Sounds like a Python issue :p
20:55:39  <LordAro> yeah :D
20:55:44  * LordAro shoves it in glx's direction
20:55:49  <TrueBrain> Hahaha
20:55:50  <glx> what says "git describe --tags --long" ?
20:55:52  <DorpsGek_III> [OpenTTD/nml] LordAro merged pull request #113: Rewrite release workflow https://git.io/Jf3tn
20:55:53  <DorpsGek_III> [OpenTTD/nml] LordAro closed issue #111: Versioned source tarball not available through github https://git.io/Jf3IB
20:55:58  * LordAro and runs away
20:56:07  <TrueBrain> Hahahaha, how evil
20:56:15  <LordAro> 0.5.1-RC4-0-gfdb82f2
20:56:28  <glx> ok I know what's wrong
20:56:32  <TrueBrain> Guess the - is an issue? :p
20:56:33  <LordAro> is it --long ?
20:56:42  <glx> it's -RC1 ;)
20:56:46  <TrueBrain> Hahaha
20:56:47  <LordAro> haha
20:56:51  <TrueBrain> Predictable :D
20:57:01  <TrueBrain> We made the same mistakes with OpenTTD
20:57:16  <TrueBrain> It is impossible to remember all those quirks
20:57:17  <LordAro> so if i'd stuck with my old fake tags of 0.5.0alpha1, it would've been fine? :p
20:57:27  <TrueBrain> Hours ago
20:57:53  <glx> LordAro: yes, but we would not have noticed this bug :)
20:58:03  <LordAro> TrueBrain: but at least the checkout is faster now :p
20:58:18  <TrueBrain> Fuck yeah
20:58:32  <glx> so I have a bug to fix now
20:59:47  <TrueBrain> At least you can release 0.6.1 or whatever safely
20:59:57  <glx> true
21:00:16  <andythenorth> we should
21:00:30  <LordAro> do it andythenorth
21:00:45  <LordAro> although maybe #112 needs resolving first?
21:00:53  <TrueBrain> So after all it is frosch123 that ran away .. interesting :D
21:08:10  <blathijs> Yay for resolving #112 too, pretty please? :-)
21:22:44  *** Wormnest_ has quit IRC
21:23:18  <Samu> when is this coming back? https://dev.openttdcoop.org/projects
21:27:38  *** Samu has quit IRC
21:34:40  <andythenorth> will it come back before or after samu? :P
21:34:47  <andythenorth> is it bedtime?
21:35:22  *** nielsm has quit IRC
21:37:16  *** frosch123 has quit IRC
21:39:37  *** Wormnest_ has joined #openttd
21:48:13  *** andythenorth has quit IRC
22:02:14  <DorpsGek_III> [OpenTTD/nml] glx22 opened pull request #114: Fix: tags could contain -RCx https://git.io/Jf33X
22:03:02  *** gelignite has quit IRC
22:12:43  <DorpsGek_III> [OpenTTD/nml] TrueBrain commented on pull request #114: Fix: tags could contain -RCx https://git.io/Jf3sf
22:13:58  <TrueBrain> glx: or does rsplit exist in Python?
22:14:05  <TrueBrain> On mobile, sorry :p
22:14:17  <TrueBrain> rsplit(.., 3) would work too
22:14:24  <TrueBrain> If it exists
22:15:03  <TrueBrain> Ah, seems to exist :)
22:15:09  <glx> but PEP440 doesn't want - :)
22:15:43  <TrueBrain> What ... has that to do with PEP 440? I am talking about the Python code :p
22:16:05  <glx> so anyway I need to convert 0.5.1-RC1 to 0.5.1rc1
22:16:13  <TrueBrain> Honest question btw, I am missing something :)
22:16:32  <TrueBrain> Ah! Well, that at least needs a comment :p
22:17:56  <TrueBrain> I would still suggest an rsplit. After that check for dash in tag, and fix up the tag? Might be easier to understand for future us that way?
22:18:29  <glx> yes would be cleaner to split correctly and fix
22:18:43  <TrueBrain> But please add a comment for sure :D
22:19:04  <TrueBrain> Why do we have so many ways to make versions :p :D
22:27:38  <DorpsGek_III> [OpenTTD/nml] glx22 updated pull request #114: Fix: tags could contain -RCx https://git.io/Jf33X
22:31:08  <TrueBrain> You don't want a comment to state why you do this? (I am fine both ways)
22:36:45  <DorpsGek_III> [OpenTTD/nml] TrueBrain approved pull request #114: Fix: tags could contain -RCx https://git.io/Jf3sX
22:38:53  <DorpsGek_III> [OpenTTD/nml] glx22 dismissed a review for pull request #114: Fix: tags could contain -RCx https://git.io/Jf3sX
22:38:53  <DorpsGek_III> [OpenTTD/nml] glx22 updated pull request #114: Fix: tags could contain -RCx https://git.io/Jf33X
22:39:02  <glx> sorry :)
22:40:29  <DorpsGek_III> [OpenTTD/nml] TrueBrain approved pull request #114: Fix: tags could contain -RCx https://git.io/Jf3sA
22:44:30  <glx> hmm I should quickly try a fake release in my fork, just to be sure :)
22:49:17  *** sla_ro|master has quit IRC
22:54:14  *** supermop_Home has joined #openttd
22:54:53  <DorpsGek_III> [OpenTTD/nml] glx22 merged pull request #114: Fix: tags could contain -RCx https://git.io/Jf33X
23:27:10  *** Wolf01 has quit IRC
23:38:08  <FLHerne> glx: Edge-case with the new version of the versioning
23:38:31  <glx> where ?
23:38:45  <FLHerne> If I've made changes to a tagged version, in a git repository, but without committing
23:39:06  <glx> ah yes there's the m
23:39:07  <FLHerne> It generates a version like "0.42.0m", which setuptools complains is invalid
23:39:19  <FLHerne> I don't know if this is worth caring about
23:39:24  <glx> but it's a modified release so not a release :)
23:39:46  <glx> I considered it's ok like that
23:40:26  <FLHerne> I agree, probably
23:45:08  *** Progman has quit IRC
23:50:16  <LordAro> tbh i don't really care about the non-tagged version being "invalid" at all
23:50:21  <LordAro> it's not like it matters
23:50:36  <DorpsGek_III> [OpenTTD/nml] glx22 opened pull request #115: Fix #112: prevent setup.py regeneration for source package https://git.io/Jf3ZN
23:55:36  <DorpsGek_III> [OpenTTD/nml] glx22 updated pull request #115: Fix #112: prevent setup.py regeneration for source package https://git.io/Jf3ZN

Powered by YARRSTE version: svn-trunk