IRC logs for #openttd on OFTC at 2019-02-09
⏴ go to previous day
00:00:56 <peter1138> Fix: is the prefix, heh.
00:01:03 <peter1138> But the message is a bit wonky now.
00:02:28 <peter1138> So basically the line was ignoring the fact that the train was flagged to reverse.
00:02:28 <andythenorth> Fix: bridges in tunnels should be tunnels in bridges
00:03:03 <m3henry> I had noticed it ages ago but never thought to investigate
00:03:52 <peter1138> This has no bearing on trains not finding a depot when they have a reserved path, does it?
00:04:33 <DorpsGek_II> [OpenTTD/OpenTTD] LordAro approved pull request #7192: Codechange: making the style of MakeVoid calls uniform https://git.io/fhH2y
00:04:49 <peter1138> m3henry, nit-picking but I would rather the blank line was kept.
00:05:28 <DorpsGek_II> [OpenTTD/OpenTTD] LordAro merged pull request #7192: Codechange: making the style of MakeVoid calls uniform https://git.io/fhHUg
00:05:46 <peter1138> Whitespace (including vertical whitespace) helps readability.
00:06:15 <m3henry> I would naturally have omitted the curly braces
00:06:45 <peter1138> Blank line after the closing curly brace.
00:07:33 <m3henry> Anyway, isn't this a minor thing that a reviewer edit would be sufficient for?
00:08:58 <peter1138> Creates a new commit.
00:10:19 <m3henry> Would no curly braces be compliant with the SG?
00:10:46 <peter1138> Nope. It is what it is.
00:11:36 <DorpsGek_II> [OpenTTD/OpenTTD] M3Henry updated pull request #7196: Fix trains pending reversal + ordered to depot not behaving correctly https://git.io/fhH2V
00:12:21 <LordAro> no braces is only allowed when the statement and condition are on the same line
00:12:40 <LordAro> just leads to a goto fail incident otherwise
00:13:57 <Samu> * @param st_ind_id Industry Index of the station that is accepting (such as Oil Rig stations and Oil Rig industries)
00:14:46 <Samu> is this param name confusing?
00:15:46 <Samu> or maybe, how to english
00:16:24 <peter1138> Industry index of the station. What does that even mean?
00:16:58 <peter1138> Is it an index into Station::industries_near?
00:17:08 <nielsm> Samu did you get the writeup of how I'd do that patch earlier?
00:17:19 <Samu> yes, i'm trying to deal with it
00:17:24 <nielsm> because with that method you don't need to alter any function signatures and tiles don't need to know who's asking
00:18:00 <peter1138> LordAro, hmm, looks like the CI's running individual tasks, and the original combined job.
00:18:13 <LordAro> that seems undesirable
00:18:14 <Samu> you see a oil rig industry, it has a neutral station attached to it
00:18:33 <nielsm> oil rigs can end up with no station attached
00:18:44 <nielsm> when there's far too many of them
00:18:59 <LordAro> hmm, it does look that way
00:19:22 <Samu> that isn't supposed to happen :|
00:19:23 <LordAro> wait no, the individual tasks link through to the same build
00:20:06 <peter1138> LordAro, indeed, I guess it is a grouping, but it's not clear :-)
00:21:19 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN approved pull request #7196: Fix trains pending reversal + ordered to depot not behaving correctly https://git.io/fhH2A
00:23:43 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN merged pull request #7196: Fix trains pending reversal + ordered to depot not behaving correctly https://git.io/fhH2V
00:23:56 <Samu> in my good english: * @param st_ind_id Industry Index that the station attached to this industry is corresponded with
00:24:34 <peter1138> That makes no sense
00:24:41 <LordAro> now you're just repeating yourself
00:25:06 <LordAro> a) "ind_id" does not need to be in the variable name, that's what the type is for
00:25:11 <Samu> I got a station, it's an oil rig station, it is attached to an industry
00:25:17 <LordAro> b) "Industry Index" is just repeating the type again
00:25:26 <Samu> or actually, I provide the index
00:25:57 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh commented on issue #7125: Convert singleplayer game to multiplayer server https://git.io/fhH2h
00:26:22 <peter1138> IndustryID associated with Station
00:26:31 <peter1138> IndustryID associated with Station (e.g. oil rig)
00:26:34 <peter1138> Something like that?
00:27:17 <Samu> i will be using it for comparison with other stations that are also neutral, but must ensure they belong to the same industry
00:27:43 <LordAro> or you could pay attention to what nielsm said about it, and not touch the signature at all
00:28:11 <peter1138> Sometimes you gotta write the patch yourself.
00:28:46 <nielsm> you're not getting my idea then
00:29:04 <nielsm> don't tell who's asking and have the answer depend on that
00:29:15 <nielsm> instead avoid asking at all
00:29:32 <Samu> i must compare 2 industry indexes
00:29:36 <nielsm> and have industry tiles whose industry owns a station lie about their acceptance (claim they don't accept anything)
00:30:14 <nielsm> player-owned stations query acceptance as normal, from the surrounding tile area
00:30:48 <nielsm> industry-owned stations do not query acceptance from tiles at all, instead just assume the accepted cargos of the industry it belongs to should be what the station accepts
00:31:29 <nielsm> and with industry tiles belonging to an industry with a station lying and claiming they accept nothing, player-owned stations won't get acceptance from those
00:32:08 <nielsm> but in addition to that, when cargo is delivered on a player-owned station, when deciding which industries to deliver the cargo to, never consider an industry that owns a station
00:32:26 <nielsm> and when an industry produces cargo, if it owns a station only deliver it on that station
00:33:21 <nielsm> this way, industry tiles only need to care about the industry itself they belong to, and never about any stations
00:33:30 <nielsm> (the industry should know whether it has a station attached)
00:34:13 <nielsm> (and you won't need to know which station id that is, just that one exists or does not exist)
00:40:43 <Samu> AddAcceptedCargo_Industry
00:40:56 <Samu> but it's still the same function that computes acceptance either way
00:42:04 <nielsm> you're still not reading what I write
00:42:19 <nielsm> where does station acceptance get calculated?
00:42:35 <nielsm> the answer is station_cmd.cpp line 562 UpdateStationAcceptance
00:42:54 <nielsm> that function knows what station is looking for acceptance
00:42:58 <Samu> it will inevitably reach this function
00:43:06 <nielsm> yes and that's too deep
00:43:26 <nielsm> change UpdateStationAcceptance such that industry-owned stations never ask any tiles at all about acceptance
00:44:15 <Samu> but it has to ask for industry tiles still
00:44:20 <nielsm> then AddAcceptedCargo_* functions don't need to know which station is asking
00:44:22 <Samu> even if it doesn't ask the others
00:44:38 <nielsm> no it does not, it can synthesize the acceptance from the industry itself
00:44:50 <nielsm> not the industry tiles, the industry
00:48:43 <Samu> I can't see this the way you see
00:49:21 <Samu> it has to at least iterate over its own tiles to compute acceptance for the neutral station
00:56:28 <nielsm> okay may not work since Industry::GetByTile requires that the tile is known to be an industry
00:57:14 <nielsm> should be fixed by: if (st->owner == OWNER_DEITY && IsTileType(st->xy, MP_INDUSTRY))
00:57:49 <Samu> but the tile can have acceptance different than that of the industry
00:57:54 *** snail_UES_ has joined #openttd
00:58:22 <Samu> a station around it, says it accepts passengers, but the steel mill itself, doesn't
00:58:24 <nielsm> yes but name one industry that creates stations and has that property
00:58:52 <Samu> but oil rig doesn't accept mail
00:59:03 <Samu> it's the tile that accepts mail
00:59:26 <Samu> it has to iterate the industry tiles
01:00:36 <Samu> one of fishing grounds tile accepts passengers
01:00:45 <Samu> but fishing grounds doesn't accept anything itself
01:01:37 <nielsm> and those examples are something you should have brought up 45 minutes ago
01:01:43 *** tokai|noir has joined #openttd
01:01:43 *** ChanServ sets mode: +v tokai|noir
01:02:06 <nielsm> if you have a proof by contradiction then present the contradiction immediately
01:03:11 <glx> <@peter1138> LordAro, hmm, looks like the CI's running individual tasks, and the original combined job. <-- that's the azure pipelines update, it now reports tasks individually
01:05:56 <nielsm> actually what would be good to have was a better way for industries to indicate they want an attached oil-rig style station
01:06:21 <LordAro> nielsm: istr trying that opens a rather large can of worms
01:07:03 <nielsm> yeah it might cause lots of trouble :/
01:08:33 <nielsm> the simplest would probably be to add another flag to industrytile definitions to control it
01:09:00 <nielsm> and give the currently magical oil rig tile that property, and remove the special-casing by tile type id
01:09:45 <DorpsGek_II> [OpenTTD/OpenTTD] glx22 commented on pull request #7163: Add: [AzurePipelines] split the CI in two parts: building and commit checking https://git.io/fhHai
01:12:11 *** Thedarkb-T60 has joined #openttd
01:30:03 *** snail_UES_ has joined #openttd
01:31:45 <Samu> * @param st_ind IndustryID associated with Station, if any (e.g. oil rig or INVALID_INDUSTRY)
02:16:35 <Samu> I had Lonnpool Shoals accepting Engineering Supplies
02:16:58 <Samu> because there was a Dredging Site nearby that accepted it
02:19:12 <Samu> I no longer pass a bool, now I pass IndustryID!
02:19:41 <Samu> INVALID_INDUSTRY is the new false
02:20:10 <Samu> 'actual industryid of the oil rig' is the new true!
02:23:11 <Samu> now i just got to not pass this around
02:26:28 <Samu> rereading what nielsm said
02:29:28 <Samu> i think there may be a way
02:32:17 <Samu> still needs to run the iterator in GetAcceptanceAroundTiles, but instead of asking through AddAcceptedCargo
02:32:38 <Samu> it asks directly to AddAcceptedCargo_Industry
02:33:45 <Samu> but... AddAcceptedCargo_Industry is part of a proc thingy, it expect a certain number of parameters
02:33:55 <Samu> and I'm adding one to it
02:44:38 <Samu> perhaps I may not need to add an extra parameter, must test
03:27:12 <Samu> now who's the bool expert?
03:28:31 <Samu> wait, maybe this is not it, yet :|
03:29:07 <Samu> must confirm with town tiles if it's correct
03:36:06 <Samu> ah crap it's not correct yet
03:58:31 <Samu> bool expert needed to simplify this code
04:02:48 *** Wormnest has joined #openttd
05:58:27 *** HerzogDeXtEr has joined #openttd
07:09:15 *** rocky11384497 has joined #openttd
08:26:21 <Eddi|zuHause> ok, so i tried launching astroneer, and i get 2fps in the intro screen
09:00:05 *** andythenorth has joined #openttd
09:08:03 <andythenorth> the rule I've found works
09:08:08 <andythenorth> one never replies to SYL
09:08:18 <andythenorth> or, in fact, in any thread where SYL is active
09:12:19 <Eddi|zuHause> if only the forum had a useful ignore function
09:12:51 <Eddi|zuHause> "this person that you ignored has posted here, click to reveal" is not useful
09:14:13 *** Progman has joined #openttd
09:14:55 <peter1138> andythenorth, the info is still useful for other people reading the thread, regardless of whether SYL understands it.
09:15:30 <peter1138> So did you find any issues with your playtesting last night?
09:15:57 <andythenorth> I found that updated NRT Road Hog isn't working
09:16:00 <andythenorth> but the old version does
09:16:13 <andythenorth> currently I'm blaming mercurial for that
09:18:18 <andythenorth> why is mercurial so totally useless?
09:20:44 <andythenorth> using it for branching workflow is absolutely dangerous
09:20:52 <andythenorth> found the issue I think
09:21:30 <andythenorth> I should just move everything to github ASAP
09:21:41 <andythenorth> but then I lose bundles server publishing
09:25:19 <andythenorth> yeah merging default into notroadtypes branch of hog
09:25:20 <peter1138> andythenorth, move to git primarily, and then when you want an updated build, just copy everything over the hg repo.
09:25:27 <andythenorth> has basically blitzed the notroadtypes changes
09:25:42 <andythenorth> 'oh this is different to default, I'll just replace it'
09:25:48 <andythenorth> mercurial is a twat
09:40:39 <peter1138> Might do some NRT squashing later.
10:04:37 <Eddi|zuHause> andythenorth: surely bundles can be set up to a git repo
10:05:06 <andythenorth> I'm 99% certain it can
10:05:12 <andythenorth> but I haven't done the work yet
10:06:27 <andythenorth> can we fix the issue with the UI not updating in pause mode?
10:11:08 <andythenorth> oof now I have to compile master to check it
10:15:56 <DorpsGek_II> [OpenTTD/OpenTTD] andythenorth opened issue #7197: Vehicle UI not showing clone button for newly created vehicles in depot when paused https://git.io/fhHoD
10:16:18 <andythenorth> I've got a £0.02 bet on what causes that
11:08:17 *** m3henry has joined #openttd
11:08:50 <andythenorth> peter1138: eh Pikka's AI is building HAUL road pieces :D
11:08:56 <andythenorth> so it has broken lots of towns :P
11:09:16 <andythenorth> eh no, it's the town actually
11:10:25 <Wolf01> How it could do it? Did you apply the patch for roadsubtype support for towns?
11:10:32 <m3henry> I'm going to have to ask, what does "o/" mean?
11:10:54 <andythenorth> well Wolf01, I've changed grfs on a running gmae
11:11:01 <andythenorth> and the newgrf might be broken anyway :P
11:11:21 <Wolf01> <m3henry> I'm going to have to ask, what does "o/" mean? <- waving hello
11:11:33 <LordAro> m3henry: generic wave emoticon
11:12:16 <m3henry> That one might bee o/o/o/o/o/
11:12:37 <Wolf01> But that looks a cyclope
11:14:45 <Eddi|zuHause> <Wolf01> Or nazi hail.. <-- assuming that the character depicted is facing you, that's the wrong arm for that :p
11:15:30 <Eddi|zuHause> i'm not going down this road
11:15:32 <Wolf01> Oh, yes, it's a silhouette, you can't tell if it's facing you or it's the back
11:16:48 <Eddi|zuHause> there's this rotating silhouette illusion that you can view as either left or right turning
11:18:24 <Wolf01> I know, that's wonderful, every time I show it to someone it says the video is changing direction
11:20:01 <m3henry> Looking at town_road.png. It seems it would be a good idea to separate the road surface sprite from the grass sprite and then render the road on top of the grass to get the final road tile for the grass-desert/snow transition pieces
11:21:50 <nielsm> (it's very exploitable)
11:22:41 <Eddi|zuHause> nielsm: with crazy non-rectangular stations (or with huge gaps) you can make a station that can deliver to the industry but not receive any cargo back
11:23:11 <Wolf01> Or you can cover an entire city with just 2 tiles
11:23:22 <Eddi|zuHause> nielsm: as the industry will search for station tiles in its rectangle (using the station radius)
11:23:57 <Wolf01> I never understood that...
11:24:46 <Eddi|zuHause> with non-rectangular industries you can also get funny effects
11:24:47 <Wolf01> It makes sense because it's the industry which wants to push the goods
11:26:30 <nielsm> stations are weirder than the sum of their parts
11:28:10 <andythenorth> hmm, pax trains are printing money
11:28:22 <andythenorth> I tried nerfing pax costs, but I've hit the limit
11:28:26 <Wolf01> Strange, I have the opposite problem
11:28:33 <andythenorth> Wolf01: different game? :P
11:28:45 *** frosch123 has joined #openttd
11:29:41 <andythenorth> maybe I nerf pax capacity more
11:32:37 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh updated pull request #6965: Add: Option for population-linear town cargo generation https://git.io/fpkqa
11:42:06 <andythenorth> so the thing with pax is that it loads both legs of an A-B route
11:42:19 <andythenorth> whereas freight only loads A-B and B-A is deadhead
11:42:43 <andythenorth> I've knocked freight engine run costs down by 33% to account for this
11:42:51 <andythenorth> but freight rates are much lower than pax also
11:44:31 <Wolf01> I would play TF, but in this game I can't kickstart and I got bored :(
11:45:04 <Wolf01> I'm "surviving" vehicles repay themselves and make like 100$/year
11:45:27 <Eddi|zuHause> you have like 10M loan?
11:45:47 <Wolf01> I took the full loan to kickstart
11:46:05 <Eddi|zuHause> you need a real moneymaker route at the start
11:46:25 <Eddi|zuHause> in my game that was ships
11:47:04 <andythenorth> I could nerf pax cars with high run costs
11:47:11 <m3henry> I've found RRT2/3 to have an excellent economic model for this sort of thing
11:47:24 <Wolf01> But also a route which doesn't clog the production, like if you start with steel you end up building the entire network because industries stockpile byproducts and stop production
11:47:28 <andythenorth> it's integrated, no grfs m3henry
11:47:48 <m3henry> I still make my point
11:47:48 <Eddi|zuHause> yes, you need to do a full network for freight
11:49:00 <Eddi|zuHause> ships are great early game because you don't need much infrastructure and they're rather cheap, but they're also bad for a full freight network, as they can only carry two cargos at a time
11:49:28 <Wolf01> The only freight you can rely is food, I'm producing a lot of food and deliver to all the neighbour cities, but it doesn't pay much, also I have passenger routes which are used by.... 4 people
11:49:56 <andythenorth> imagine ships with more than one cargo :o
11:50:03 <Eddi|zuHause> passengers are a bit time sensitive, and you need good internal connections in the town
11:50:35 <Eddi|zuHause> andythenorth: later ships can carry 4 cargo types, but by then the routes are so full that it's impossible to have enough ships
11:51:23 <Eddi|zuHause> ship route capacity is limited by docking/undocking procedure, which takes ages
11:51:57 <Wolf01> That's why I play with 4x time
11:52:30 <Wolf01> So I can play fast forward and still take ages to pass a year :P
11:55:19 <Eddi|zuHause> dancer still rotating in the same direction...
11:55:23 <Eddi|zuHause> i made it work once...
12:12:33 *** gelignite has joined #openttd
12:18:19 *** drac_boy has joined #openttd
12:19:33 <drac_boy> just curious how any of you feel about locomotives changing over time? (same vehid in list but slight different spec that is)
12:20:02 <Eddi|zuHause> bad idea, messes with autoreplace
12:24:07 <drac_boy> yeah I thought as much, was wondering about the grftables that seem to do that
12:24:18 <drac_boy> ah well, just had to ask :)
12:32:17 <drac_boy> anything interesting or just same old for now eddi?
12:56:23 <andythenorth> eh I should turn breakdowns on
12:57:26 <andythenorth> ha ha that ruins my network
12:57:36 <andythenorth> servicing was disabled, so everything is 0%
13:03:01 <drac_boy> andy I'm just asking but generally how hard was it for you to code that thing where one industry can come in different sprites matched with randomized outputs?
13:04:14 <Eddi|zuHause> <andythenorth> servicing was disabled, so everything is 0% <-- we might want to disable reliability decay if servicing is disabled?
13:13:42 *** synchris has joined #openttd
13:14:22 <DorpsGek_II> [OpenTTD/OpenTTD] Eddi-z opened pull request #7199: Change: skip reliability decay if servicing is disabled https://git.io/fhH61
13:14:35 <Eddi|zuHause> andythenorth: care to test? :p
13:15:21 <Eddi|zuHause> 4 cases to test: breakdowns on/off, service with no breakdowns on/off
13:15:35 <TrueBrain> okay, that update of Azure Pipelines is pretty nice .. that is much more verbose
13:15:59 *** HerzogDeXtEr has joined #openttd
13:17:38 <TrueBrain> LordAro: what are the odds on a beta this weekend? :D
13:18:05 *** Thedarkb-T60 has joined #openttd
13:18:06 <LordAro> well the PR has been approved...
13:18:33 <LordAro> as soon as that's been merged, it can be tagged, afaik
13:19:17 <TrueBrain> lemme know before you tag and everything else is in place; means we can do some dryruns to see how it looks :)
13:19:23 <TrueBrain> (removing tags is such an annoying thing to do :P)
13:19:43 <LordAro> i'm not planning on tagging anything myself
13:20:05 <TrueBrain> well, first we need to merge everything that is needed anyway :D
13:20:15 <TrueBrain> anything I can help with?
13:20:59 <Eddi|zuHause> on the chance of derailing your focus, any thoughts on enabling downloading CI builds for the "i can't compile myself" testing cases? (possibly with dev approval?)
13:21:36 <LordAro> other than that, unknown
13:21:44 <TrueBrain> Eddi|zuHause: downloading CI builds: never (infosecs worst nightmare). Seperate on-demand request to kick of a 'nightly' like build based on PR code: sure, we should do that
13:21:51 <TrueBrain> but in my list of things it is number ... 10? :)
13:21:59 <TrueBrain> so if you want it quicker, someone will have to stand up and help :)
13:22:25 <Eddi|zuHause> i don't think i'll figure out how this compile farm thing works any time soon
13:22:30 <TrueBrain> LordAro: I like how you predated that :P
13:22:38 <TrueBrain> Eddi|zuHause: not with that attitude :D
13:23:05 <Eddi|zuHause> i mean, not without someone giving me a step by step introduction
13:23:29 <LordAro> TrueBrain: it was originally last sunday :p
13:23:33 <TrueBrain> LordAro: oeh, new release notes :D
13:23:35 <Eddi|zuHause> instead of a "here, this is a result after 20 steps, go on"
13:23:51 <LordAro> TrueBrain: and i presume you've seen how much nicer the individual checks are now?
13:23:54 <TrueBrain> Eddi|zuHause: yeah, as always, the issue of someone telling you that step by step, means someone could also have done it him/herself ;)
13:24:05 <TrueBrain> LordAro: yeah .. happy they finally deployed that version
13:24:12 <TrueBrain> it looks so much better
13:24:22 <Eddi|zuHause> TrueBrain: it's worse, that person could have done it multiple times
13:24:36 <TrueBrain> Eddi|zuHause: but both me and nielsm just started trying things out and the result is now in the repo
13:24:41 <TrueBrain> it is also a matter of just doing, and trying things
13:25:01 <TrueBrain> it is easy to want someone else to shew your food; but in the end, that is not how we progress as species :D
13:25:19 <TrueBrain> but I understand that is is difficult Eddi|zuHause :)
13:25:52 <TrueBrain> Eddi|zuHause: given the link LordAro just gave me, I say: do not do ANYTHING till the feb 4th version of Azure Pipelines is deployed :D
13:26:00 <TrueBrain> that solves 80% of the hard work required to make it happen
13:26:05 <Eddi|zuHause> TrueBrain: the thing is, i'm much more of a high-level-problem kind of mindset, and every time it devolves into the nitty-gritty details, i get stuck
13:26:27 <TrueBrain> Eddi|zuHause: so we have to find other people who are willing to take on these jobs :D
13:28:07 <peter1138> 10:08 < andythenorth> peter1138: eh Pikka's AI is building HAUL road pieces :D
13:28:17 <peter1138> Not possible unless HAUL is road type index 0.
13:28:47 <TrueBrain> Eddi|zuHause: but in this case, seriously, if that Azure Pipelines version is deployed, it is like 1 hour work for me; so I can prioritize that if that version is deployed :)
13:29:03 <TrueBrain> that addition is really awesome .. Azure Pipelines is really understanding what GitHub projects need :D
13:29:04 <andythenorth> peter1138: yeah no it's the town
13:29:15 <TrueBrain> LordAro: can I merge it too?
13:29:30 <andythenorth> so I should fix that
13:29:54 <LordAro> TrueBrain: if you're actually happy with it :p
13:30:07 <LordAro> i still think the changelog needs cutting down a bit further, but i'm not sure how
13:30:26 <peter1138> andythenorth, add the "town can build" flag
13:30:38 <TrueBrain> LordAro: I have no opinion about it, what-so-ever
13:30:41 <peter1138> andythenorth, if it's index 0 it would've always done that.
13:30:52 <peter1138> andythenorth, bit 3 in the road type flags means "town can build this"
13:31:17 <peter1138> if you have introduction dates, that also affects it.
13:31:21 <TrueBrain> frosch123: do you want to give the changelog a second look, or are you also #dontcare ? :D
13:32:05 <planetmaker> <TrueBrain> (removing tags is such an annoying thing to do :P) <-- we could simply tag alphas ;)
13:32:37 <TrueBrain> planetmaker: or just let me do my thing before we tag :)
13:33:13 <TrueBrain> GitHub Actions also start to become more available .... that is a nice addition to Azure Pipelines .. \o/
13:35:42 <LordAro> TrueBrain: what would that mean?
13:36:29 <TrueBrain> but mostly, it can also trigger things like deployments
13:36:40 <TrueBrain> currently we do that via Azure Pipelines, but that is a bit annoying
13:36:46 <TrueBrain> integration with GitHub would be ... very awesome
13:37:05 <TrueBrain> (it is a docker based action system)
13:37:11 <LordAro> and people said MS buying GH would be a bad thing :p
13:37:22 <TrueBrain> so far, they are not disapointing
13:38:05 <TrueBrain> flake8 and stuff can be done much more easily
13:38:32 <peter1138> andythenorth, hmm, unless you have bit 3 set on HAUL for some reason
13:39:03 <peter1138> andythenorth, have we lost some nrt changes at some point...? cos bit 2 & 3 were not used, but I've seen one NRT set use bit 3.
13:39:42 <TrueBrain> that action would be awesome :D
13:40:54 <LordAro> wait hang on, that 404s
13:41:10 <peter1138> How long has George been playing this game... #7198 :p
13:41:58 <TrueBrain> I thought they were public
13:42:01 <TrueBrain> I signed up for the beta
13:42:22 <TrueBrain> so basically, you can type: /rebase, in a PR
13:42:24 <TrueBrain> and it rebases the PR
13:42:35 <peter1138> andythenorth, what was that about locks?
13:43:18 <Wolf01> Ok, with ships I'm losing $250k/year, without I'm earning $20k/year
13:44:05 <TrueBrain> just too bad you cannot see which version of Azure Pipelines is deployed
13:45:09 <drac_boy> wolf01 sounds like imcomptent orders than actual ship profit problem to me? dunno tbh :)
13:46:29 <Wolf01> People don't like to use ships, using buses instead
13:47:33 <frosch123> i need to look up some static code analysers for python
13:48:09 <frosch123> restarting eints for every single typo is mehj
13:48:21 <Eddi|zuHause> <peter1138> How long has George been playing this game... <-- dunno, but loads of people have been finding stuff and went "wtf? that has been there for the last 10 years"
13:48:30 <LordAro> frosch123: pylint, pycodestyle (formerly pep8), flake8
13:49:00 <Eddi|zuHause> and honestly, the way catchment area works is crazy and should be fundamentally reworked
13:49:31 <LordAro> Eddi|zuHause: sounds like an OTTD2.0 thing
13:50:09 <Eddi|zuHause> also, more things in this game should be euclidean
13:50:10 <LordAro> frosch123: i believe so, the translation should've been invalidated
13:50:12 <drac_boy> eddi heh, sadly I also still wonder how many people are "its been 10 years why is a simple feature still completely missing?" type .. but what do I know tho
13:50:16 * drac_boy is one of these mind you
13:50:32 <frosch123> LordAro: they were invalidated, but eints does not delete invalid translations
13:50:36 <peter1138> frosch123, I think so.
13:50:47 <peter1138> frosch123, we had deleted them manually, but eints put them back.
13:50:53 <frosch123> LordAro: with that diff eints has more info than the repository
13:52:14 <frosch123> peter1138: that is not how eints works. eints always writes a whole langfile. it does not make "updates"
13:52:22 <andythenorth> peter1138: screenshot :P
13:54:21 <drac_boy> thats HEQS right? :p
13:54:31 <drac_boy> only released freight tram I recall of afaik :)
13:54:49 <andythenorth> there are a bunch of tram grfs
13:55:31 <peter1138> frosch123, i'm not sure what you mean, but it definitely but back the bad translations which we removed.
13:55:48 *** Thedarkb1-T60 has joined #openttd
13:55:59 <frosch123> peter1138: yes, and that is by design
13:59:54 <TrueBrain> haha, I found a nasty bug: if there is a commit after the last nightly but before the date changes, the nightly version is a bit weird
14:00:12 <TrueBrain> for example, the last nightly version was called '20190207-master-g5e4f76f2f9', but produced on 20190208
14:01:07 <drac_boy> andy really? never seen any that wasn't just HEQS but hm
14:01:23 <TrueBrain> "Latest release in master is 20190207-master-g5e4f76f2f9, released on 2019-02-08 19:01 UTC."
14:01:32 <TrueBrain> it is not wrong, but also not really right
14:01:34 <peter1138> andythenorth, yes, I saw the picture, but what's the point of it?
14:05:04 <TrueBrain> we should add dorpsgek to it :)
14:05:11 <peter1138> If eints puts back invalid stuff we've removed, that's eints bug ;p
14:05:19 <TrueBrain> frosch123: 2 sentence son the first sentence? That is not git-like :D
14:05:38 <Eddi|zuHause> TrueBrain: but using the build date instead of the commit date doesn't make any sense either
14:05:45 <frosch123> TrueBrain: do you want one or two \n ?
14:06:05 <TrueBrain> frosch123: git best practice says: <line>\n\n<more info which might include a \n>
14:06:05 <LordAro> Eddi|zuHause: doesn't it? the build will not be produced if there are no new commits
14:06:11 <andythenorth> peter1138: can't build a lock under a bridge :P
14:06:13 <TrueBrain> Eddi|zuHause: there is no solution, I think :D
14:06:58 <Eddi|zuHause> TrueBrain: well, we could do "version master-gXYZ built on blah-mm--dd"
14:07:27 <TrueBrain> Eddi|zuHause: I was more wondering about ingame
14:07:32 <TrueBrain> on 2 days you see the same 'date'
14:07:36 <TrueBrain> for a nightly, this might be confusing
14:07:37 <peter1138> Did we add nielsm's versioning changes yet?
14:07:54 <Eddi|zuHause> i think we included the changes for multiplayer version
14:07:58 <frosch123> oh, eints repo does not require reviews
14:08:13 <frosch123> do you need to setup this stuff for every single repository?
14:08:19 <TrueBrain> frosch123: I think so
14:12:56 <drac_boy> ehhh hmm just have to ask but I mean I know a boxcab is a boxcab .. but does eg a SBB Re 460 still fall under that termiology too or it would had rather gone by a different name?
14:14:15 <peter1138> We don't use that term at all, so...
14:15:39 <Eddi|zuHause> i'm not familiar with that term either
14:17:40 <andythenorth> I will never win this silicon valley game :P
14:17:45 <andythenorth> spending too much time choosing liveries
14:17:54 <TrueBrain> frosch123: just to avoid wrong assumptions: you will also bump eints-openttd and update the source on the server, and restart eints? or can I help with any of it?
14:18:02 <Eddi|zuHause> andythenorth: you know the pause key, right?
14:18:45 <frosch123> TrueBrain: yes, rebase eints-openttd. however, i don't think i have enough permissions on th eserver
14:19:03 <TrueBrain> frosch123: let me know if/where I can help out
14:20:09 <frosch123> so, is there a gh way to do the branch-rebase, or is is force-push time?
14:20:27 <TrueBrain> rebased are always force pushes
14:20:38 <TrueBrain> rebases are always force pushes
14:20:56 <TrueBrain> this is a git thing; so github doesn't have anything for it :)
14:25:10 <frosch123> TrueBrain: ok, i pushed the branch
14:25:22 <frosch123> on the server there is still a hg clone :)
14:25:28 <TrueBrain> you sure you pushed it?
14:25:57 <TrueBrain> that tree overview is misleading
14:26:02 <TrueBrain> shows the author date I guess :D
14:26:07 <Eddi|zuHause> concerning NRT and level crossings: can we have separate ways to specify the lower and upper half of the crossing barriers? i still have this "make level crossings work as one unit" and "diagonal crossings" patch lying around that always lacked graphics support
14:26:30 <TrueBrain> frosch123: do you want me to fix the server, or will you? (just to avoid us both working on it)
14:26:55 <frosch123> i failed to su translators :)
14:27:08 <frosch123> no idea how to do it otherwise
14:27:18 <TrueBrain> su to root first :)
14:27:55 <peter1138> sudo -i -u translators?
14:28:27 <TrueBrain> let me create a backup first frosch123 :)
14:31:34 <frosch123> oh, wait... cloning eints does not take as long as cloning openttd :p
14:31:47 *** sla_ro|master has joined #openttd
14:32:04 <TrueBrain> dont forget to copy the user.cfg :)
14:36:15 *** Thedarkb1-T60 has joined #openttd
14:43:24 <TrueBrain> nice work frosch123 :)
14:43:28 <frosch123> triggered a push, looks good
14:47:10 <TrueBrain> curious what the branch name is if I triggered a 'nightly'-like build on a pull request ..
14:48:17 <TrueBrain> "20190208-head-gc124344005"
14:48:21 <TrueBrain> that is not ... useful :D
14:48:54 <nielsm> version numbers are all in your head...?
14:49:48 <TrueBrain> " git checkout -B ${BUILD_SOURCEBRANCHNAME}"
14:50:41 <TrueBrain> so how can I queue a PR in the same way as Azure does via the GitHub app .. hmmm
14:59:20 *** snail_UES_ has joined #openttd
15:04:20 <peter1138> So many snacks in my pantry
15:04:58 <TrueBrain> happy you say pantry, not pants
15:05:34 <TrueBrain> "Merge c1243440052e19b4eb846b90f8bb8b0bd1b32dbd into de1278290b7ede2fa9e4a7455abfeb8487496dcf" <- well, this is not really helping now is it? :D
15:08:30 <frosch123> but i guess that does not translate to english
15:11:15 <frosch123> it's different than before, but i guess that's due to git :)
15:12:08 <frosch123> issue numbers in front instead of in back. but no idea why the old changelog did that
15:12:18 <frosch123> anyway, the worst commit message i found was my own :p
15:13:25 <frosch123> about the pr: all files that need changing were changed, so that is fine
15:13:48 <TrueBrain> known-bugs.txt is the only odd-one-out, I guess
15:13:52 <drac_boy> which two would you had prefer to use more of the two? 'bogie' or 'truck' re non-2axle wagons
15:13:53 <TrueBrain> already at 1.9.0, with wrong date
15:14:07 <drac_boy> I did think of articulated but it seem to sound very misleading at the end
15:15:31 <TrueBrain> LordAro: do you have the time and would you mind updating known-bugs.txt too, and set the dates to today (instead of tomorrow)? :)
15:15:43 <TrueBrain> and the commit message is weird
15:15:59 <TrueBrain> "Update: add changelog for 1.5.0-beta1 and prepare for release" ?
15:16:38 <frosch123> LordAro: in that case, i would like to update an entry in the changelog :p
15:17:07 <frosch123> i had no idea what it meant, though apparently i made it :p
15:17:34 <TrueBrain> "BUILD_SOURCEBRANCHNAME=merge" <- well ... that is not useful
15:17:59 <TrueBrain> tnx for the poke glx; it indeed improved a lot (Azure Pipelines) :)
15:18:33 <glx> yes and the warning count seems more accurate on the checks page
15:18:45 <TrueBrain> no; only for Windows builds still :P
15:19:07 <glx> well at least it shows 40 and not 11 :)
15:21:03 <Samu> the bool expert is asleep
15:27:59 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain opened pull request #7201: Fix: [AzurePipelines] in case of a Pull Request, use that number (prefix with 'pr') as branchname https://git.io/fhHPG
15:30:04 <TrueBrain> hmm .. why doesn't that work ..
15:30:32 <TrueBrain> that variable is only set when triggered from Github
15:32:04 <TrueBrain> so to also support 'manual' builds, it needs more magic
15:35:06 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain updated pull request #7201: Fix: [AzurePipelines] in case of a Pull Request, use that number (prefix with 'pr') as branchname https://git.io/fhHPG
15:37:21 <TrueBrain> Eddi|zuHause: ^^ that should make possible to publish 'nightly'-like binaries for Pull Requests :) (after Azure Pipelines bot updates to allow the building-from-comments ability :P)
15:37:32 <TrueBrain> just every PR first needs to rebase after this is merged before that works :P
15:39:01 <TrueBrain> glx: what part exactly?
15:39:19 <TrueBrain> yeah; doesnt hurt, but can be removed
15:39:22 <TrueBrain> please make a PR :)
15:41:29 <Samu> why am i getting mails about 7170?
15:42:19 <glx> you commented in it maybe
15:42:47 <Samu> what the heck? I posted this years ago, it didn't post it?
15:43:12 <Samu> forget it, i cancelled review
15:43:37 <LordAro> it's quite easy to forget to actually complete a review
15:44:03 <TrueBrain> they are pretty good in improving their UI :)
15:45:04 <peter1138> Right, now to add some compatibility functions...
15:47:11 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain requested changes for pull request #7202: Update: NSIS is now part of the Hosted image https://git.io/fhHPg
15:47:14 <peter1138> Hmm, do I have to update all the bin/*/compat_*.nut files? Or is there some source elsewhere?
15:48:16 <TrueBrain> otherwise these commits are rather confusing in my opinion :)
15:48:26 <glx> yeah I forgot that part :)
15:48:30 <LordAro> peter1138: all of them
15:48:54 <Samu> i opened them in visual studio then paste paste paste in every one of them
15:49:22 <peter1138> I've made way more changes than you ;(
15:49:26 <DorpsGek_II> [OpenTTD/OpenTTD] glx22 updated pull request #7202: Update: NSIS is now part of the Hosted image https://git.io/fhHPz
15:49:59 <glx> peter1138: maybe a new script :)
15:50:03 <peter1138> But then again I just a Galaxy bar. (Albeit a mini celebrations one)
15:50:03 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain approved pull request #7202: Update: NSIS is now part of the Hosted image https://git.io/fhHPV
15:50:12 <TrueBrain> glx: does that mean you review my change? :D
15:50:49 <Samu> how do I unsubscribe to 7170? still receiving mails
15:51:07 <glx> well I read them TrueBrain :)
15:52:27 <Gabda> Samu: in the meanwhile I've subscribed you to another PR :)
15:53:55 <DorpsGek_II> [OpenTTD/OpenTTD] glx22 approved pull request #7201: Fix: [AzurePipelines] in case of a Pull Request, use that number (prefix with 'pr') as branchname https://git.io/fhHPo
15:54:11 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain merged pull request #7201: Fix: [AzurePipelines] in case of a Pull Request, use that number (prefix with 'pr') as branchname https://git.io/fhHPG
15:55:25 <TrueBrain> peter1138: would you mind rebasing NRT soon-ish? :D
15:56:05 <glx> you want to test your change ?
15:56:19 <peter1138> AIs broken at the moment until I write the compat stuff.
15:56:35 <TrueBrain> well, it is tested; I just want to make it useful :P
15:56:46 <TrueBrain> peter1138: I just want to try to build binaries out of them which we can publish :)
15:57:10 <Samu> im renaming on_water to st_ind
15:57:11 <peter1138> Actually I can stash these changes and push that.
15:57:56 <peter1138> No rebase conflicts, phew.
15:58:59 <TrueBrain> Azure Pipelines will be busy for a few :D We might have a lot of jobs queued :P
15:59:22 <glx> CI should not try to build if src or projects are untouched ;)
15:59:44 <glx> but just check the commit
16:00:23 <TrueBrain> feel free to add that logic :D
16:00:31 <TrueBrain> but I am guessing it gets out of hand REALLY quick :P
16:00:50 <TrueBrain> so .. I think it is okay to take the penalty ;)
16:00:59 <Samu> second part, no changes needed
16:01:17 <Samu> third part, not done, don't think it's needed
16:01:20 <glx> yeah we rarely commit something not touching the code anyway
16:02:49 <Samu> down from touching 17 files to 13
16:02:50 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain merged pull request #7202: Update: NSIS is now part of the Hosted image https://git.io/fhHPz
16:03:36 <Samu> could go further down to 12 if I remove static inline bool IsOilRigTile(TileIndex t) { return IsTileType(t, MP_STATION) && IsOilRig(t); }
16:04:33 <glx> hmm should I try to merge generate_widget and squirrel_export ? (or at least run both from a single script)
16:05:41 <TrueBrain> 'Version: 20190209-pr6811-g0b1ef5244a'
16:05:43 <TrueBrain> oeh, that seems to work :D
16:06:07 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN commented on pull request #7120: Codechange: Improve performance of closest town lookups with cache https://git.io/fhHPd
16:08:43 <peter1138> TrueBrain, you realise if you release an NRT test build, and a 1.9 beta at the same time, probably nobody will test the 1.9 beta? ;)
16:09:13 <TrueBrain> the beta is also more to make sure the infrastructure is correct ;)
16:10:51 <Gabda> peter1138: I'll look into it. I am not familiar with the std::vector yet.
16:11:16 <Gabda> but standard classes are easy to use usually
16:11:41 <peter1138> I might be wrong, mind you.
16:12:03 <peter1138> Opinions here, new code, just use std::vector etc instead of our custom implementations?
16:12:52 <peter1138> I did wonder if you could just use FOR_ALL_TOWNS throughout the code but it doesn't look that simple.
16:18:36 <Samu> what should the behaviour be for a water industry under construction that provides its own station but the game runs out of stations to create one?
16:19:05 <Samu> treat the industry as just a regular industry without station?
16:19:18 <peter1138> Samu, industry should not be generated.
16:19:36 <Samu> it is under construction already
16:19:44 <peter1138> Station should be allocated before construction starts.
16:20:28 <peter1138> Sounds like a PR ;)
16:20:36 <peter1138> Or at least an issue.
16:20:58 <peter1138> I guess this is a "hit 65534-or-so stations" thing?
16:21:21 <Samu> easy to achieve for ais that mass buoys
16:21:23 <peter1138> Hmm, so in the code I was modifying ScriptRoad. Does this become AIRoad in the compat stuff?
16:22:16 <planetmaker> I think it does. AI:.. derives from script classes
16:24:30 <frosch123> peter1138: imho std::vector all the way
16:24:48 <peter1138> frosch123, 's what I thought :D
16:25:09 <Gabda> the long term plan is to change smallvec to std::vector behind GUIList, or leave behind GUIList entirely?
16:25:11 <frosch123> ScriptRoad is turned into AIRoad and/or GSRoad depending on the @api in the doxygen
16:25:51 <frosch123> Gabda: the idea of smallvec is copy-elision, which is obsolete with c++11
16:26:24 <Samu> if (IsTileType(other, MP_STATION) && IsOilRig(other) || IsTileType(other, MP_INDUSTRY) && GetIndustryGfx(other) == GFX_OILRIG_1 && i->index == GetIndustryIndex(other) && !IsIndustryCompleted(other)) {
16:26:49 <nnyby> i'm playing on git master and it seems like shift is fast-forwarding the game instead of tab, is this intentional?
16:27:00 <drac_boy> samu not related but that certain uk-style very-flat map that was in tt/ttdx .. it was easy to nearly run out of locomotives to the point that if you only had one slot left but bought an asiastar/tim, you would only get the head part alone :-)
16:27:06 <frosch123> nnyby: that's the case for debug builds
16:27:10 <Samu> then it assumes the best, that it will get a oilrig station
16:27:36 <frosch123> originally because people on 2005's single-screen setups wanted to alt+tab to their editor/ide without ff
16:27:38 <Samu> but if it's completed, and no station is already there, then returns false
16:29:23 *** Eddi|zuHause has joined #openttd
16:29:23 *** TinoDidriksen has joined #openttd
16:29:23 *** OsteHovel has joined #openttd
16:29:23 *** michi_cc has joined #openttd
16:29:23 *** Hazzard has joined #openttd
16:29:23 *** TrueBrain has joined #openttd
16:29:23 *** Webster has joined #openttd
16:29:23 *** XeryusTC has joined #openttd
16:29:23 *** Terkhen has joined #openttd
16:29:23 *** ^Spike^ has joined #openttd
16:29:23 *** Hirundo has joined #openttd
16:29:23 *** coulomb.oftc.net sets mode: +vov michi_cc Terkhen Terkhen
16:29:43 <peter1138> Cool, failed to compile compat_1.8.nut, and openttd crashes :D
16:29:49 <TrueBrain> owh, no, they are back
16:30:11 <TrueBrain> ubuntu bionic deb seems to work, as does the windows 64bit
16:30:15 <LordAro> TrueBrain: nah, you are back
16:30:22 <TrueBrain> would be nice if someone can check the OSX too
16:30:57 <TrueBrain> please check if you see anything weird etc
16:32:19 <TrueBrain> (there is an Artifact link on these pages, where you can download the result)
16:32:46 <Samu> down to 12 files touched
16:33:08 <peter1138> "Your script made an error: the index 'IsRoadTypeAvail' does not exist"
16:34:58 <peter1138> Now to C&P and hope.
16:35:12 <peter1138> Hmm, can we use includes in squirrel .nut files?
16:36:19 <drac_boy> hm huh I didn't think..should include one or two dual-end locomotives just for player's fun too
16:36:42 <peter1138> TrueBrain, it runs :D
16:36:53 <peter1138> 20190209-pr6811-g0b1ef5244a
16:40:48 <frosch123> noai.openttd.org seems to have some bonus magic
16:40:57 <frosch123> "main page" links to data structures
16:41:08 <frosch123> and "data structures" has some extra text links
16:43:04 <TrueBrain> now it needs a nice place on www.openttd.org :)
16:43:43 <TrueBrain> hmm .. possibly we want this per PR or something
16:44:18 <frosch123> i would not adverise random prs builds on openttd.org
16:44:29 <peter1138> ocal cost = AIRail.GetBuildCost(roadType, AIRail.BT_TRACK)
16:44:35 <peter1138> Huh, well, that's a bug in the AI.
16:44:55 <TrueBrain> frosch123: no, but it is the place that makes a pretty HTML page for it, so it is easier to find your download
16:45:03 <TrueBrain> in the future we can even publish it as comment in the PR
16:45:09 <TrueBrain> but .. someone needs to figure out how that works
16:45:14 <TrueBrain> as I have no clue currently :D
16:45:29 <frosch123> sounds like a job for DorpsGek_III
16:46:10 <TrueBrain> okay, I think releases should work too
16:46:17 <TrueBrain> at least, I do not see anything that would stand in the way ..
16:46:37 <TrueBrain> except an automated trigger, I guess
16:46:41 <TrueBrain> that should be doable
16:47:49 <peter1138> Ok, AIs are building roads :D
16:50:41 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain opened pull request #7203: Add: [AzurePipelines] build a stable release if a tag is created https://git.io/fhHXo
16:51:07 <peter1138> Bah, can't use require.
16:52:32 <peter1138> Can if you use the right syntax ;?
16:55:49 *** andythenorth has joined #openttd
16:56:04 <TrueBrain> andythenorth: can you help me out with some OSX testing? :D
16:57:12 <TrueBrain> especially the first is kinda important that it "just works" :D
16:58:56 <peter1138> ^ AI support. I hope.
16:58:57 <andythenorth> ok I'll test shortly
16:59:14 * andythenorth house full of people asking for roblox credits
16:59:16 <peter1138> Obviously current AIs don't do anything with it.
16:59:27 <peter1138> They can only build the first road/tram type.
16:59:31 <TrueBrain> andythenorth: tnx; no rush etc :)
16:59:32 <TrueBrain> Please be careful, and don't use these binaries for long-term gameplay.
16:59:38 <TrueBrain> is that a good text for pull request binaires?
17:01:20 <TrueBrain> "Latest release in pr6811 is 20190209-pr6811-g0b1ef5244a, released on 2019-02-09 15:05 UTC." :)
17:02:34 <Samu> 'continue' is better than 'else'
17:04:55 <DorpsGek_II> [OpenTTD/website] TrueBrain opened pull request #50: Add: support openttd-releases and openttd-pullrequests on new infrastructure https://git.io/fhHXx
17:05:51 <TrueBrain> okay ... checklist: OSX, PR7203 approved, PR50 approved. If those 3 are done, I think we can release 1.9.0-beta1 :)
17:05:55 <andythenorth> TrueBrain: so I found a huge zip of all binaries
17:06:01 <andythenorth> "openttd-20190209-master-g8c0bfb4637-macosx.dmg" is in there, testing that?
17:06:19 <peter1138> andythenorth, test the pr6811 build ;)
17:06:27 <TrueBrain> andythenorth: (also the zip plz)
17:06:34 <frosch123> TrueBrain: "Savegames can be incompatible with other versions of OpenTTD and also with other versions of this PR."
17:06:43 <andythenorth> yeah gatekeeper :P
17:07:01 <TrueBrain> frosch123: good point!
17:07:14 <DorpsGek_II> [OpenTTD/website] TrueBrain updated pull request #50: Add: support openttd-releases and openttd-pullrequests on new infrastructure https://git.io/fhHXx
17:08:05 <andythenorth> TrueBrain: both binaries (dmg and zip) start
17:08:10 <andythenorth> I didn't test more than that
17:08:16 <TrueBrain> sweet :) Are they also what you expect?
17:08:19 <TrueBrain> (in terms how OSX works)
17:08:23 <andythenorth> they look correct to me
17:08:52 <peter1138> TrueBrain is a magician.
17:09:35 <andythenorth> I'll try the PR one
17:09:36 <peter1138> Who ares to approve 6811?
17:09:40 <drac_boy> anyway think going figure out the callback issues another day instead...lunchtime
17:09:46 <DorpsGek_II> [OpenTTD/OpenTTD] LordAro approved pull request #7203: Add: [AzurePipelines] build a stable release if a tag is created https://git.io/fhH1f
17:10:07 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain merged pull request #7203: Add: [AzurePipelines] build a stable release if a tag is created https://git.io/fhHXo
17:10:11 <andythenorth> why is the 6811 bundles.zip so much smaller?
17:10:22 <TrueBrain> andythenorth: release vs nightly
17:11:05 <peter1138> Yeah, tends to be easier for Linux users to compile themselves.
17:11:43 <TrueBrain> okay ... so ... I guess we can tag? :D
17:11:47 <TrueBrain> who wants to try that out :D
17:12:05 <peter1138> Or was it 1.9.0-beta1? :p
17:12:10 <andythenorth> 6811 dmg starts fine and loads my savegame
17:12:23 <TrueBrain> I kinda don't want to do it, as I want to know if how I created it makes sense :)
17:12:29 <TrueBrain> so .. who wants to have the honor? :D
17:12:35 <TrueBrain> andythenorth: awesome!! Tnx a lot :)
17:12:42 <andythenorth> peter1138: so is this moved to the new host?
17:12:53 <peter1138> andythenorth, turns out SYL was right? ;(
17:13:16 <TrueBrain> LordAro? frosch123? peter1138? :)
17:13:24 <peter1138> TrueBrain, not pushing a tag?
17:13:37 <andythenorth> so...can I just go back to making newgrf now? :P
17:13:42 <andythenorth> as we've won development
17:13:43 <TrueBrain> peter1138: I think doing it via github releases is better
17:13:56 <peter1138> TrueBrain, probably, you get more meta data I think.
17:14:06 <peter1138> And there's a checkbox for prerelease.
17:14:07 <TrueBrain> and it avoids silly mistakes :)
17:14:20 <frosch123> let lordaro do it, so we can also make him update the wiki "todos for releases" afterwards :)
17:14:36 <peter1138> LordAro can be our new release manager? :D
17:15:34 <andythenorth> he's not doing a bad job so far
17:16:20 <TrueBrain> Eddi|zuHause: I named it 'openttd-pullrequests'
17:16:26 <TrueBrain> (as openttd-testing was used for betas)
17:16:30 <frosch123> prxyz is a branch, no tag
17:16:59 <frosch123> openttd-pullrequests/123/latest.html ? too much?
17:16:59 <TrueBrain> and the filename is currently the full version name; I might work on that after I have seen it work :)
17:17:13 <TrueBrain> frosch123: yeah, I was thinking something like that too
17:17:24 <TrueBrain> just it is completely against the simplicity we already have in place :P
17:17:31 <TrueBrain> but ... rules are meant to be broken :D
17:17:46 <Eddi|zuHause> it's like complexity is always the enemy of simplicity :p
17:17:49 <TrueBrain> planetmaker: I happen to see your comment, but those comments tend to hide. Comments on existing code is better put in an issue, so they are more visible
17:18:02 <TrueBrain> but first I want this 1.9.0-beta1 done
17:18:08 <TrueBrain> than we can look back to PR binaries :)
17:18:53 <frosch123> hmm, did i put pm on my ignore list? that's the second time today tb adresses him
17:19:24 <TrueBrain> frosch123: I havent seen him talk here, no :)
17:19:41 <TrueBrain> frosch123: I just always hope people have a long enough backlog :D
17:19:59 <TrueBrain> which only became visible after I commented on his comment :P
17:22:15 <TrueBrain> frosch123: now he talked :P
17:22:52 *** Wormnest has joined #openttd
17:23:23 <frosch123> at least i did not misclick something
17:23:24 <TrueBrain> I miss "topics" on IRC
17:23:30 <TrueBrain> crosstalk is a mess :P
17:23:47 <frosch123> @topic set 1 1.9.0-beta1, 1.8.0
17:23:47 *** DorpsGek changes topic to "1.9.0-beta1, 1.8.0 | Website: *.openttd.org (source: github, translator: translator, server list: servers, wiki: wiki, release info: finger) | Don't ask to ask, just ask | 'Latest' is not a valid version, 'Most recent' neither | English only | Logs: @logs | #openttd.dev if this channel is really spammy"
17:23:58 <TrueBrain> frosch123: LIES! :D
17:24:25 <TrueBrain> okay, I made the release myself now
17:24:41 <planetmaker> hm, I commented on existing code, not PR... meh :P
17:25:08 <frosch123> comment on revision... yeah, no idea why that even exists
17:25:09 <TrueBrain> planetmaker: but your comment is not wrong :)
17:25:19 <TrueBrain> planetmaker: just not something for that PR :P
17:25:20 <planetmaker> yeah, will make a PR
17:25:37 <planetmaker> I thought I was commenting on one and trying to improve it ;)
17:26:35 <TrueBrain> 'Version: 1.9.0-beta1'
17:26:37 <TrueBrain> okay, so far so good
17:27:25 <TrueBrain> PRs in folder by PR .. hmmm
17:27:35 <TrueBrain> I did not design the system to handle that, ffs
17:29:17 <DorpsGek_II> [OpenTTD/website] claman commented on pull request #50: Add: support openttd-releases and openttd-pullrequests on new infrastructure https://git.io/fhH1Z
17:29:35 <Samu> i think my feature may cause desyncs
17:30:26 <DorpsGek_II> [OpenTTD/website] TrueBrain commented on pull request #50: Add: support openttd-releases and openttd-pullrequests on new infrastructure https://git.io/fhH1c
17:30:29 <Eddi|zuHause> that's the problem with designs, they tend to collide with the reality afterwards
17:32:14 <LordAro> TrueBrain: DorpsGek_II should shout about tags
17:32:46 <TrueBrain> LordAro: I agree; can you make an issue out of it in DorpsGek-github? Tnx!
17:34:08 <TrueBrain> planetmaker: that PR copies the files that already exist
17:34:09 <Eddi|zuHause> i'm still not entirely sure i understood why we need both DorpsGek and DorpsGek_II
17:34:16 <TrueBrain> so if I start to apply these kind of request sin these, where do I end?
17:34:26 <TrueBrain> it is a chain of events that never stops :P
17:34:31 <TrueBrain> I really like to scope PRs for what they are meant
17:34:36 <TrueBrain> in this case, moving a bit of infrastructure
17:34:40 <TrueBrain> a new PR can address these texts
17:34:43 <peter1138> Eddi|zuHause, high availability.
17:34:48 <frosch123> Eddi|zuHause: do you have any siblings?
17:34:56 <TrueBrain> Eddi|zuHause: old vs new ;)
17:34:57 * peter1138 ponders spending money on a new bike.
17:35:00 <TrueBrain> but .. migrations are a bitch :)
17:35:29 <Eddi|zuHause> or was it __42__?
17:35:30 <LordAro> TrueBrain: hmm. the irc bot i occasionally work on did just add (github) webhook support...
17:35:34 <TrueBrain> just because a PR is open, doesn't mean it is the right place to fix it :)
17:35:46 <LordAro> and is otherwise a fully functional irc bot
17:35:58 <TrueBrain> what is a "fully functional" irc bot :P
17:36:04 <TrueBrain> that is a rather strange thing to say :P
17:36:20 <LordAro> well, "has other functions"
17:36:25 <TrueBrain> LordAro: before Azure Pipelines, there was a very high chance we needed something that handled GitHub events, and something that did IRC stuff
17:36:30 <TrueBrain> so they are split over 2 "bots"
17:36:40 <LordAro> calc, seen, topic, etc
17:36:54 <TrueBrain> that is why we have DorpsGek-github, which handles the GitHub events
17:36:57 <TrueBrain> and forwards it to DorpsGek-IRC
17:37:12 <TrueBrain> for DorpsGek-irc I just used a simple IRC library, as I really was in no mood to look into IRC bots
17:37:21 <TrueBrain> (as ..... they are WEIRD these days :P)
17:37:45 <LordAro> that is honestly impressive
17:37:47 <peter1138> I am somewhat not a happy bunny.
17:37:48 <TrueBrain> but with AzurePipelines (and GitHub actions) it is unlikely DorpsGek willg et any further additions
17:38:19 <Samu> my build is being called 1.9.0-beta1~7, is that normal?
17:38:20 <TrueBrain> so I wouldnt mind if someone makes a Docker that runs a proper IRC bot and also talks correctly (via an App) to GitHub :)
17:39:30 <DorpsGek_II> [OpenTTD/website] TrueBrain commented on pull request #50: Add: support openttd-releases and openttd-pullrequests on new infrastructure https://git.io/fhH10
17:39:47 <LordAro> TrueBrain: i can certainly investigate doing such a thing. i think the "main" instance of it was recently switched to a docker instance anyway, so that would certainly work
17:40:03 <TrueBrain> LordAro: ideally it supported the @seen database of Supybot
17:40:11 <TrueBrain> I think many people would welcome that :)
17:40:22 <Eddi|zuHause> hm... so when i try to run astroneer, i get 60% of time spent in "read_hpet" in perf top
17:40:28 <LordAro> ah, yeah, that could be an issue
17:40:37 <LordAro> but not too difficult to convert, i suspect?
17:40:48 <TrueBrain> from what I saw, indeed
17:41:26 <TrueBrain> LordAro: but one can wonder, if we should have a bot that does all that, AND talks to GitHub
17:41:31 <Eddi|zuHause> last time i had that i had problems with pulseaudio, but i thought i fixed that
17:41:33 <TrueBrain> possibly it might be better and easier to have 2 bots
17:42:05 <peter1138> OpenTTD 1.9.0-beta1
17:42:29 <peter1138> Funny how you don't have to check out a tag for it to still be this version.
17:42:48 <DorpsGek_II> [OpenTTD/website] planetmaker opened pull request #51: Change: Improve description of what is really needed to run OpenTTD https://git.io/fhH1g
17:43:18 <Samu> looks like no desync after all
17:43:25 <Samu> openttd code is mucho smart already
17:43:25 <LordAro> peter1138: git is clever
17:44:23 <DorpsGek_II> [OpenTTD/website] TrueBrain commented on pull request #51: Change: Improve description of what is really needed to run OpenTTD https://git.io/fhH12
17:45:25 <TrueBrain> crap, publishing failed ... oh-oh :D
17:45:25 <peter1138> Hmm, git show on a merge seems unuseful. Just blank.
17:45:31 <LordAro> the bootstrap stuff is *usually* able to download a graphics set, right?
17:45:49 <peter1138> There's no dedi-server bootstrap :(
17:46:28 <glx> bootstrap requires GUI for now IIRC
17:48:45 <DorpsGek_II> [OpenTTD/website] TrueBrain merged pull request #50: Add: support openttd-releases and openttd-pullrequests on new infrastructure https://git.io/fhHXx
17:49:17 <DorpsGek_II> [OpenTTD/website] TrueBrain commented on pull request #51: Change: Improve description of what is really needed to run OpenTTD https://git.io/fhH1K
17:49:38 <TrueBrain> okay, the files are there ... now fixing up the website ...
17:49:45 <DorpsGek_II> [OpenTTD/website] planetmaker updated pull request #51: Change: Improve description of what is really needed to run OpenTTD https://git.io/fhH1g
17:50:46 <TrueBrain> planetmaker: if you don't mind, I can also fix up your PR? (as I created the conflict, I dont mind fixing it :D)
17:51:01 <glx> ok so running generate_widget.sh and squirrel_export.sh on mingw now always change EOL :(
17:52:05 <Eddi|zuHause> anyone have perf-fu? i'm running "perf top" and it shows a high usage in one function, can i see a kind of call stack for that function?
17:55:37 <DorpsGek_II> [OpenTTD/website] TrueBrain opened pull request #52: Add: put openttd-pullrequests per Pull Request in its own folder https://git.io/fhH1X
17:55:42 <TrueBrain> Eddi|zuHause: ^^ :D I found a nice way :)
17:58:08 <planetmaker> TrueBrain, so now please walk me through how I tell git to solve the conflict in a nice manner :)
17:58:29 <LordAro> is it any different to hg?
17:58:32 <planetmaker> just rebase --continue ?
17:58:39 <planetmaker> I've no clue, LordAro ;)
17:58:44 <TrueBrain> planetmaker: no :) You have to resolve the conflict
17:58:51 <TrueBrain> if you rebase, you get a conflict on openttd-releases.md
17:58:59 <TrueBrain> if you open this file, you see <<<< >>> markings
17:59:02 <Eddi|zuHause> planetmaker: git diff to show the conflicts, edit the file, then git add the file, and git rebase --continue
17:59:04 <planetmaker> meh... and no nice merge tool :|
17:59:07 <TrueBrain> modify the file till you are happy again
17:59:12 <TrueBrain> git rebase --continue
17:59:17 <TrueBrain> there are tools that can assist; I never use them
17:59:32 <TrueBrain> planetmaker: but as I made the conflict, I have no issues fixing it? :)
18:00:03 <planetmaker> nah, I want to go through it
18:00:08 <Eddi|zuHause> stupid thing about rebase: if you "pick" the commit and it conflicts, you have to "add" and "--continue", but if you "edit" the commit you need to "add", "commit --amend" and then "--continue"
18:00:13 <planetmaker> gotta learn this with git
18:00:25 <TrueBrain> well, the above is what I just did to see what I broke
18:00:41 <TrueBrain> it aint pretty, but it is mostly easier than to try and understand a tool doing this for you :P
18:00:41 <peter1138> Eddi|zuHause, because one is resolving a conflict, and the other is editing a commit.
18:00:44 <TrueBrain> they confuse the hell out of me
18:01:16 <Eddi|zuHause> peter1138: but i see no reason for the added step
18:01:32 <Samu> yes, hope nielsm likes it now
18:02:18 <Eddi|zuHause> peter1138: what's worse is if you "commit --amend" during conflict resolution, you edit the previous commit
18:02:37 <DorpsGek_II> [OpenTTD/OpenTTD] SamuXarick opened pull request #7204: Feature: Game setting to define how industries with neutral stations accept and supply cargo from/to surrounding stations. https://git.io/fhH19
18:03:26 <Samu> oh, the CI looks different
18:06:38 <Samu> commit checker did not like it
18:06:47 <DorpsGek_II> [OpenTTD/website] TrueBrain commented on pull request #51: Change: Improve description of what is really needed to run OpenTTD https://git.io/fhH1N
18:07:30 <TrueBrain> planetmaker: how is it going? :)
18:07:42 <Samu> *** b/src/industry_cmd.cpp:417: Trailing whitespace: ' * Check whether an oil rig station belongs to the given industry '
18:08:36 <peter1138> I have the git hooks locally, it works wonders, but also probably not on Windows.
18:11:40 <LordAro> there is an editorconfig file
18:11:48 <LordAro> though i imagine VS doesn't support such a thing
18:12:19 <TrueBrain> just have a decent editor that removes trailing whitespaces on save
18:12:26 <frosch123> TrueBrain: assuming we would also publish stuff like jgrpp
18:12:31 <frosch123> what folder would it use?
18:13:13 <frosch123> maybe "openttd-branches/pr123" would be better than "openttd-pullrequests/pr123"
18:13:32 <TrueBrain> frosch123: would it be a pull request?
18:13:45 <TrueBrain> ah :D You scared me for a second :P
18:14:05 <TrueBrain> possibly we can simply do: openttd-jgrpp/ ?
18:14:15 <TrueBrain> or 'jgrpp-nightlies'
18:14:29 <frosch123> ok, so jgrpp would be entirely different from prs?
18:14:39 <TrueBrain> I think that makes things easier
18:14:53 <TrueBrain> but please tell me if you see that different
18:15:23 <TrueBrain> I was thinking that our 'master' is just a name, and so I prepared the system to handle any branch in the same flow
18:15:31 <frosch123> i have no idea about the infrastructure :)
18:15:31 <TrueBrain> master is mapped to openttd-releases and openttd-nightlies
18:15:44 <TrueBrain> we can make jgrpp to jgrpp-releases and jgrpp-nightlies
18:15:58 <TrueBrain> PRs are just one huge mess in that idea, as .. well .. everything is different there :(
18:16:22 <michi_cc> LordAro: VS2017 does.
18:17:18 <DorpsGek_II> [OpenTTD/website] planetmaker updated pull request #51: Change: Improve description of what is really needed to run OpenTTD https://git.io/fhH1g
18:18:21 <Eddi|zuHause> all remotely relevant problems i can find with read_hpet being performance relevant is relating to VMs
18:18:27 <Eddi|zuHause> which doesn't seem right
18:18:29 <TrueBrain> planetmaker: something went wrong :D
18:19:08 <Samu> is it waterborne or waterbourne?
18:19:08 <TrueBrain> not sure what, tbh :)
18:19:29 <planetmaker> possibly. I wonder what?
18:19:50 <TrueBrain> hard to tell, honestly
18:19:57 *** gelignite has joined #openttd
18:19:58 <TrueBrain> possibly if you run: git rebase -i upstream/master
18:20:08 <TrueBrain> that shows you all the commits that is has ready for commit
18:20:19 <planetmaker> I renamed your last commit it seems :D
18:20:42 <Eddi|zuHause> planetmaker: you wrongfully used "commit --amend"?
18:21:00 <planetmaker> I used that, yes. Seemingly not correctly. I did a lot of stuff I cannot reproduce
18:21:01 <DorpsGek_II> [OpenTTD/website] TrueBrain updated pull request #52: Add: put openttd-pullrequests per Pull Request in its own folder https://git.io/fhH1X
18:21:23 <Eddi|zuHause> planetmaker: yeah, don't use that during conflict resolving. i said so :)
18:21:26 <planetmaker> rebasing, pulling, pushing, commit --amend... in various orders :P
18:21:40 <planetmaker> and reset --hard in between several times ;)
18:21:44 <TrueBrain> planetmaker: indeed, you amended my commit, instead of yours :)
18:21:46 <Eddi|zuHause> planetmaker: git reflog
18:21:54 <TrueBrain> you are not allowed to rewrite history! :P
18:22:15 <planetmaker> I'm allowed and forced to rewrite the history of my own repo - so that the PR is updated
18:22:31 <TrueBrain> yes, your own history is fine :)
18:22:34 <TrueBrain> but sorry, I was wrong
18:22:37 <planetmaker> well. git forced me to the merge when I updated / pulled
18:22:40 <TrueBrain> you simply made a merge out of it
18:22:41 <planetmaker> no way around that
18:22:48 <TrueBrain> I never ever pull :)
18:22:54 <TrueBrain> you can learn pull to rebase, instead of merge
18:22:58 <TrueBrain> but I always fetch, and do it myself
18:22:59 <peter1138> planetmaker, don't pull inside a branch!
18:23:03 <DorpsGek_II> [OpenTTD/OpenTTD] SamuXarick updated pull request #7204: Feature: Game setting to define how industries with neutral stations accept and supply cargo from/to surrounding stations. https://git.io/fhH19
18:23:05 <TrueBrain> as ... git annoys the fuck out of me in those regards :P
18:23:09 <planetmaker> I did both... inside and outside...
18:23:09 <Eddi|zuHause> planetmaker: don't pull, rebase to origin/master
18:23:24 <peter1138> You can pull master, but not the branch.
18:23:33 <Eddi|zuHause> or upstream/master, depending on your setup
18:23:52 <planetmaker> I can do lots. But I'm not any wiser despite (or because) of the many comments ;)
18:23:55 <TrueBrain> planetmaker: if you like: go to your branch, git fetch upstream, git rebase upstream/master
18:23:59 <TrueBrain> this gives you (again) a merge conflict to resolve
18:24:02 <Eddi|zuHause> "git fetch origin master" [without /]
18:24:15 <TrueBrain> that still works, and is what most people do I guess :)
18:24:16 <Eddi|zuHause> and then "git rebase -i origin/master" [with /]
18:24:45 <planetmaker> first things first: which rev to checkout now
18:24:50 <TrueBrain> right .. that is the first check: how are your remotes :D
18:25:11 <TrueBrain> what does it tell you?
18:26:00 <Eddi|zuHause> planetmaker: a slightly simplified way: "git checkout master; git pull; git checkout branch; git rebase -i master"
18:26:17 <TrueBrain> that fetches the latest upstream version
18:26:28 <TrueBrain> now jump into your branch (req_desc)
18:26:31 <TrueBrain> git checkout req_desc
18:26:41 <TrueBrain> shows you 3 commits till openttd/master
18:26:49 <TrueBrain> twice the same commit, once mine
18:26:59 <planetmaker> right. But what version to checkout now?
18:27:10 <TrueBrain> "version" in git terms is not really something that has an answer
18:27:16 <planetmaker> before I pull anything?
18:27:21 <planetmaker> version. hash. whatever
18:30:07 <peter1138> You don't want to pull at all.
18:32:36 <peter1138> Hmm, I guess the AI interface could be expanded. Let's see what RailTypes get.
18:33:25 *** supermop_work_ has quit IRC
18:34:37 *** grossing has joined #openttd
18:34:37 *** LordAro has joined #openttd
18:34:37 *** m1cr0man has joined #openttd
18:34:37 *** seatsea has joined #openttd
18:34:37 *** Alkel_U3 has joined #openttd
18:34:37 *** Hobbyboy has joined #openttd
18:34:37 *** argoneus_ has joined #openttd
18:34:37 *** DorpsGek_II has joined #openttd
18:34:37 *** ZirconiumX has joined #openttd
18:34:37 *** KenjiE20 has joined #openttd
18:34:37 *** Sheogorath has joined #openttd
18:34:37 *** Mahjong2 has joined #openttd
18:34:37 *** Lejving has joined #openttd
18:34:37 *** dwfreed has joined #openttd
18:34:37 *** greeter has joined #openttd
18:34:37 *** innocenat_ has joined #openttd
18:34:37 *** mikegrb has joined #openttd
18:34:37 *** ccfreak2k has joined #openttd
18:34:37 *** Antheus has joined #openttd
18:34:37 *** ToBeFree has joined #openttd
18:34:37 *** reldred has joined #openttd
18:34:37 *** AKTheKnight has joined #openttd
18:34:37 *** daspork has joined #openttd
18:34:37 *** Smedles_ has joined #openttd
18:34:37 *** techmagus has joined #openttd
18:34:37 *** Laedek_ has joined #openttd
18:34:37 *** Extrems has joined #openttd
18:34:37 *** charon.oftc.net sets mode: +ovv orudge orudge glx
18:34:37 *** Gustavo6046 has joined #openttd
18:34:37 *** supermop_work_ has joined #openttd
18:34:37 *** quiznilo has joined #openttd
18:34:37 *** rocky11384497 has joined #openttd
18:34:37 *** synchris has joined #openttd
18:34:37 *** Wormnest has joined #openttd
18:34:37 *** gelignite has joined #openttd
18:34:37 *** andythenorth has joined #openttd
18:34:37 *** Hirundo has joined #openttd
18:34:37 *** ^Spike^ has joined #openttd
18:34:37 *** Terkhen has joined #openttd
18:34:37 *** XeryusTC has joined #openttd
18:34:37 *** Webster has joined #openttd
18:34:37 *** TrueBrain has joined #openttd
18:34:37 *** Hazzard has joined #openttd
18:34:37 *** michi_cc has joined #openttd
18:34:37 *** OsteHovel has joined #openttd
18:34:37 *** TinoDidriksen has joined #openttd
18:34:37 *** Eddi|zuHause has joined #openttd
18:34:37 *** HerzogDeXtEr has joined #openttd
18:34:40 *** frosch123 has joined #openttd
18:34:40 *** Speedy` has joined #openttd
18:34:40 *** tyteen4a03 has joined #openttd
18:34:40 *** dihedral has joined #openttd
18:34:40 *** fonsinchen has joined #openttd
18:34:40 *** charon.oftc.net sets mode: +ovv Terkhen Terkhen michi_cc
18:34:40 *** planetmaker has joined #openttd
18:34:40 *** V453000 has joined #openttd
18:34:40 *** ericnoan has joined #openttd
18:34:40 *** SpComb^ has joined #openttd
18:34:40 *** TrueBrain_ii has joined #openttd
18:34:40 *** charon.oftc.net sets mode: +ov planetmaker planetmaker
18:35:09 <DorpsGek_II> [OpenTTD/website] planetmaker updated pull request #51: Change: Improve description of what is really needed to run OpenTTD https://git.io/fhH1g
18:35:24 <Eddi|zuHause> Wolf01: "we" were here all the time
18:35:34 <DorpsGek_II> [OpenTTD/website] TrueBrain updated pull request #52: Add: put openttd-pullrequests per Pull Request in its own folder https://git.io/fhH1X
18:35:42 <planetmaker> yay :) Thanks a lot!
18:35:50 <planetmaker> for the extensive, detailed walk-through
18:36:10 <Wolf01> I was alone with just other 12 people :(
18:36:14 <planetmaker> now I can actually look at your comments there on the ticket and fix those, too ;)
18:36:35 *** snail_UES_ has joined #openttd
18:36:39 *** ChanServ sets mode: +v peter1138
18:37:13 <Wolf01> Mmmh, the last 40MB of 10GB, this download is taking AGES
18:37:19 <DorpsGek_II> [OpenTTD/website] TrueBrain commented on pull request #51: Change: Improve description of what is really needed to run OpenTTD https://git.io/fhHMu
18:37:28 <TrueBrain> (some things got lost in the rebasing :D)
18:37:49 <DorpsGek_II> [OpenTTD/website] Eddi-z commented on pull request #51: Change: Improve description of what is really needed to run OpenTTD https://git.io/fhHMz
18:38:01 <TrueBrain> Eddi|zuHause: the most fun I have seen with reflog btw was when the GC kicked in :P That made people cry :D
18:38:07 <TrueBrain> (their workflow depended on reflog, basically)
18:38:33 <Eddi|zuHause> TrueBrain: that sounds like horribly broken workflow
18:38:37 <TrueBrain> Eddi|zuHause: the typical german way of doing things is being inconsistent?! :P
18:38:42 <TrueBrain> Eddi|zuHause: owh, yes, it was :D
18:38:44 <Eddi|zuHause> TrueBrain: like relying on windows trash bin
18:38:51 <Eddi|zuHause> TrueBrain: as "backup"
18:39:30 <Eddi|zuHause> TrueBrain: no the typical german way of using "and" is without comma (except for some rare special cases)
18:39:31 <TrueBrain> I once had a customer, that called up fully in panic ... her "archive" was empty. After talking for 10 minutes I found out that she used her "trash bin" as "archive" .. and the server was out of space so cleaned the trash bins ..
18:40:02 <TrueBrain> Eddi|zuHause: same for Dutch. In English you can use both, both are equally valid (as far as I know)
18:40:09 <TrueBrain> the inconsistency was what was killing me :D
18:40:33 <TrueBrain> (the customer above was in relation to email btw :P)
18:40:48 <Eddi|zuHause> i heard english people tended to insist on the comma
18:41:00 <TrueBrain> I like the comma; makes me feel more important
18:41:21 <TrueBrain> hmm .. I was wondering why I was hungry
18:41:27 <TrueBrain> time to get myself some food I guess
18:41:31 <Eddi|zuHause> i pretty much never put the comma
18:41:56 <TrueBrain> you also have issues using the shift key, we notice
18:42:00 <frosch123> comma, make, you, appear, more, important?
18:42:13 <Eddi|zuHause> that's because that rule makes no sense
18:42:29 <TrueBrain> many rules make no sense
18:42:34 <frosch123> hmm, now i remember dale'stan
18:42:36 <TrueBrain> life makes no sense!
18:42:51 <TrueBrain> the only thing that makes sense .. ARE KITTENS
18:42:58 <DorpsGek_II> [OpenTTD/website] planetmaker updated pull request #51: Change: Improve description of what is really needed to run OpenTTD https://git.io/fhH1g
18:43:36 <planetmaker> TrueBrain, I disagree with the comment on "exactly one base set" (as it doesn't hurt to have 12.4. The rest should be amended accordingly
18:44:26 <Eddi|zuHause> TrueBrain: at some point in the last 10 years i decided i want to actively oppose that rule
18:44:52 <TrueBrain> planetmaker: you now no longer like to openttdcoop at all?
18:45:31 <TrueBrain> planetmaker: if you are new to OpenTTD, the sentence: "OpenTTD requires one version of base graphics" , reads really weird
18:45:38 <planetmaker> Well, I do. But I want to move the repos to github as well. It's basically me alone to maintain the whole server. That's not something I'm comfortable with
18:45:38 <TrueBrain> is there anything we can do to make it more clear?
18:45:48 <TrueBrain> planetmaker: fair enough :)
18:46:16 <Eddi|zuHause> "OpenTTD requires a separate base graphics set"
18:46:26 <planetmaker> ^ sounds good to me
18:46:34 <frosch123> planetmaker: if you want a repository moved, just shout :)
18:46:35 <TrueBrain> "base graphics set" .. how do you know what that is?
18:46:45 <Eddi|zuHause> ", which can be downloaded through the installer"
18:46:49 <frosch123> i did not continue on the issue moving though, but andy closes them anyway
18:47:02 <Eddi|zuHause> "in case you want to manually download it, you can get it from:"
18:47:18 <TrueBrain> OpenTTD requires a separate "base graphics set", a set which contains all the graphics for the game.
18:47:28 <TrueBrain> possibly without "separate"
18:47:56 <planetmaker> I don't think 'base graphics set' needs extensivel explanation at that point. We have a readme to explain details
18:48:18 <Eddi|zuHause> TrueBrain: "separate" was intended as a short for "is independent from the binary"
18:48:35 <TrueBrain> Eddi|zuHause: ah; I did not read it as such :D
18:48:40 <TrueBrain> meh .. english ... :P
18:48:55 <planetmaker> frosch123, thanks! I will come back to that. Before we move more repos, I would like to overhaul and test the newgrf build system to be more sustainable. And maybe work the same way and on the same infrastructure as OpenTTD
18:49:29 <TrueBrain> planetmaker: I can strongly advise Azure Pipelines :D Saves you a lot of "maintaining a server" shit :P
18:50:08 <planetmaker> there are alternatives. But if the same infrastructure is used... I actually would make that NewGRF-CI part of the OpenTTD project as well
18:50:11 <planetmaker> if you don't mind
18:50:23 <planetmaker> but... I first have to get *something* working at all
18:50:39 <planetmaker> and it's a slow and steep learning curve for me
18:50:40 <TrueBrain> if you need help etc, let us know; few of us now start to gain some experience with Azure Pipelines :D
18:50:42 <DorpsGek_II> [OpenTTD/website] Eddi-z commented on pull request #51: Change: Improve description of what is really needed to run OpenTTD https://git.io/fhHMi
18:50:51 <TrueBrain> it is really ugly to get to know AP :)
18:51:26 <DorpsGek_II> [OpenTTD/website] TrueBrain commented on pull request #51: Change: Improve description of what is really needed to run OpenTTD https://git.io/fhHMM
18:51:43 <nielsm> how come the windows version is able to start in a "blank" mode where it can offer to download opengfx automatically, but non-windows versions can't?
18:51:45 <TrueBrain> right, enough nitpicking :)
18:51:50 <nielsm> what is windows-specific about that?
18:52:03 <TrueBrain> nielsm: Eddi|zuHause just suggested it also works on Linux :P
18:52:05 <TrueBrain> (no clue if it does)
18:52:18 <TrueBrain> bad Eddi|zuHause :D
18:52:18 <planetmaker> eddi is right with it
18:52:19 <DorpsGek_II> [OpenTTD/website] Eddi-z commented on pull request #51: Change: Improve description of what is really needed to run OpenTTD https://git.io/fhHMD
18:52:42 <planetmaker> it works on linux w/o it present, offering it for download. did last time I tested it
18:52:53 <TrueBrain> conflicting results, it seems :D
18:53:00 <Eddi|zuHause> i was pretty certain it worked on linux
18:53:06 <Eddi|zuHause> just not on macos
18:53:23 <planetmaker> but please agree on 'download' vs. 'install' and I'll push yet another update :)
18:53:30 <TrueBrain> "You must own a version of Transport Tycoon Deluxe, they cannot be downloaded." <- you missed a possibility to use ; there :)
18:53:43 <TrueBrain> "download" in my mind makes no sense. Just downloading it is not enough
18:53:47 <Eddi|zuHause> i don't see the point of ;
18:53:47 <TrueBrain> installing is the part that matters
18:54:08 <TrueBrain> Eddi|zuHause: yeah, you seem to have more of those :D
18:54:34 <Eddi|zuHause> dunno, "installing with the installer" seems redundant
18:54:50 <TrueBrain> @choice download install
18:55:13 <TrueBrain> LordAro: where is that bot of yours? :P
18:55:16 <frosch123> nielsm: at some point it was missing for osx
18:55:21 <frosch123> but win and linux works
18:55:24 <planetmaker> I'm with eddi on that it sounds nicer. And... and a download by an installer implies that it is being installed
18:55:56 <TrueBrain> planetmaker: that is fair; I was mostly triggered by the "In case you want to manually download it"
18:55:59 <glx> last time I needed it, it worked on linux, but only in GUI mode
18:56:02 <TrueBrain> In case you want to manually download änd install it," ?
18:56:07 <TrueBrain> best of both worlds :P
18:56:15 <nielsm> frosch123: my fresh 1.9.0-beta1 build on fedora does not
18:56:19 <TrueBrain> okay, I stopped to care about these details I noticed :D What ever is pushed next gets approved :P
18:56:39 <TrueBrain> änd .. that is a new "and"
18:56:43 <glx> nielsm: starting as dedicated or normal ?
18:56:45 <TrueBrain> did not know greykeys were not on :)
18:57:10 <DorpsGek_II> [OpenTTD/website] planetmaker updated pull request #51: Change: Improve description of what is really needed to run OpenTTD https://git.io/fhH1g
18:57:44 <nielsm> (the _baseset dir I move back in contains original game graphics)
18:58:02 <Eddi|zuHause> planetmaker: my second suggestion got lost? "Alternatively, the original Transport Tycoon Deluxe data files can be used as base graphics set. You must own a version of Transport Tycoon Deluxe, they cannot be downloaded."
18:58:08 <TrueBrain> tnx a lot planetmaker :) As you can see, these changes can take a few iterations :D but I really appreciate you doing this :)
18:58:39 <glx> on my debian VM it worked when I needed it last month
18:58:45 * LordAro takes a look at some words
18:58:46 <frosch123> nielsm: can you also move your openttd.cfg
18:58:52 <Eddi|zuHause> nielsm: maybe depends on not having openttd.cfg
18:59:44 <planetmaker> I also agree with eddi's latest comment. But I would like that to be the last change :P
18:59:53 <planetmaker> yes, I only saw it now, sorry
18:59:56 <TrueBrain> I agree with planetmaker's last comment :)
19:00:05 <nielsm> nope still not working after removing openttd.cfg
19:00:15 <Eddi|zuHause> TrueBrain: race condition :p
19:00:24 <DorpsGek_II> [OpenTTD/website] LordAro commented on pull request #51: Change: Improve description of what is really needed to run OpenTTD https://git.io/fhHMN
19:00:51 <TrueBrain> you want to help out, and you are beaten to death with comments :P
19:00:55 <TrueBrain> we are not nice people :( :P
19:01:15 <Eddi|zuHause> nielsm: works here if i rename ~/.openttd
19:01:38 <planetmaker> sometimes bike-shedding can be fun :P
19:01:53 <DorpsGek_II> [OpenTTD/website] planetmaker updated pull request #51: Change: Improve description of what is really needed to run OpenTTD https://git.io/fhH1g
19:02:49 <DorpsGek_II> [OpenTTD/website] LordAro approved pull request #52: Add: put openttd-pullrequests per Pull Request in its own folder https://git.io/fhHMp
19:02:59 <TrueBrain> dont merge please :) planetmaker will hate me
19:03:27 <planetmaker> LordAro, base graphics are required at every run ;)
19:04:07 <nielsm> weird I can't make it work
19:04:09 <TrueBrain> I think he meant the saying is: "on first run"
19:04:19 <planetmaker> ah... hm. at vs. on
19:04:20 <TrueBrain> "at first run" is not the correct .. what is that 'at/on' word?
19:04:44 <TrueBrain> nielsm: did you build it? :D
19:04:53 <Eddi|zuHause> nielsm: maybe you have another openttd.cfg in weird places?
19:05:51 <nielsm> can't bother to try more right now
19:06:10 <planetmaker> I just checked: I'm offered the choice of OpenGFX for download when I delete all basesets
19:06:30 <nielsm> the git hash being "0" in rev.cpp for the 1.9.0-beta1 tag is more serious
19:06:41 <nielsm> can anyone confirm that?
19:07:12 <TrueBrain> I cannot build OpenTTD, as there is a space in my folder name .... *facepalm*
19:07:29 <Eddi|zuHause> nielsm: can we agree on that more people seem to have it working than not working, and thus the "Linux" mention is justified? :p
19:07:46 <TrueBrain> planetmaker: will you be fixing the at/on, or shall I just hit merge?
19:08:18 <planetmaker> In case you want to manually download and install it, some options are:
19:08:40 <planetmaker> if we do bike shedding, let's do it properly ;)
19:08:50 <planetmaker> I'm learning git, doing so
19:08:52 <TrueBrain> fine; than remove the double newlines on top too :P
19:09:06 <TrueBrain> and the double space behind '-' somewhere in the middle :P
19:09:19 <LordAro> planetmaker: sounds good
19:10:19 <TrueBrain> nielsm: const char _openttd_revision_hash[] = "0";
19:10:26 <DorpsGek_II> [OpenTTD/website] planetmaker updated pull request #51: Change: Improve description of what is really needed to run OpenTTD https://git.io/fhH1g
19:10:35 <planetmaker> anyone want to approve? :)
19:10:53 <TrueBrain> but I am going to approve :P
19:11:21 <peter1138> I guess 1.9.0-beta1 is bad? :p
19:11:23 <TrueBrain> (you fixed 2 of the 3 files in regards to newlines :D)
19:11:23 <DorpsGek_II> [OpenTTD/website] LordAro approved pull request #51: Change: Improve description of what is really needed to run OpenTTD https://git.io/fhHDU
19:11:41 <TrueBrain> owh .. that is wrong
19:11:59 <TrueBrain> it does list the beta
19:12:16 <planetmaker> but that's not stable. It's testing
19:12:24 <TrueBrain> planetmaker: yeah, caching is still acting up. Still not sure how to fix it properly
19:12:51 <TrueBrain> planetmaker: that is what is wrong, yes :)
19:13:13 <Eddi|zuHause> is there no generic linux binary anymore?
19:13:22 <peter1138> nielsm, it's 0 for any build.
19:13:36 <TrueBrain> a penny for every time I read that question
19:13:43 <TrueBrain> < 1% of the downloads, but people seem to care :D
19:13:44 <peter1138> Not just the 1.9.0-beta1 version.
19:14:11 <Eddi|zuHause> "< 2%" is not insignificant
19:14:31 <TrueBrain> hmm ... something went wrong with stable/testing :D lol
19:14:41 <TrueBrain> this .. cannot work :D
19:15:09 <DorpsGek_II> [OpenTTD/website] TrueBrain merged pull request #51: Change: Improve description of what is really needed to run OpenTTD https://git.io/fhH1g
19:16:44 <DorpsGek_II> [OpenTTD/website] TrueBrain commented on pull request #51: Change: Improve description of what is really needed to run OpenTTD https://git.io/fhHDI
19:16:59 <DorpsGek_II> [OpenTTD/website] TrueBrain dismissed a review for pull request #52: Add: put openttd-pullrequests per Pull Request in its own folder https://git.io/fhHMp
19:17:00 <DorpsGek_II> [OpenTTD/website] TrueBrain updated pull request #52: Add: put openttd-pullrequests per Pull Request in its own folder https://git.io/fhH1X
19:17:25 <peter1138> GITHASH := $(shell echo "$(VERSIONS)" | cut -f 3 -d' ')
19:17:29 <peter1138> ^^ it's field 4, not 3
19:17:32 <peter1138> field 3 is "modified"
19:17:44 <TrueBrain> I see a beta2 in our near future :)
19:18:33 <peter1138> Delete and make a new release ;)
19:20:03 <glx> or at least it should be correct
19:20:24 <peter1138> So we have linux builds that can't connect to windows :-)
19:22:01 <TrueBrain> Liquid (Jekyll templating bla) doesn't have a not operator :o
19:23:34 <frosch123> intesting strategy to promote de-morgan
19:24:06 <TrueBrain> it also has 'contains'
19:24:11 <TrueBrain> so I have no way to check if something is NOT in there
19:24:24 <peter1138> Yeah but I have a pint of beer :D
19:24:29 <peter1138> Well, slightly less now.
19:24:46 <frosch123> TrueBrain: "unless"
19:24:49 <TrueBrain> (beer is like a sandwich with cheese, right)
19:25:15 <frosch123> hmm, that's liquid, not jekyll
19:25:55 <TrueBrain> okay, I will fiddle with this a bit later ..
19:26:17 <DorpsGek_II> [OpenTTD/website] TrueBrain merged pull request #52: Add: put openttd-pullrequests per Pull Request in its own folder https://git.io/fhH1X
19:29:49 <planetmaker> my... OpenTTD reports 1.9.0-beta1 when I checkout that tag
19:30:23 <peter1138> It's the network revision that's messed up.
19:30:59 <peter1138> game and network revision are decoupled.
19:31:18 <planetmaker> yep, I didn't miss that
19:32:11 <nielsm> if you try 'gamelog' in the console you'll see the gamelog claims version is g0
19:32:34 <nielsm> I thought I had tested the bash script for this
19:33:08 <LordAro> sound like it needs an assert_compile(lengthof(revision_hash) == 10)
19:33:12 <LordAro> or whatever it actually is
19:33:16 <planetmaker> and you probably did. And then fiddled around a bit more...
19:33:53 <nielsm> well it may be an empty string in the case of no version control present and no .ottdrev file
19:34:17 <glx> no the problem is in makefile.src.in
19:34:33 <planetmaker> content download is ... VERY slow :(
19:35:07 <glx> wrong field is used for githash
19:35:28 <glx> nielsm: ] <@peter1138> GITHASH := $(shell echo "$(VERSIONS)" | cut -f 3 -d' ')
19:35:56 <glx> but for windows it's correct
19:36:04 <glx> (not checked gamelog stuff)
19:36:06 * nielsm blames whoever reviewed the PR
19:42:54 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh opened pull request #7205: Fix 5f8354f3: Non-Windows builds did not get correct git hash https://git.io/fhHDW
19:44:01 <DorpsGek_II> [OpenTTD/OpenTTD] glx22 approved pull request #7205: Fix 5f8354f3: Non-Windows builds did not get correct git hash https://git.io/fhHD4
19:44:31 <nielsm> the algorithm used to generate network revision string will still have a funny effect on 1.9.0-beta1 though, it will probably be rewritten to 1.9.0-gwhateverhash
19:45:05 <nielsm> dunno if that's "good enough" or that should also get a fix
19:45:59 <peter1138> Check if hash is the correct lenght?
19:46:06 <glx> depends if it's the string shown to players in the network window
19:46:16 <nielsm> it will get shown to players yes
19:46:17 <peter1138> I guess it's not always used, if there is no hash.
19:46:40 <glx> then 1.9.0-beta1 is required I think
19:46:59 <nielsm> I think I'd like an "is tagged version" flag in rev.cpp too
19:47:17 <nielsm> to indicate that the main revision string should never be mangled
19:49:29 <glx> just checking if the string already fit is not enough ?
19:51:07 <nielsm> not quite, because of the "git hash only" check for network version compat
19:51:20 <nielsm> that could make 1.9.0-beta1 and 2.0.0-beta1 compare equal
19:52:54 <peter1138> Oh, I just noticed that OzTrans is back on the forums, heh.
19:53:07 <peter1138> And suggesting use of... GRM. How oldschool.
19:53:40 <glx> still disabling his grfs in openttd ?
19:54:09 <planetmaker> hm, SilverSurfer banned and him not?
19:54:27 <peter1138> Hmm, where/how do we publish AI/GS documentation?
19:54:44 <peter1138> planetmaker, I don't think orudge does permabans.
19:54:55 <planetmaker> peter1138, he does
19:55:00 <peter1138> planetmaker, ah, it's automatic?
19:55:15 <planetmaker> the docs should be automatically updated upon build
19:56:10 <peter1138> It's not automatic cos it's slow.
19:56:12 <TrueBrain> updated where, I wonder :)
19:56:20 <peter1138> Hmm, lots of warnings.
20:01:23 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh merged pull request #7205: Fix 5f8354f3: Non-Windows builds did not get correct git hash https://git.io/fhHDW
20:02:11 <peter1138> beta1 wasn't officially released yet... so...
20:02:25 <peter1138> const char _openttd_revision_hash[] = "0151fe998a999b48b67afa5b96d9a4cd72246455";
20:02:25 <TrueBrain> nielsm: indeed, beta2 :)
20:02:33 <peter1138> Of course, our repos all have it tagged now :p
20:03:13 <TrueBrain> nielsm: guess we first need a commit to do all kind of bumps for that :)
20:03:27 <peter1138> LordAro's our release man ;)
20:03:35 <TrueBrain> we are going for the shortest changelog here, right? :)
20:03:57 <planetmaker> -Fix: Network revision was reported wrongly
20:04:08 <planetmaker> is it longer than "No changes"?
20:04:10 <peter1138> incorrectly reported
20:04:27 <nielsm> now it's just inconveniently reported
20:05:16 <peter1138> Could be worse. It's only a beta, after all.
20:05:57 <planetmaker> No harm done there except a dent in some imaginary crown ;)
20:06:36 <TrueBrain> I found another bug :D
20:06:41 <TrueBrain> but that is in the release pipeline
20:07:09 <peter1138> frosch123, now that was a mistake :D
20:09:29 <Eddi|zuHause> 0.4.0 was the first version of openttd i downloaded
20:10:12 <Eddi|zuHause> dunno how i managed that exact time window :p
20:10:30 <TrueBrain> hmm .. is there a way to detect if this was a stable version or not, from findversion.sh ?
20:12:03 *** supermop_Home has joined #openttd
20:13:36 <nielsm> unless you do some kind of regex on branch and/or tag name
20:14:28 <TrueBrain> meh; solved it differently
20:14:43 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain opened pull request #7206: Fix: [AzurePipelines] manifest.sh didn't know when it was a stable release https://git.io/fhHDb
20:14:47 <TrueBrain> another reason to do beta2 ^^ :D
20:16:24 <TrueBrain> "Latest release in beta1 is 1.9.0-beta1, released on 2019-02-09 16:26 UTC."
20:16:27 <TrueBrain> oef, that is also wrong
20:19:43 <frosch123> TrueBrain: we have a release flag
20:20:27 <TrueBrain> frosch123: from findversion.sh?
20:21:19 <michi_cc> Depending on the GitHub repo config, it is probably possible to redo the tag, but I've got no idea if that updates the sources code zip files in the GitHub release.
20:21:59 <frosch123> there are many changes between svn-branches and svn-tags
20:22:04 <frosch123> assertions and stuff
20:22:05 <TrueBrain> michi_cc: we could, but we shouldn't :)
20:22:32 <TrueBrain> frosch123: that is not findversion.sh :) This is for the build system, which only uses findversion.sh :)
20:22:42 <TrueBrain> I now have an external flag to indicate if the intend is to build a stable
20:23:11 <DorpsGek_II> [OpenTTD/website] TrueBrain opened pull request #53: Fix: 'testing' and 'stable' were always the latest version https://git.io/fhHyJ
20:23:25 <TrueBrain> okay .. with these 2 fixes the website will also show the correct version now
20:24:07 <TrueBrain> one downside, I noticed: 'openttd-releases/latest.html' will always point to the latest; there is no URL to get the latest stable
20:24:10 <TrueBrain> latest might be a beta, etc
20:24:13 <michi_cc> It is basically still unreleased, or will your CDN go nuts about the changes hash? :)
20:24:13 <TrueBrain> not sure that is a bad thing
20:24:48 <TrueBrain> michi_cc: I would need to scrub the CDN, local copies of people might be confused, people might already have downloaded the version, etc etc
20:24:52 <TrueBrain> it really doesn't hurt to make a beta2
20:24:56 <TrueBrain> it is, after all, a beta
20:25:00 <TrueBrain> bugs are expected to be found
20:25:09 <TrueBrain> lets not make things more complicated by making it something more :D
20:25:52 <frosch123> shouldn't there be a openttd-testing/latest.html ?
20:27:04 <TrueBrain> frosch123: we used to do that, but it was also annoying
20:27:11 <TrueBrain> but we can build that again, if that helps
20:27:16 <TrueBrain> I was more wondering if it mattered
20:27:45 <TrueBrain> main issue with 'testing' is what to do with it if 'stable' is newer
20:28:03 <TrueBrain> its a pretty amount of logic to work around that
20:29:32 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain updated pull request #7206: Fix: [AzurePipelines] manifest.sh didn't know when it was a stable release https://git.io/fhHDb
20:29:38 <Eddi|zuHause> what about the corner case of "newer" (as in later release date) stable bugfix release, that is technically derived from an older branch version?
20:29:55 <TrueBrain> why is this never straight forward! :P
20:30:06 <Eddi|zuHause> e.g. release 1.9.0-beta and then 1.8.1
20:30:18 <TrueBrain> the main website will show the right thing btw
20:30:25 <TrueBrain> those links are always correct
20:30:31 <TrueBrain> but what 'latest.html' is, that is the question :D
20:32:02 <TrueBrain> LOL! The nightly just also build 1.9.0-beta1 :D
20:32:16 <Eddi|zuHause> btw, i don't think that case actually ever happened
20:33:47 <DorpsGek_II> [OpenTTD/website] TrueBrain merged pull request #53: Fix: 'testing' and 'stable' were always the latest version https://git.io/fhHyJ
20:34:52 <Eddi|zuHause> i guess the findversion script needs to check which branch is actually checked out, instead of just looking for whether it has a version tag?
20:35:14 <TrueBrain> the nightly is correct, as in, it is identical to the 1.9.0-beta1 release
20:35:25 <TrueBrain> just .. it shows 1.9.0 on the banner
20:35:31 <Eddi|zuHause> i wasn't questioning that
20:35:38 <TrueBrain> because it assumes it is like date-branch-hash
20:36:02 <TrueBrain> Eddi|zuHause: what I meant was, I think the website is wrong :P
20:36:21 <TrueBrain> now I come to think of it
20:36:48 <TrueBrain> possibly this nightly simply shouldn't have run
20:37:14 <Eddi|zuHause> i was assuming the build script recognized that there were no commits after the tag was added, and thus built the binary according to the tag
20:37:45 <TrueBrain> findversion.sh returns the tag if the current commit has any
20:38:01 <TrueBrain> and a nightly is also fully compatible with the release, so it is right in doing so
20:38:29 <TrueBrain> so .... what do we want to do here :s
20:38:37 <Eddi|zuHause> maybe we should just do nothing and accept the oddity
20:38:48 <TrueBrain> so only the website is wrong in saying 1.9.0
20:38:51 <TrueBrain> not mentioning it is a beta
20:39:25 <Eddi|zuHause> the website is just confused about the different versioning scheme?
20:40:02 <Eddi|zuHause> was expecting $1-$2-$3 and prints $1
20:40:36 <Eddi|zuHause> dunno how to detect that in a clever way
20:40:52 <TrueBrain> or do we just accept this happening once in a blue moon, basically
20:41:12 <michi_cc> If num - <> 2 print all else print $1 ? :p
20:41:42 <Eddi|zuHause> yeah, check whether the name fits the scheme, or print the whole name otherwise
20:41:52 <TrueBrain> {{ latest_nightly.version | split: "-" | slice: 0 }}
20:41:59 <TrueBrain> really difficult to do that, in this templating language :)
20:42:10 <TrueBrain> is it worth the fuzz?
20:42:37 <Eddi|zuHause> but i dunno anything about the templating language
20:42:52 <Eddi|zuHause> there needs to be a clever functional expression
20:43:13 <TrueBrain> {% if latest_nightly.version != latest_stable.version and latest_nightly.version != latest_testing.version %}
20:43:18 <TrueBrain> in that case, don't show the nightly
20:43:51 <Eddi|zuHause> that sounds like it might cause confusion
20:44:03 <TrueBrain> showing 1.9.0-beta1 will cause confusion anyway :P
20:44:21 <planetmaker> There is no harm in having a nightly identical to another version
20:44:29 <TrueBrain> that we established yes :)
20:44:45 <Eddi|zuHause> but certainly less confusion than pretending there was no nightly
20:44:59 <planetmaker> And also no harm in showing identical versions
20:45:16 <planetmaker> That happens like once in a blue moon anyway
20:45:27 <TrueBrain> Eddi|zuHause: you could make the case it is better for people to download the testing over the nightly, in these cases ;)
20:45:32 <planetmaker> (though new moon was like 3 days ago... but yes ;) )
20:46:01 <planetmaker> TrueBrain, conversely you can argue it doesn't matter - and offering both is less support work
20:46:22 <TrueBrain> planetmaker: no, offering both is more work
20:46:26 <TrueBrain> it is only a VISUAL thing we talk about :)
20:46:28 <Eddi|zuHause> planetmaker: "blue moon" is when two full moons happen in one month
20:46:46 <Eddi|zuHause> which isn't actually all that rare
20:46:48 <TrueBrain> it is only about the banner in the top of the website
20:46:52 <planetmaker> which happens like once a year
20:47:35 <TrueBrain> basically, what we are doing now is fixing an edge case with another edge case
20:47:41 <TrueBrain> which might or might not break the next time the edge case happens :D
20:47:44 <TrueBrain> so .. this will be fun ;)
20:48:07 <planetmaker> ^^ that's why I think we should simply offer identical versions. And also display them all
20:48:21 <TrueBrain> which can only be fixed by adding code for this edge case
20:48:24 <TrueBrain> which is the whole point
20:48:35 <TrueBrain> read a bit high, 20:40
20:48:36 <Eddi|zuHause> it occasionally happens twice a year. when it happens in january, it also might happen in march again (skipping a full moon in february)
20:48:40 <TrueBrain> eddi explained well
20:50:11 <DorpsGek_II> [OpenTTD/website] TrueBrain opened pull request #54: Fix: in a very rare case, it might happen nightly version == testing version https://git.io/fhHyO
20:50:20 <TrueBrain> so this fixes that edge case, but I strongly doubt it will ever work again :D
20:50:23 <TrueBrain> it just fixes THIS edge case
20:50:28 <TrueBrain> by adding 5 more lines of code ..
20:50:39 <TrueBrain> yeah maintainability ...........
20:51:24 <Eddi|zuHause> <TrueBrain> {% if latest_nightly.version != latest_stable.version and latest_nightly.version != latest_testing.version %} <-- is there a reason why you can't use that check to switch between showing slize 0 or whole name?
20:51:41 <TrueBrain> that is what I am doing, not?
20:51:47 <TrueBrain> I don't follow, sorry :)
20:51:56 <peter1138> I padded out my spaghetti bolognese with enough stuff to make it last two meals... but it only has 1 portion (and small at that) of beef mince in it... Woe.
20:52:04 <Eddi|zuHause> haven't looked at the PR yet
20:52:30 <TrueBrain> but ... it is very fragile, to check the version like that
20:52:33 <Eddi|zuHause> right, then scratch the question
20:52:38 <TrueBrain> it is not really solving the problem, honestly
20:52:42 <TrueBrain> but the problem has no solution :D
20:52:52 <TrueBrain> okay, done with this, fine for now :P
20:52:55 <TrueBrain> up to the testing/stable issue
20:53:11 <Eddi|zuHause> TrueBrain: well, alternative would be to check the number of slices
20:53:26 <TrueBrain> Eddi|zuHause: yeah; but that is also very error prune honestly
20:53:29 <Eddi|zuHause> or whether the second slice is "master"
20:53:47 <TrueBrain> that alternative I can work with
20:53:58 <TrueBrain> if someone changes 'master' into something else ..
20:54:06 <TrueBrain> so weird to have that hidden in your website
20:54:12 <Eddi|zuHause> then we need to update the website anyway
20:54:36 <TrueBrain> euh, no, that master idea fails
20:54:41 <TrueBrain> as there is no second slice if it is a release
20:55:04 <Eddi|zuHause> that's where the "has 3 slices" check comes in?
20:55:11 <DorpsGek_II> [OpenTTD/website] TrueBrain merged pull request #54: Fix: in a very rare case, it might happen nightly version == testing version https://git.io/fhHyO
20:55:16 <TrueBrain> that is very annoying in this templating stuff
20:55:25 <TrueBrain> what-ever; if we do what we have been doing for the next 15 years
20:55:27 <Eddi|zuHause> assuming it's lazy evaluation of &&
20:55:51 <TrueBrain> I really cannot be bothered too much with these edge cases that only happens once a year :)
20:56:44 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain updated pull request #7206: Fix: [AzurePipelines] manifest.sh didn't know when it was a stable release https://git.io/fhHDb
20:57:23 <TrueBrain> who wants to make a PR for a news item about 1.9.0-beta1?
20:58:01 <TrueBrain> do we care if latest.html always points to what-ever is last?
20:58:07 <TrueBrain> or do we want a known url for stables?
21:00:12 <TrueBrain> Eddi|zuHause: thinking a bit more about your beta vs 1.8.1, that indeed never happens for OpenTTD
21:00:22 <TrueBrain> as when we start a new minor, we stop doing anything for the others, I guess
21:01:42 <TrueBrain> guess we never backport CVE-stuff :P
21:01:48 <Eddi|zuHause> so we can assume that continues to never happen
21:01:51 <TrueBrain> I thought we once did, but I cannot find anything back
21:02:55 <glx> anyway a new release get an increase build number
21:03:34 <TrueBrain> glx: not sure how that helps in this case?
21:03:49 <glx> for the website maybe not
21:04:12 <TrueBrain> okay, so the current assumption the website currently makes is this:
21:04:34 <TrueBrain> the latest release is always either a beta/RC or stable
21:04:47 <TrueBrain> there is always a stable dating before a beta/RC, if there is any
21:05:00 <TrueBrain> show both if there is a beta/RC, otherwise only stable
21:05:08 <TrueBrain> latest always points to either the beta/RC, or the stable
21:05:21 <TrueBrain> nothing points to only a beta/RC, and nothing only to a stable
21:05:27 <TrueBrain> are we okay with that?
21:06:27 <Eddi|zuHause> i'd be more comfortable with a "latest for only-stable"
21:06:36 <glx> hmm latest should always be the stable I think, and beta/RC are testing
21:06:41 <TrueBrain> I agree, latest should be stable
21:06:46 <TrueBrain> and beta/RC should be special
21:07:09 <TrueBrain> but the build farm has no knowledge of this information
21:07:30 <Eddi|zuHause> "/download/stable/latest.html" and "/download/testing/latest.html"?
21:07:44 <glx> can be derived from version name I guess
21:07:44 <TrueBrain> but what should testing/latest.html point to if there is no testing?
21:07:48 <TrueBrain> mind you, I cannot do redirects
21:08:04 <Eddi|zuHause> testing should include stable
21:08:42 <TrueBrain> not easy, but I can work with that
21:08:56 <Eddi|zuHause> doesn't need redirecting, just hide it from the website
21:09:10 <TrueBrain> yeah; but some people will bookmark it
21:09:13 <TrueBrain> so they will go there
21:09:27 <TrueBrain> but showing a stable in these cases is fine
21:09:29 <Eddi|zuHause> will have duplicate information then
21:09:48 <TrueBrain> okay ..... so ..... this needs rewriting a bit ..
21:10:37 <TrueBrain> I guess we are fine with a ltest.html and a testing.html
21:10:43 <TrueBrain> not a real need to have them in different folders, I guess
21:18:16 <TrueBrain> planetmaker: are you going to use that latest.txt we talked about, or is the first line from listing.txt also fine by you?
21:20:18 <planetmaker> listing.txt is required in order to obtain versions. Latest.txt is nice when it comes to updating the server as it requires less parsing
21:20:56 <planetmaker> but ... not strictly needed as I could get 'latest' by parsing the dates in listing
21:21:07 <TrueBrain> latest, as it turns out, only works for nightlies
21:21:16 <TrueBrain> latest for releases, as we just figured out, is a bit more difficult :D
21:21:21 <TrueBrain> as .. which latest :P
21:21:34 <TrueBrain> so it is easier for me to just remove it, and keep listing.txt
21:21:39 <TrueBrain> so everyone can pick what he wants from it :)
21:21:42 <TrueBrain> but .. I wanted to check
21:25:11 <TrueBrain> okay, I will remove latest.txt for now; we can always revisit :)
21:26:12 * peter1138 ponders squashing a bit of NRT, or leaving that til... the end.
21:31:41 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN commented on pull request #7204: Feature: Game setting to define how industries with neutral stations accept and supply cargo from/to surrounding stations. https://git.io/fhHy5
21:35:14 <DorpsGek_II> [OpenTTD/OpenTTD] LordAro approved pull request #7206: Fix: [AzurePipelines] manifest.sh didn't know when it was a stable release https://git.io/fhHyb
21:38:42 <TrueBrain> ty ty ty LordAro :)
21:38:42 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain merged pull request #7206: Fix: [AzurePipelines] manifest.sh didn't know when it was a stable release https://git.io/fhHDb
21:39:04 <TrueBrain> okay, this really is complex to get right, this testing/stable stuff :D
21:44:19 <TrueBrain> Eddi|zuHause: how bad do you think it would be if testing is always pointing to an beta/RC, even if there is a stable?
21:44:35 <TrueBrain> (the website shows it correct, but if you have a bookmark, you get "old" data)
21:45:10 <TrueBrain> meh; so many lines of code!
21:52:14 <peter1138> Bah, I wish it was easy to return two parameters.
21:53:13 <frosch123> TrueBrain: it's probably the best solution
21:53:55 <peter1138> nielsm, hah! I wonder if that works with squirrel :p
21:54:33 <nielsm> and it only has single return values because they wouln't want to beat lua :D
21:54:56 <DorpsGek_II> [OpenTTD/website] TrueBrain opened pull request #55: Fix: point latest.html to stable and testing.html to testing for releases https://git.io/fhHSt
21:55:01 <TrueBrain> okay ... that wasn't easy ...
21:55:15 <TrueBrain> but 'testing.html' is now either the beta/RC or stable, what ever is newer
21:55:20 <TrueBrain> and 'latest.html' is always stable
21:56:11 <peter1138> nielsm, currently I have AIVehicle::GetRoadType() and AIVehicle::GetRoadSubType() calls, which is less than ideal, but will work.
21:56:12 <DorpsGek_II> [OpenTTD/website] TrueBrain updated pull request #55: Fix: point latest.html to stable and testing.html to testing for releases https://git.io/fhHSt
21:57:08 <TrueBrain> do we wait for a beta2 btw, or do we want to release that today too?
21:57:35 <nielsm> any other easy fixes to integrate?
22:00:32 <DorpsGek_II> [OpenTTD/website] TrueBrain updated pull request #55: Fix: point latest.html to stable and testing.html to testing for releases https://git.io/fhHSt
22:00:58 <TrueBrain> glx: indeed, warnings and errors are a lot better. The docker pull error is not the only error anymore, also other entries later on. A lot more clear (not there yet, but more clear)
22:01:03 <TrueBrain> nielsm: not from my side
22:01:37 <planetmaker> tags are cheap. No reason really to delay it, is there?
22:01:54 <TrueBrain> "cheap" .. it creates a full build etc :)
22:01:59 <TrueBrain> and someone needs to bump a few files in the repository
22:02:00 <glx> ok I manage to fix EOL issues for generate_widget and squirrel_export (again)
22:02:02 *** HerzogDeXtEr has joined #openttd
22:02:07 <TrueBrain> (known-bugs, os/debian/control, ..)
22:02:15 <peter1138> ^^ let's have glx's stuff :D
22:02:26 <peter1138> It has no bearing on it ;)
22:02:29 <planetmaker> I don't mind how much the CF is used... the only thing that really matters is manpower
22:02:39 <glx> trying to do it for generate too
22:03:12 <TrueBrain> planetmaker: fair enough; most time is in the commit to prepare for a release
22:03:35 <TrueBrain> the rest is hitting a few buttons on the GitHub release page :P
22:03:41 <glx> anyway I found 2 issues in generate.vbs and generate, both not related to EOL ;)
22:04:21 <TrueBrain> that script really is getting out of hand tbh :D
22:04:35 <TrueBrain> still think they should be on the CDN, so they can be created when a build runs
22:04:39 <TrueBrain> removes most of the complexity
22:05:39 <TrueBrain> I found one bug still in the release process .. the nightly now has a name 'beta1'. But this name is currently not used anyway, so meh
22:08:08 <peter1138> Oh my, AI is building trams as well.
22:08:19 <TrueBrain> OOOEEEEHHHHH, trammmsssss :D
22:08:35 <peter1138> But because it doesn't know about tramtypes, it can only build the shittest slowest type :p
22:09:48 <peter1138> I could make the compat stuff pick the best type available, but that might change mid-game.
22:10:00 <peter1138> And "best" doesn't mean compatible. Hmm.
22:10:22 <peter1138> On the other hand, it's doing the same as before I changed the AI interface, so that's good.
22:12:24 <Samu> peter1138: do you want me to rename IsOilRig function?
22:12:40 <Samu> it already exists in master
22:12:51 <peter1138> No, that's to check if it's an oil rig.
22:15:57 <peter1138> I can't think of a good short-hand for "serve industries with attached neutral station" either.
22:16:07 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain opened pull request #7207: Update: Add changelog for 1.9.0-beta2 and prepare for release https://git.io/fhHSR
22:16:15 <frosch123> peter1138: you could also split VehicleType VT_ROAD in the api to VT_ROAD and VT_TRAM. then GetRoadType/GetTramType matches GetRailType
22:16:23 <TrueBrain> frosch123: where was your handy list again?
22:17:01 <frosch123> it's very out of date with git
22:17:35 <peter1138> api currently uses "road type" to mean road or tram.
22:17:35 <frosch123> peter1138: it would match the toolbar UI
22:18:08 <frosch123> though iirc vehicle lists are still shared in nrt
22:18:15 <TrueBrain> frosch123: mostly wanted to know if I had all the files I needed to update :) Seems I did :)
22:18:32 <frosch123> TrueBrain: make sure to update the build number for msvc
22:19:02 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain updated pull request #7207: Update: Add changelog for 1.9.0-beta2 and prepare for release https://git.io/fhHSR
22:19:36 <DorpsGek_II> [OpenTTD/OpenTTD] frosch123 requested changes for pull request #7207: Update: Add changelog for 1.9.0-beta2 and prepare for release https://git.io/fhHSu
22:20:24 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain updated pull request #7207: Update: Add changelog for 1.9.0-beta2 and prepare for release https://git.io/fhHSR
22:20:28 <TrueBrain> possibly '2' was a better value, but okay
22:20:48 <frosch123> rc1 will be 2, and stable will be 3
22:20:55 <frosch123> so there is no sense in it anyway
22:21:24 <TrueBrain> weird, it didnt dismiss your review
22:21:43 <frosch123> it doesn'T show any change yet as well
22:21:53 <frosch123> still says "0" for me
22:22:10 <TrueBrain> it is VERY hidden :P
22:22:20 <TrueBrain> (almost on the top)
22:23:41 <DorpsGek_II> [OpenTTD/OpenTTD] frosch123 approved pull request #7207: Update: Add changelog for 1.9.0-beta2 and prepare for release https://git.io/fhHS2
22:25:06 <peter1138> but that doesn't exist :/
22:26:28 <peter1138> This seems to be tons of API changes :(
22:26:44 <frosch123> well, if you break the script api unrecoverable, we can also upgrade to squirrel3 :p
22:27:24 <peter1138> That's not quite my aim!
22:31:18 <TrueBrain> but a nice benefit? :P
22:32:52 <DorpsGek_II> [OpenTTD/website] TrueBrain merged pull request #55: Fix: point latest.html to stable and testing.html to testing for releases https://git.io/fhHSt
22:32:55 <andythenorth> major version API change :P
22:33:21 <TrueBrain> any objections against me merging this beta2?
22:35:33 <peter1138> I think splitting road and tram just for the AI is too intrusive.
22:36:27 <peter1138> Would end up with tons of compat shims as well.
22:39:21 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain merged pull request #7207: Update: Add changelog for 1.9.0-beta2 and prepare for release https://git.io/fhHSR
22:39:54 <TrueBrain> I started 1.9.0-beta2 build :)
22:42:34 <TrueBrain> so is anyone willing to write a news post? :D Otherwise I just push this to production :P (website)
22:44:12 *** supermop_Home has joined #openttd
22:46:06 <TrueBrain> hmm .. I should wait till beta2 is done building :D
22:49:39 <peter1138> Oh yeah, I ought to make regression. Hmm.
22:51:45 <peter1138> Ah, it doesn't use any compat shims.
22:51:58 <peter1138> Damn it, I need to make new regression tests :-(
22:59:49 <peter1138> How do I update result.txt?
23:01:43 <peter1138> This options parsing is terribel :)
23:06:47 <peter1138> Sorry, no VT_TRAM ... yet?
23:07:05 <TrueBrain> lol .. of course the weekly build to update the dependencies of the CF kicked in .. there goes all the jobs
23:08:18 <TrueBrain> nah; my jobs are queued before yours :P
23:08:23 <TrueBrain> it was the weekly job that stole all the agents
23:08:57 <TrueBrain> I created a race condition based on a race condition :D
23:09:48 <peter1138> Good job it's staging.
23:09:58 <TrueBrain> I just pushed this version to production too :P
23:10:21 <TrueBrain> not going to fix something that most likely only happens once in 10 years :P
23:10:45 <TrueBrain> (it broke because we had a beta before the nightly and after the nightly, on the same day)
23:10:50 <TrueBrain> I mean .. how often will that happen :D
23:11:50 <TrueBrain> yes; and it will most likely never happen again :D
23:12:23 <TrueBrain> but I will fix this! >:D
23:12:27 <TrueBrain> *goes into evil mode*
23:17:42 <TrueBrain> thar, 'fixed' it :P
23:17:48 <TrueBrain> sed -i s/1\.9\.0\)/1.9.0-beta1\)/ */*.html
23:18:06 <TrueBrain> peter1138: is NRT stable to be build?
23:18:25 <peter1138> Should be, regression tests pass now.
23:18:35 <peter1138> Candidate for 1.9 I reckon ;p
23:18:39 <TrueBrain> hmm ... no, I am going to wait till the next nightly
23:18:45 <TrueBrain> otherwise I have to fix the production pods again :P
23:19:42 <TrueBrain> and no peter1138! No! 2.0 ! :D
23:20:47 <TrueBrain> peter1138: a nightly of 2.0, sure :P
23:21:20 <TrueBrain> nah; I don't really care :)
23:21:26 <peter1138> Maybe we'll have time to rewrite it then.
23:21:36 <peter1138> Split road and tram into separate features like frosch123 said.
23:21:56 <peter1138> (Not really going to work)
23:21:57 <TrueBrain> with all this new infrastructure, releases are a lot easier. So possibly we can go to 2 a year :P
23:22:20 <peter1138> Hmm, how long did newstations take me...
23:22:32 <peter1138> Felt like years at the time, but probably wasn't.
23:23:23 <TrueBrain> @topic set 1 1.9.0-beta2, 1.8.0
23:23:23 *** DorpsGek changes topic to "1.9.0-beta2, 1.8.0 | Website: *.openttd.org (source: github, translator: translator, server list: servers, wiki: wiki, release info: finger) | Don't ask to ask, just ask | 'Latest' is not a valid version, 'Most recent' neither | English only | Logs: @logs | #openttd.dev if this channel is really spammy"
23:24:11 *** DorpsGek sets mode: +o TrueBrain
23:24:52 <TrueBrain> hmm .. changing topic with this client is not possible, it seems ... lol
23:25:02 *** DorpsGek sets mode: -o TrueBrain
23:25:38 <frosch123> what's wrong with it?
23:26:02 <TrueBrain> few things are outdated, but .. hmm
23:26:09 <TrueBrain> nielsm: it should, yes
23:26:21 <frosch123> it's so complex that noone understands anything of it :p
23:26:34 *** TrueBrain is now known as TrueBrain_iii
23:26:38 *** TrueBrain_ii is now known as TrueBrain
23:26:40 *** DorpsGek sets mode: +o TrueBrain
23:27:13 *** TrueBrain changes topic to "1.9.0-beta2, 1.8.0 | Website: *.openttd.org (source: github, translator: translator, server list: servers, wiki: wiki) | Don't ask to ask, just ask | 'Latest' is not a valid version, 'Most recent' neither | English only"
23:27:21 <TrueBrain> I cut a bit of the fat
23:27:27 *** DorpsGek sets mode: -o TrueBrain
23:27:31 *** TrueBrain is now known as TrueBrain_ii
23:27:37 *** TrueBrain_iii is now known as TrueBrain
23:28:04 <TrueBrain> right, changelog .. so what broke there
23:29:09 <frosch123> doesn'T it always only show changes since last release?
23:29:13 <TrueBrain> nielsm: turns out, that is intended, to always do this
23:29:29 <TrueBrain> not what I expected, but indeed
23:30:01 <TrueBrain> except for releases
23:30:04 <TrueBrain> those list everything
23:30:21 <TrueBrain> but .. azure-pipelines/changelog.sh
23:31:11 <TrueBrain> yeah, only for releases it walks every prerelease
23:31:36 <peter1138> Server version: 1.9.0-g6e21190858
23:32:03 <peter1138> It's dropped the beta2 :/
23:32:46 <TrueBrain> something to fix for beta3? :D
23:33:01 <peter1138> Hmm, server list behaves oddly.
23:33:02 <TrueBrain> nielsm / frosch123: not sure why we only show latest release, but it is an easy fix to show all; not sure what we want?
23:33:19 <peter1138> (In game list, that is)
23:33:36 <peter1138> When I click on my server, the list jumps to the bottom and some other server is selected.
23:33:48 <nielsm> imo for changelogs it should always show everything since last stable release
23:33:49 <TrueBrain> oddly defines that well :D
23:34:29 <peter1138> Well, anyway, I have a 1.9.0(-beta2) server up :p
23:34:39 <peter1138> How do you play this game?
23:35:27 <TrueBrain> does anyone disagree with nielsm? :)
23:35:34 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain opened pull request #7208: Fix: [AzurePipelines] always list the full changelog since last stable https://git.io/fhH9g
23:36:04 <frosch123> TrueBrain: that kind of list is actually only needed for the nightly. tags should instead link to the written changelog.txt
23:37:04 <frosch123> so, changelog.txt would be a build artifact like the readme.md
23:37:18 <TrueBrain> there is only a script that makes it smaller
23:37:26 <TrueBrain> as someone found it important somewhere in the last 15 years :D
23:37:38 <TrueBrain> (azure-pipelines/changelog.sh)
23:38:01 <TrueBrain> nielsm: updated the beta2 changelog :)
23:38:12 <frosch123> oh. all fine then :)
23:38:36 <TrueBrain> frosch123: we can still copy the whole changelog
23:38:44 <TrueBrain> but I guess the reasoning was: THAT IS WASTING BYTES, or something
23:39:13 <frosch123> do you know the reason why c uses == for comparison and = for assignment? :p
23:39:43 <peter1138> To annoy VB developers.
23:40:31 <TrueBrain> anyway, we have to go to 2.0, not 1.10, otherwise the version will be 1.10.-gblabla
23:41:25 <frosch123> BCPL had the usual := and =. B changed them to = and == to save memory, since source code contains more assignments than comparisons
23:41:32 <TrueBrain> you just showed -beta2 doesnt fit peter1138 :P
23:41:34 <peter1138> Hmm, this very expensive ginger hot-chocolate is... a bit rubbish :/
23:41:51 <peter1138> TrueBrain, I don't think it's the length :)
23:42:06 <TrueBrain> that was an assumption on my part :P
23:42:15 <frosch123> TrueBrain: can we also go to 2020 ?
23:42:20 <TrueBrain> guess the same issues as I have with the nightly are also in the game :P
23:42:50 <TrueBrain> I don't follow, sorry :(
23:42:52 <TrueBrain> its getting late :(
23:43:11 <frosch123> i like that 1.0 and 1.9 match up with the release year, makes it easy to associate features
23:43:26 <TrueBrain> even more reason to go to 2.0 :D
23:43:28 <frosch123> so we could change ottd versioning to release-year
23:43:35 <nielsm> feature request for version 2.0: change the original vehicle set and adjust monorail and maglev to more closely reflect the reality of today :P
23:43:42 <peter1138> Damn, too many passengers already.
23:44:06 <peter1138> nielsm, put all the default vehicles into a NewGRF and then... have no defaults!
23:44:43 <nielsm> get rid of climates and have it all be grf-defined
23:45:13 <nielsm> (a great way to break all backwards compat)
23:45:17 <peter1138> Hmm, not making money yet.
23:45:30 <frosch123> nielsm: also squirrel3 :p
23:45:33 <peter1138> Maybe I should add 15 AIs and just watch them.
23:45:44 <TrueBrain> guess we should make a list of shit we want to break :P
23:46:04 <TrueBrain> frosch123: honestly, from what I have seen, squirrel3 doesn't really break anything
23:46:08 <peter1138> Hmm, should I lay some tracks.
23:46:16 <TrueBrain> I have yet to scan all the AIs, but it is easily fixed, most things
23:46:17 <nielsm> scrap newgrf and squirrel, all mods are now javascript packages with png files
23:46:27 <TrueBrain> nielsm: javascript? JAVA!
23:46:33 <TrueBrain> and make it Minecraft compatible!
23:47:15 <frosch123> of course ottd should become a factorio mod
23:47:26 <nielsm> implement ottd in minecraft
23:47:35 <TrueBrain> you guys are crazy :P
23:48:02 <nielsm> so, about the 1.9.0 title game...
23:48:16 <nielsm> I made this little toyland title game for fun a while ago
23:48:29 <TrueBrain> that or hope someone makes a contest out of it again
23:48:33 <TrueBrain> *pauses for effect*
23:50:57 <andythenorth> peter1138: break passengers :P
23:51:00 <peter1138> People have been making savegames for 1.9 on reddit.
23:51:17 <andythenorth> wasn't me on reddit btw
23:51:21 <peter1138> Wasn't there a patch for it?
23:51:43 <peter1138> Ooh, a river heads right past a factory.
23:51:54 <peter1138> Maybe I should prevent the town bridging it.
23:52:20 <andythenorth> remember the nuke button in lemmings? o_O
23:52:24 <andythenorth> can we have that in ottd?
23:52:32 <peter1138> I'm looking for a farm I can supply it from.
23:52:40 <andythenorth> there is a good JS version of lemmings btw
23:55:26 <andythenorth> wolf vs chainsaw
23:56:09 <peter1138> That needs VR support.
23:57:29 <TrueBrain> yeah; the demo is fun :)
23:57:49 <peter1138> Oh, they released the game now, early access.
23:57:53 <TrueBrain> derailed so many trains ...
23:58:08 <peter1138> There's no chainsaws and wolves though.
23:58:20 <andythenorth> derail looks less good :P
continue to next day ⏵
