IRC logs for #openttd on OFTC at 2019-03-04
⏴ go to previous day
00:00:27 <Beerbelott> and I'm no developer, that's the funny part :D
00:00:41 <peter1138> Renaming a PR is easy. Just don't bother renaming branches, that just leads to pain.
00:01:29 <Beerbelott> peter1138: I already made a mistake or deleting a branch a few days ago, remember? ;)
00:01:42 <peter1138> And someone else did it today...
00:01:45 <peter1138> " If you wish, you can delete your fork of OpenTTD/OpenTTD. "
00:01:54 <peter1138> Cheers GitHub, yeah, I'd really want to do that...
00:01:55 <Beerbelott> my branch is called 'heightmap' anyway. can't think of a better name :D
00:02:12 <LordAro> peter1138: i guess if you've forked just to fix one particular issue...
00:02:13 <Beerbelott> I'm branching as hell on my part...
00:02:32 <peter1138> Yeah but it's so prominent on something I perceive as rare.
00:02:35 <peter1138> Maybe it's not rare for them.
00:02:52 <Eddi|zuHause> <peter1138> " If you wish, you can delete your fork of OpenTTD/OpenTTD. " <-- i also saw this button today, and was wondering "that looks awfully easy to press accidentally" :p
00:02:56 <peter1138> They probably have more metrics on us that I have ;)
00:03:38 <peter1138> m3henry, table/newgrf_debug_data.h has a load as well.
00:04:03 <peter1138> I really wish this was a default-on check :/
00:04:19 <peter1138> If you have nothing marked as override then it doesn't care.
00:04:43 <DorpsGek_II> [OpenTTD/OpenTTD] M3Henry updated pull request #7324: Codechange: [C++11] Use override specifer for overriding member declarations https://git.io/fhxq0
00:08:54 <Beerbelott> I suppose GitHub hates all those fork sitting there idle ;) Less forks = more gc
00:10:00 *** Lejving has joined #openttd
00:11:13 <TrueBrain> LordAro: sentences end with a dot
00:11:26 <LordAro> TrueBrain: sorry about the local branch, i couldn't (trivially) figure out a way to create the branch on my fork via the web interface
00:11:37 <DorpsGek_II> [OpenTTD/OpenTTD] M3Henry updated pull request #7324: Codechange: [C++11] Use override specifer for overriding member declarations https://git.io/fhxq0
00:11:38 <TrueBrain> you are them all :p
00:11:58 <TrueBrain> the same way you did on the main branch :)
00:12:11 <TrueBrain> but it doesn't auto sync.. so yeah
00:12:26 <TrueBrain> just delete it afterwards ;)
00:12:28 <LordAro> it'll be gone in a few moments, anyway :p
00:12:53 <TrueBrain> after you found the dot key on your keyboard :p
00:13:54 <peter1138> m3henry, just 4 left in newgrf_debug_data from what I can see.
00:14:00 <peter1138> m3henry, :363 365 529 and 537
00:14:03 <Eddi|zuHause> (i had to try whether that works)
00:14:09 <TrueBrain> still more missing dots :p
00:14:14 <peter1138> I dunno about other non-Linux builds though.
00:14:22 <TrueBrain> I like this collaborated way of news writing tbh
00:14:34 <TrueBrain> possibly we should always do it like this
00:15:38 <TrueBrain> can't approve on mobile :(
00:17:04 <TrueBrain> time to get some sleep; night all
00:17:38 <DorpsGek_II> [OpenTTD/OpenTTD] M3Henry updated pull request #7324: Codechange: [C++11] Use override specifer for overriding member declarations https://git.io/fhxq0
00:18:43 <peter1138> I wonder if there are any classes which are missing.
00:19:29 <LordAro> some static analyser should be able to tell you
00:19:38 <LordAro> cppcheck has something, i think
00:21:21 <peter1138> "Failed to load std.cfg. Your Cppcheck installation is broken"
00:22:12 <Eddi|zuHause> so who can merge the news post? it says i can't.
00:28:13 <snail_UES_> peter1138: saw your comment
00:28:25 <snail_UES_> I’m using a version of OTTD from July 2018, has the code changed since?
00:32:20 <snail_UES_> then there must be something weird with my NFO
00:36:29 <peter1138> Did you accidentally go with A0 instead of 0A? :p
00:37:05 <Samu> why was it closed? I told you what happened :|
00:37:49 <Samu> just downloaded RC-1, gonna test some AIs
00:46:43 <snail_UES_> I wrote: “ 17 B6 FA 0A 00 "
00:50:08 <supermop_Home_> safe to assume nrt isn't going to move forward any time soon?
00:50:20 <peter1138> supermop_Home_, nope
00:50:31 <peter1138> supermop_Home_, it will
00:50:34 <supermop_Home_> so i don't need to start working on my RV set again yet
00:50:35 <peter1138> supermop_Home_, 1.9 was branched.
00:50:51 <peter1138> m3henry, I still see some /* virtual */ in cpp files.
00:51:07 <supermop_Home_> hmm. i miss a good generic station set
00:51:29 <peter1138> You're right, there are none. You should make one ;)
00:51:33 <supermop_Home_> i just use chips all the time lately.
00:51:40 <Eddi|zuHause> newstats is a passenger station set
00:51:42 <supermop_Home_> i am waiting for NML
00:52:05 <Eddi|zuHause> i wouldn't hold my breath on that :p
00:52:28 <supermop_Home_> chips is ok for cargo but a little bland - all those paved flat lots stacked 10m high with crates
00:53:16 <supermop_Home_> the default stations with just a tad more richness would be 90% of what I desire
00:53:17 <Beerbelott> What are the valid prefixes for commit messages again? I got a list from a fail from commit-checker the other day but did not make a printscreen
00:53:23 <Eddi|zuHause> i was thinking of a distribution of which cargo type is displayed on each tile, so not only the highest one gets shown, but it gets somewhat mixed
00:53:47 <supermop_Home_> i can't even understand the code i wrote for MLSS ~9 years ago at this point
00:54:31 <Beerbelott> LordAro: Many thx!
00:54:44 <Eddi|zuHause> not sure how up to date that is
00:54:54 <supermop_Home_> hmm i did try some rendered station sprites about 4 years ago
00:55:32 <supermop_Home_> but I've sort of wandered away from rendering things that are not for work these days
00:55:46 <Eddi|zuHause> it seems updated, but with wiki you can never be sure :p
00:56:18 <supermop_Home_> Eddi|zuHause a 60-40 mix of cargo for chips would be nice
00:56:53 <supermop_Home_> as is, my station with 1000T of coal looks like it has none if it has 1001T of Iron Ore
00:57:38 <Eddi|zuHause> (not 100% serious)
00:59:24 <supermop_Home_> of course if you build a Victorian passenger terminal hall for your coal mine, should it still show coal piled up on the platforms?
01:00:12 <Eddi|zuHause> that is a bit of a problem, if you "hide" some cargos, you're not sure if the remaining cargos can be displayed on any given tile...
01:00:48 <peter1138> Station-cargo-storage.
01:01:12 <Eddi|zuHause> the station tile would have to provide a list of cargos to the distribution algorithm
01:01:20 <supermop_Home_> yeah i guess you could just make people build coal hoppers to see coal
01:01:40 <Eddi|zuHause> so you can mix bulk pile tiles with crate stack tiles
01:01:41 <supermop_Home_> people hoppers to see passengers
01:02:19 <supermop_Home_> Eddi|zuHause essentially ISR, but i think something between chips and ISR would be nice
01:02:28 <Eddi|zuHause> to prevent the distribution algorithm to assign the crate cargos to the bulk tiles
01:02:53 <Eddi|zuHause> and end up with empty tiles
01:03:38 <supermop_Home_> it would be nice if this algorithm could pick what tiles to get what class
01:03:49 <supermop_Home_> so you don't need to build them
01:04:36 <supermop_Home_> chips is nice in that you can quickly build a generic station of the size you need and it will look not too out of place
01:10:20 <m3henry> peter1138: I wasn't sure if those would still be wanted, so I left them. What do others think?
01:24:37 <Samu> is there anything wrong with the PR other than missing a description?
01:24:48 <Samu> just added a description w
01:27:35 <Samu> You're making me feel useless here :(
01:53:19 *** drac_boy has joined #openttd
01:55:07 <drac_boy> are they generally independent of each others or could you somehow have an industry that has two outputs with the second one being always around 10% of the first one?
01:55:14 <drac_boy> its single input for the note
01:58:44 <supermop_Home_> i think FIRS steeltown slag works that way
01:59:43 <supermop_Home_> it might not be 10%, but blast furnace, arc furnace, and steel furnace all produce some small % of slag compared to the metal output
02:00:22 <drac_boy> hmm I'll see if I want to try sort out that source but thanks for the extra thought btw :)
02:01:59 <drac_boy> hows the homebound mop tonight btw?
02:07:51 * drac_boy just has a long messy industry list to sort :->
02:08:55 <peter1138> Surely you just need to announce big plans, get overwhelmed with the work required, ask other people to do it for you, and then leave the forums in a rage quit, optionally deleting/sabotaging everything you posted... No?
02:09:25 <drac_boy> hmm nope not even my type :)
02:09:43 * drac_boy actually is going start with a somewhat bare train release and add onto it over time
02:20:43 <Eddi|zuHause> opening 10 threads and requesting an own subforum sometimes helps?
02:21:32 * drac_boy will rather just post in the graphic dev section like everyone else
02:36:22 <drac_boy> anyway enjoy your day/night both of you
03:51:07 *** Wormnest has joined #openttd
04:43:26 *** Beerbelott has left #openttd
04:52:13 *** tokai|noir has joined #openttd
04:52:13 *** ChanServ sets mode: +v tokai|noir
05:35:41 *** supermop_Home_ has quit IRC
07:01:58 *** snail_UES_ has joined #openttd
08:05:55 *** HerzogDeXtEr has joined #openttd
08:06:21 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN commented on pull request #7324: Codechange: [C++11] Use override specifer for overriding member declarations https://git.io/fhxs8
08:17:31 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN commented on pull request #7324: Codechange: [C++11] Use override specifer for overriding member declarations https://git.io/fhxs2
09:20:55 <DorpsGek_II> [OpenTTD/OpenTTD] M3Henry commented on pull request #7324: Codechange: [C++11] Use override specifer for overriding member declarations https://git.io/fhxG4
09:51:18 *** m3henry has joined #openttd
10:31:16 *** WWacko1976-work has joined #openttd
10:34:34 *** peter1138 has joined #openttd
10:34:34 *** ChanServ sets mode: +o peter1138
10:34:49 <peter1138> Well, VPS got rebooted :/
10:42:10 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN commented on pull request #7235: Change: Non-rectangular sparse station catchment area https://git.io/fhxZP
10:56:06 <peter1138> Weird, forum says I have an unread post in Graphics Development, but there isn't one there.
11:09:20 <Eddi|zuHause> yeah. it does that sometimes
11:09:31 <Eddi|zuHause> never figured out why
11:15:47 <peter1138> Might be a moderated post.
11:16:09 <planetmaker> are there ever moderated posts?
11:16:22 <Eddi|zuHause> these were both my exact thoughts :p
11:17:20 <peter1138> You're a mod, you'd know :p
11:17:23 <planetmaker> I can't come up with any instance where it was used... posts are moved in worst case to spam (which is not visible to anyone but mods), but not hidden or deleted
11:17:41 <peter1138> Yup. And that would be a moderated post.
11:17:57 <planetmaker> it's not in graphics forum then anymore. it's in spam forums then
11:18:00 <Eddi|zuHause> planetmaker: well, but some caching might still show up the moved post as "unread" in the original forum
11:18:28 <planetmaker> it's a separate high-level entry in the menu like OpenTTD itself
11:18:32 <Eddi|zuHause> planetmaker: in some weird corner cases.
11:20:17 <Eddi|zuHause> planetmaker: something like 1) someone posts, 2) the forum marks the forum as "has unread posts", 3) post is moved, 4) "has unread posts" flag is not correctly removed from the forum
11:21:19 <Eddi|zuHause> planetmaker: well, you could at least look whether there were recently moved posts. i don't have that ability
11:21:19 <peter1138> Hmm, another Debian OpenSSH update.
11:21:35 <peter1138> Nah, if it's moved, it's moved.
11:22:01 <peter1138> I don't see anything recent in the spam bin.
11:22:36 <peter1138> I have now clicked on "mark all read" and will get on with my life :-)
11:22:43 <planetmaker> owen might use special owner priviliges to actually mass-delete stuff from mass-spammers. maybe that affects things differently
11:23:19 <peter1138> m3henry, is submitted a PR to you useful? I figured directly pushing to your branch would be impolite.
11:30:59 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN commented on pull request #7301: Feature: When filtering purchase list by cargo type, make buy button perform a refit if required. https://git.io/fhxnH
11:58:29 *** WWacko1976-work has quit IRC
12:26:22 <peter1138> dpkg: warning: downgrading openttd from 1.9.0~beta3-0 to 1.9.0~RC1-0
12:30:10 <Heiki> rc1 would have been better than RC1
12:30:54 <planetmaker> it usually should have a sequential release numer afaik... alas
12:31:27 <peter1138> I also wonder if 1.9.0 will be considered newer.
12:35:46 <planetmaker> if b>R, then R<b<[:space:]
12:38:50 <Heiki> and the footnote states “One common use of ~ is for upstream pre-releases. For example, 1.0~beta1~svn1245 sorts earlier than 1.0~beta1, which sorts earlier than 1.0.”
12:40:27 <peter1138> Ok, so ~ is fine, just RC should be rc?
12:41:28 <Heiki> (or beta should have been Beta or BETA or BeTa or something)
12:42:33 <planetmaker> ~rc next time, I guess :)
12:42:46 <Heiki> yes, caps is for MS-DOS users
12:46:32 <peter1138> Should the tag itself be lowercase? Hmm.
12:54:35 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh commented on pull request #7235: Change: Non-rectangular sparse station catchment area https://git.io/fhxCd
12:56:02 <LordAro> peter1138: never has been previously
12:56:38 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh commented on pull request #7301: Feature: When filtering purchase list by cargo type, make buy button perform a refit if required. https://git.io/fhxCF
12:58:17 <Eddi|zuHause> so, next developer blog comprehensive overview what will happen with NRT?
13:00:41 <peter1138> It'll sit in a branch for 2 years ;D
13:05:52 <peter1138> LordAro, okay, so the stuff that makes the packages needs to take care of lowercasing the tag.
13:23:28 <peter1138> LordAro, now I feel bad. I'm going through all my PRs adding proper descriptions!
13:47:30 <Eddi|zuHause> so it must be 1.10~alpha, 1.10.0~beta and 1.10.0~rc?
14:03:05 *** goodger has joined #openttd
14:04:43 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN updated pull request #7235: Change: Non-rectangular sparse station catchment area https://git.io/fh5s1
14:10:57 <Eddi|zuHause> it's now done? :p
14:11:11 <peter1138> I think it's done enough to be merged.
14:11:28 <peter1138> Unless someone wants to go back and rewrite it to not use two separate types, or something.
14:12:01 <Eddi|zuHause> all the other options were also not terribly optimal
14:12:37 <peter1138> This one has the advantage of having been written, and having NewGRFs already produced for it, and being fairly well tested, but probably not extensively.
14:13:41 <Eddi|zuHause> "lieber den Spatz in der hand als die Taube auf dem Dach"
14:16:27 <peter1138> Sparrows and pigeons, eh?
14:17:03 <Eddi|zuHause> it's a silly saying, about how you should go with the thing you already hold in hands instead of striving for the potentially bigger thing that is just out of reach
14:17:52 <Eddi|zuHause> (assuming you can't just reach on your roof)
14:22:15 <planetmaker> a bird in the hand it worth two in the bush ;)
14:30:32 <Heiki> in Finnish it’s “parempi pyy pivossa kuin kymmenen oksalla” – “a hazel grouse in the hand is better than ten on a branch”
14:30:57 <peter1138> Well, all you need to do is review the patch and... approve it ;)
14:34:13 <Eddi|zuHause> oh, if that is all... :p
14:47:30 *** sla_ro|master has joined #openttd
15:02:24 *** Beerbelott has joined #openttd
15:11:19 <Eddi|zuHause> so i've been downloading this 13GB thing, and for the last two hours it has said "remaining time: 2 hours"
15:18:36 <LordAro> peter1138: am without bike until (at least) thursday :(
15:20:03 <Eddi|zuHause> easy, just use your car. (or don't leave your house) (or use your legs) (or ...)
15:21:00 <Eddi|zuHause> ok, so i managed to get pidgin to connect to steam, but all my steam friends are offline so i don't know whether sending/receiving messages works :p
15:22:12 <LordAro> Eddi|zuHause: well yes, i have to bus it
15:22:23 <LordAro> but that takes 3 times as long as biking
15:23:10 <planetmaker> LordAro, my way to work even taks by car longer than by bike ;)
15:23:15 <Eddi|zuHause> i once had a commute that took 40 minutes by car, or 2 hours by bus/train
15:24:16 <LordAro> peter1138: the shop rang me up and said something along the lines of "the headset fell apart in my hands"
15:24:27 <LordAro> oh, and they need to order a new bottom bracket
15:24:48 <LordAro> turns out winter is pretty hard on bikes
15:24:55 <Eddi|zuHause> that means they smashed it with a heavy automatted tool :p
15:25:28 <Eddi|zuHause> damn, who invented this winter thing. can't global warming happen already?
16:00:41 <sara> how can i download pipelines
16:00:52 <nielsm> oh yes you could have houses with production callback so some house types produce school children at some rate, and you have to deliver those to school industries
16:01:12 <sara> how to download pipelines?
16:01:21 <nielsm> the school produces educated young adults which you then have to deliver to other industries
16:01:45 <nielsm> sara: pipalines, as in large pipes to transport fluid goods?
16:01:58 <nielsm> (because those don't exist, as far as I know)
16:08:38 <peter1138> So your linear population thing...
16:09:08 <peter1138> if (GB(_tick_counter, 8, 2) == GB(tile, 0, 2)) {
16:10:00 <peter1138> Maybe wants a comment that it spreads out overall generation.
16:10:10 <peter1138> Oh wait, why am I talking here :D
16:10:31 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN commented on pull request #6965: Add: Option for population-linear town cargo generation https://git.io/fhxB2
16:11:48 <DorpsGek_II> [OpenTTD/OpenTTD] nikolas commented on pull request #7086: Change #6173: Update SDL driver to use SDL 2.0 https://git.io/fhxBV
16:12:37 <DorpsGek_II> [OpenTTD/OpenTTD] nikolas updated pull request #7086: Change #6173: Update SDL driver to use SDL 2.0 https://git.io/fhamZ
16:23:49 <DorpsGek_II> [OpenTTD/OpenTTD] nikolas commented on issue #7296: Can't type non-latin symbols from keyboard in game on linux https://git.io/fhxBF
16:24:41 <blelch> Is there a rule of thumb for when one should use Roads and when, Rails?
16:26:52 <Eddi|zuHause> use roads (or trams) if you have low space
16:28:47 <peter1138> Rail is more profitable.
16:29:55 <Eddi|zuHause> especially when you go diagonal
16:30:54 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN commented on pull request #6965: Add: Option for population-linear town cargo generation https://git.io/fhxBj
16:32:57 *** octernion has joined #openttd
16:36:31 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh commented on pull request #6965: Add: Option for population-linear town cargo generation https://git.io/fhxRt
16:39:03 <DorpsGek_II> [OpenTTD/OpenTTD] Eddi-z commented on pull request #6965: Add: Option for population-linear town cargo generation https://git.io/fhxRY
16:46:45 <DorpsGek_II> [OpenTTD/OpenTTD] Nik-mmzd commented on issue #7296: Can't type non-latin symbols from keyboard in game on linux https://git.io/fhxRw
16:49:18 <nielsm> I found multi-track mode
16:54:07 <blelch> Eddi|zuHause: there are Trams in the game!!??
16:54:47 <blelch> peter1138: Why is rail more profitable?
17:00:40 *** synchris has joined #openttd
17:00:51 <peter1138> Basically trains can go faster and carry more.
17:02:46 <Eddi|zuHause> blelch: yes, but you need NewGRFs to activate them
17:03:17 <Eddi|zuHause> HEQS has some nice trams for cargos
17:03:41 <Eddi|zuHause> they look like mini trains
17:11:38 *** Wormnest has joined #openttd
17:15:18 <DorpsGek_II> [OpenTTD/OpenTTD] GabdaZM commented on pull request #7250: K-d tree data structure for spatial lookups https://git.io/fhx0l
17:18:04 <Gabda> nielsm: you can discard any of my comment, I am just trying to learn how to give a review
17:18:46 <Gabda> so far I've only looked into the first commit
17:39:38 <Beerbelott> Is there a way to accept a PR on a branch used for another PR on GitHub in another branch?
17:41:14 <Eddi|zuHause> how do you mean?
17:44:06 <Beerbelott> I had 3 options on GitHub: merge, squash & rebase
17:44:23 <Beerbelott> But all those options alter the branch on which the PR was made
17:44:58 <Beerbelott> and since this branch was used in another PR (on OpenTTD), it triggered validation automatically (and failed them which was expected)
17:45:21 <Beerbelott> Is there a way I could have accepted the PR, but by making another branch on GitHub?
17:45:23 <glx> yes I made this PR for the branch you used as PR
17:45:52 <glx> and no PR are merged only in the intended branch
17:46:14 <Beerbelott> glx: Ah damn that means there was no was to trigger validation on ur changes
17:46:29 <Beerbelott> I needed some time to integrate & squash them into my own changesets
17:46:30 <glx> but you could have edited the squash message :)
17:47:03 <glx> you can still use "git rebase -i HEAD^" to edit the last commit
17:47:23 <Eddi|zuHause> or git commit --amend
17:47:26 <Beerbelott> The problem is not that valdiation failed, I don't care, it's that your PR was part of WIP, not supposed to be pushed to the other PR
17:47:45 <Beerbelott> I do all that stuff once I can pull them from GitHub yes
17:47:54 <Beerbelott> But 1st you have to accept the PR :)
17:48:01 <Beerbelott> and that's where the problem lie
17:48:17 <Eddi|zuHause> i still don't understand your problem
17:48:29 <Beerbelott> I know all of rebasing & cherry-picking now... I practiced quite a lot to push some clean commits to my PR :)
17:48:44 <Beerbelott> (I mean 'all'... well... 'enough' ;) )
17:48:46 <glx> that's how github works, but once the PR is merged to your branch you can do anything to it
17:49:02 <Beerbelott> glx: Yes I know... and did ;)
17:49:13 <Beerbelott> I pushed clean commits again
17:49:25 <glx> anyway forced push are allowed, so not a real issue ;)
17:50:02 <Beerbelott> it's just that in the meantime temporary work was pushed to the OpenTTD PR... But well not that big of a problem anyhow
17:50:38 <Beerbelott> I happily was able to start a test OpenTTD dedicated network on a heightmap file.
17:50:51 <Beerbelott> It *seems* to work :\
17:51:10 <Beerbelott> (ofc it doesn't and I am still to experience it :D )
17:52:42 <Eddi|zuHause> Beerbelott: if you wanted to do other things to the PR that isn't just merge/squash/rebase, then you could modify the PR beforehand, or you can pull it to your local repo and cherry-pick it
17:53:01 <Eddi|zuHause> (and then dismiss the PR)
17:53:14 <glx> that's an option too indeed
17:56:07 <Beerbelott> Eddi|zuHause: How do you pull a PR w/o merging it?
17:56:14 <Beerbelott> That I did not know of
17:56:58 <Beerbelott> That's precisely what I would have wanted to do
17:59:02 <Eddi|zuHause> not entirely sure, refs/pull/something or so
18:00:04 <Eddi|zuHause> refs/pull/<number>/head
18:01:08 <Eddi|zuHause> make sure you pull that from your remote, not the upstream remote :p
18:07:07 <Beerbelott> Eddi|zuHause: git fetch origin pull/ID/head:BRANCHNAME
18:07:17 <Beerbelott> Waow is that undocumented?
18:07:25 <Beerbelott> Found the soluation on StackOverflow -_-'
18:07:35 <glx> no, it's documented somewhere on github
18:11:40 <glx> btw in dedicated_v.cpp you may want to use StartNewHeightMapGameWithoutGUI() ;)
18:12:15 <Beerbelott> glx: It's done through the call of SwitchToMode
18:12:26 <Beerbelott> Oh lol same link :D
18:12:41 <Beerbelott> Well I should have thoroughly searched beforehand then...
18:15:30 <Beerbelott> This workflow seems a bit odd to me
18:16:10 <Beerbelott> dedicated_v.cpp calls SafeLoad directly, and then hands over the job to SwitchToMode... which does that again and this time properly initialize stuff I suppose
18:16:27 <Beerbelott> just to check of the file is loadable (and print something specific)
18:16:28 <glx> dedicated just checks it can load
18:16:42 <glx> and fall back to random game
18:16:42 <Beerbelott> yeah but SafeLoad does more than merely checking; right?
18:17:03 <Beerbelott> it actually loads sth
18:17:11 <Beerbelott> thus there is a double load, isn't it?
18:17:28 <glx> hmm no it cancel server start if load fails
18:17:57 <Beerbelott> I do not get this direct call anyway
18:18:03 <Beerbelott> since SwitchToMode does it
18:18:38 <Beerbelott> Thus for heightmaps, I merely call SwitchToMode propagating _switch_mode
18:19:02 <glx> SwitchMode() doesn't touch _networking
18:19:29 <Beerbelott> seems to work, although I lack global understanding
18:20:16 <glx> it's used in all situations, but dedidacted server needs special care, as you don't want your headless server to return to main menu
18:22:24 <Beerbelott> but isn't calling StartNewHeightMapGameWithoutGUI from dedicated_v.cpp a problem?
18:23:24 <glx> there's already StartNewGameWithoutGUI() when not using -g
18:23:45 <Beerbelott> Isn't rather MakeNewGame that should called?
18:26:03 <Beerbelott> dedicated_v.cpp seems to place calls looking like what's done in SwitchToMode, but in its own context for dedicated server requirements, I get that
18:26:15 <Beerbelott> for SM_LOAD_GAME, that's SafeLoad
18:26:27 <Beerbelott> but for SM_START_HEIGHTMAP, it's MakeNewGame
18:26:43 <Beerbelott> (which ends up calling what u mentioned)
18:27:02 <Beerbelott> isn't it important to mimic the same calls?
18:27:08 <nielsm> ugh what's the word that's similar to "spatial" but relates to time rather than space
18:29:05 <Eddi|zuHause> i have no clue what that means
18:30:04 <glx> Beerbelott: MakeNewGame() creates a new company
18:30:06 <Eddi|zuHause> i shouldn't need to read the code to understand what the comment writer meant :p
18:30:11 *** Progman has joined #openttd
18:30:14 <glx> you don't want that for dedicated
18:30:31 <peter1138> Eddi|zuHause, read the code and write a comment ;)
18:31:38 <glx> hmm, it doesn't, there's a gui check preventing it
18:33:17 <Beerbelott> Well, as it is I was able to spawn a heightmap-base game on a test dedicated server
18:34:21 <Beerbelott> if u wanna fiddle w/ it, u might connect to it
18:35:25 <glx> oh it seems the StartNewGameWithoutGUI() in dedicated is a check call
18:37:07 <Beerbelott> MMMh thus I should replicate this function there for a heightmap version too? lemme check
18:37:19 <Beerbelott> did not see that duplicate
18:38:17 <peter1138> nielsm, I was thinking of a one-liner comment, to be honest.
18:38:19 <Beerbelott> glx: Actually, that explains why, if you 'restart' on my game, you end up in a new game, not a new heightmap game :D
18:38:21 <glx> in dedicated it calls StartNewGameWithoutGUI() then SwitchToMode() which the calls MakeNewGame()
18:38:30 <nielsm> I'll try to be concise
18:38:34 <peter1138> That's verging on Samu-territory ;)
18:38:41 <peter1138> I'll see if I can come up with one.
18:39:53 <nielsm> "9th and 10th bit of tick counter is tile loop count, comparing that to tile index gives a nicer temporal distribution than comparing to a constant."
18:42:02 <Beerbelott> glx: feels wrong *WithoutGUI functions sit in genworld_gui, though
18:43:28 <Eddi|zuHause> Beerbelott: the null video driver is still a video driver...
18:47:20 <glx> and yes, StartNewXXXWithoutGUI() stores the seed :)
18:49:12 <peter1138> nielsm, btw, bits 0-1 of tile may not be the best to use.
18:49:38 <glx> but 'restart' command will still generate a new map as explained in the command help message
18:49:41 <Beerbelott> Is there a way to manually trigger NetworkCheckRestartMap in dedicated_v.cpp?
18:49:41 <peter1138> Hmm, actually not sure it matters, so tile loop is spread out anyway.
18:50:04 <Beerbelott> It seems to be at the end of playing years, but is there a way to fast-forward that call?
18:50:47 <glx> hmm it's a setting I think
18:50:50 <Beerbelott> I'd like to see the states of variabels there...
18:51:06 <Beerbelott> well supplying a custom .cnf then :)
18:51:09 <Beerbelott> I was just lazy :p
18:51:26 <Beerbelott> I see 2 chanegs to be made so far
18:51:32 <Beerbelott> lemme test before doing that
18:52:39 <glx> but NetworkCheckRestartMap() will still generate a newgame because that's what it is supposed to do
18:53:01 <Beerbelott> Yes, but a normal game
18:53:15 <glx> yes, it can't do anything else
18:53:22 <Beerbelott> This is wrong if a heightmap was loaded, it should then trigger a random new map base on current heightmap
18:53:25 <glx> the heighmap is not stored
18:53:37 <glx> it the same if you loaded a game
18:53:40 <Beerbelott> Now u got me sad :(
18:53:58 * TrueBrain tickles Beerbelott till he smiles
18:54:16 <Beerbelott> laughing nervously does not mean someone's happy ;)
18:54:43 <TrueBrain> I am scared I am being called to HR now :P
18:54:53 <Beerbelott> glx: Tell me there is a way outta that situation
18:56:08 <Beerbelott> thus, apart from restarting issues (either manually or through auto-trigger), I do not see any other call to StartNewGameWithoutGUI which needs to be patched
18:56:16 <glx> with -g you can start with a savegame, a scenario or a heightmap (with your code) but on restart it's always a newmap
18:56:27 <Samu> the runway too short warning is not displayed when plane crashes is 0
18:56:47 <Beerbelott> Well, for dedicated server, keeping this parameter somewhere is more than useful
18:57:12 <glx> but not the object of your PR ;)
18:57:19 <Beerbelott> Oh no... Another issue? Another PR?
18:57:48 <Samu> wonder if that proves anything in my favour
18:58:22 <Beerbelott> Well I officially call my PR feature-complete... modulo bugs :\
18:58:43 <TrueBrain> that is what I say in job interviews! :P
18:58:50 <glx> because if you keep the heightmap info, you may need to also keep the savegame/scenario info
18:58:58 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN commented on pull request #6965: Add: Option for population-linear town cargo generation https://git.io/fhxup
18:59:10 <peter1138> nielsm, not perfect but brief.
18:59:16 <glx> and that's a totally unrelated issue
19:00:08 <Beerbelott> glx: Nope, not the object of mine, that's why I referenced new issue/PR
19:00:55 <Beerbelott> "I'm feature-complete": sounds arrogant :D
19:01:08 <TrueBrain> you compensate that with: "modulo bugs"
19:01:09 <Beerbelott> "... module bugs": now creepy
19:01:20 <Beerbelott> yup but bugs has many meanings :\
19:02:23 <Beerbelott> glx: You piqued y interest now, at making this information being remembered
19:02:30 <Beerbelott> No, no, no that's unreasonable
19:02:35 <Beerbelott> I'm *not* a developer
19:02:49 <TrueBrain> keep telling that to yourself
19:03:03 <peter1138> Let me just improve this one thing...
19:03:15 <Samu> do I have to open an issue?
19:03:44 *** octernion has joined #openttd
19:04:00 <peter1138> Samu, what's your point?
19:04:04 <planetmaker> <peter1138> Let me just improve this one thing... <-- sounds... how I started with OpenTTD :)
19:04:05 <TrueBrain> we still have __NDS__ code in our codebase? Owh dear ..
19:04:15 <Beerbelott> I'm normally the guy you blame because you wanna deploy all sorts of exotic programs in very specific versions, and you complain deploying whole new servers farms every morning s not done in a snap
19:04:22 <peter1138> The line you quote says that it won't automatically send large aircraft to small airports for servicing.
19:04:45 <Beerbelott> planetmaker: I did just that too...
19:04:48 <Samu> yes, but the cheat must be disabled
19:04:54 <Beerbelott> and down the road of complexity it went
19:04:59 <peter1138> Yes. And it will be.
19:05:33 <peter1138> Unless you're cheating, in which case it won't crash. So what's the problem?
19:07:10 <peter1138> God damnit, variety distribution is stupid. Who made that rubbish...
19:08:00 <Samu> what about the vehicle message warning?
19:08:18 *** m3henry has joined #openttd
19:08:32 <DorpsGek> TrueBrain: Terkhen was last seen in #openttd 3 years, 3 weeks, 6 days, 23 hours, 31 minutes, and 35 seconds ago: <Terkhen> Hello
19:08:52 <peter1138> Samu, okay, that line probably needs to be taken out, correct.
19:10:30 <DorpsGek> TrueBrain: tokai|noir was last seen in #openttd 44 weeks, 2 days, 6 hours, 45 minutes, and 4 seconds ago: <tokai|noir> TrueBrain: Oh... but I am lurking. :)
19:10:33 <peter1138> Well, that setting does not affect crashing at too-short runways.
19:10:52 <TrueBrain> tokai|noir: you still lurking? what are the odds that you can help out with MorphOS / AmigaOS? :D
19:13:02 <peter1138> Samu, what's the problem? You've highlighted an issue, which will get resolved. Thanks!
19:13:31 <Samu> why was it like that in the first place
19:14:17 <peter1138> Well until #7302 it was 'correct'
19:14:41 <m3henry> peter: would removing derived virtual destructors (to default them) make more sense as a separate commit?
19:15:18 <peter1138> Maybe. They're all pointless, as far as I can see.
19:15:58 <Samu> also found something in the wiki
19:16:54 <Heiki> trying to translate “Invalid industry production callback” into Finnish, I think I need some wine now
19:19:16 <peter1138> Samu, mentions nothing about too short runways.
19:21:28 <peter1138> wiki isn't canonical documentation anyway.
19:22:21 <peter1138> Just stop. Just don't send large aircraft to small airports.
19:22:32 <tokai|noir> TrueBrain: Pretty slim, sadly. :) I don't even have a compatible machine anymore.
19:22:43 <peter1138> If you like, you can get around it by making a NewGRF that makes all aircraft small.
19:22:53 <TrueBrain> well, I guess it is time to remove MorphOS from our codebase in that case :(
19:22:56 <TrueBrain> if you don't mind at least?
19:23:29 <nnyby> i wish i still had my amiga 2000.. amigas are so hard to find. any way to get some compatible alternative hardware these days?
19:23:52 <tokai|noir> TrueBrain: If it blocks other things, and no other dev can step in, then remove the code, sure.
19:24:00 <Samu> just wanted to point out I wasn't crazy when I said there were ppl who'd think none really meant no plane would ever crash
19:24:18 <TrueBrain> tokai|noir: yeah, we want to switch to SDL2. There is a WIP-ish SDL2 port for MorphOS, but not supported by upstream
19:24:25 <TrueBrain> so ... we have no way to knowing it works or not :D
19:24:34 <peter1138> People just documenting the behaviour they saw, despite it being wrong.
19:25:48 <planetmaker> does mean "switch to sdl2" mean that we remove sdl1.x support? Couldn't it "just" be yet another video driver?
19:25:56 <tokai|noir> TrueBrain: it's not like the old versions of OpenTTD will stop working (at least for the single player modi) :)
19:26:07 <peter1138> planetmaker, it could, if you fancied a lot of duplication.
19:26:08 <TrueBrain> planetmaker: I asked the same; I got answer that it was decided SDL1 was being removed :)
19:26:16 <TrueBrain> tokai|noir: indeed :)
19:26:24 <TrueBrain> planetmaker: but nobody knows for sure MorphOS is currently still working
19:26:30 <peter1138> The driver is mostly the same, just a few tweaks.
19:26:32 <TrueBrain> last fix for MorphOS is 2008 ..
19:26:39 <TrueBrain> so possibly it is also time to say goodbye to it :)
19:26:43 <peter1138> And some backwards compatibility cruft gets removed.
19:26:48 <TrueBrain> (no offense or anything meant :))
19:26:59 <peter1138> There's always the Allegro driver ;)
19:27:16 <peter1138> But I really see no point in maintaining both SDL 1.2 and SDL 2.0
19:27:19 <TrueBrain> I will also make a PR to remove BeOS, as I am being told Haiku replaced it
19:27:29 <TrueBrain> especially if it is only for OSes nobody is testing
19:27:35 <planetmaker> if it's mostly the same... a factory driver with a switch to v1.2 or 2.0 should work like OSX with cocoa and quickdraw?
19:28:00 * tokai|noir recently started to play around with Haiku ;)
19:28:22 <planetmaker> congrats. You just got a new job here :P
19:29:55 <tokai|noir> Recently as in like yesterday... had the strange mood to test something new. :)
19:30:28 <tokai|noir> TrueBrain: Well.. new to me. :D
19:30:37 <peter1138> planetmaker, not sure what you mean.
19:30:52 <TrueBrain> how did AmigaOS and MorphOS differ?
19:30:54 <TrueBrain> I always wondered ..
19:31:58 <tokai|noir> TrueBrain: not a lot. MorphOS was basically a AmigaOS clone with some old hardware-depending cruft removed and some new API stuff added. A bit shinier, but still AmigaOS with all its flaws. :)
19:33:58 <nnyby> there is also AROS btw, seems have similar goals of an improved AmigaOS
19:34:47 <TrueBrain> hmm .. s ome traces of NDS support
19:35:08 <glx> we used to support nintendo DS ?
19:35:23 <TrueBrain> we did? How is it playable?
19:35:31 <TrueBrain> ugh, I remember I wrote a PSP port .. I will shut up
19:35:47 <peter1138> The PSP port is why we have the libtimidy driver.
19:36:04 <TrueBrain> I will remove it in a bit :P
19:36:11 <planetmaker> peter1138, from my OSX time I recall that OSX has a a lot of abstract code to support the display driver. And this base class is sub-classed by the actual cocoa and quartz (not quickdraw) display drivers. IIRC. Let's check
19:36:34 <peter1138> TrueBrain, I think the android port uses libtimidity. If we care.
19:36:47 <TrueBrain> the port is free to care
19:36:50 <peter1138> Although, I would like to have official android support. Just haven't got there yet.
19:36:55 <TrueBrain> having untested code in your repository is bad
19:37:09 <TrueBrain> so I remove it for now in a single commit
19:37:16 <TrueBrain> reverting of that commit gives you back support
19:37:25 <peter1138> I suspect the fluidsynth stuff would work just as well, tbh.
19:37:53 <peter1138> TrueBrain, well it's git ;) not like my colleagues idea of version control... "made a backup of it"
19:37:55 <planetmaker> ah cocoa -> (quartz|quickdraw)
19:38:02 <TrueBrain> peter1138: exactly ;)
19:38:48 <planetmaker> e.g. cocoa_v.h:195-203 etc
19:39:33 <peter1138> planetmaker, okay but you get to maintain both pieces of sdl :p
19:39:33 <tokai|noir> TrueBrain: looks like libsdl2 is available for Haiku at least. If this thing is in some form posix compatible a compile of OpenTTD is probably painless, I guess.
19:39:50 <TrueBrain> tokai|noir: that was the hope and idea :)
19:40:05 <glx> tokai|noir: does it have cmake ?
19:40:33 <planetmaker> peter1138, yes... where they differ
19:40:41 <tokai|noir> glx: it's available in their repository.
19:42:29 <tokai|noir> hmm... OpenTTD code now on github? :) Lots of things changed... I still remember the pain I had to port over a broken SVN build to MorphOS to be able to check out OpenTTD's code ;)
19:42:32 <glx> so if you want to play we have PR#7270 ;)
19:42:48 <planetmaker> but cocoa_v.mm is litterally sprinkled everywhere with like #if defined(ENABLE_COCOA_QUARTZ) || defined(ENABLE_COCOA_QUICKDRAW) and alike... not nice to maintain
19:44:19 <peter1138> planetmaker, exactly
19:44:26 <peter1138> So why would you want to do that?
19:45:01 <TrueBrain> OpenBSD and OS/2 .. when did we last test those? :D
19:46:55 * tokai|noir doesn't know a thing about Haiku, doesn't remember a thing about how to build OpenTTD... wonder where that goes. ;)
19:47:15 <nnyby> hah - looks fun. maybe i'll give that a try too this evening in a haiku VM.
19:47:28 <Beerbelott> glx: I don't like signatures not matching for similar functions
19:47:34 <LordAro> TrueBrain: there was a openbsd issue opened not that long ago...
19:48:00 <LordAro> you'd have to ask orudge :p
19:50:42 <glx> hey but if master fails don't try to fix it, better test in #7270 and suggest fixes there I think
19:51:01 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh updated pull request #6965: Add: Option for population-linear town cargo generation https://git.io/fpkqa
19:51:02 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN opened pull request #7325: Fix: Runway too short for large aircraft advice should not depend on plane crashes setting. https://git.io/fhxz7
19:51:12 <TrueBrain> tokai|noir: no matter what, I would love to see what #7270 does on Haiku; just to get a sense for it :)
19:51:26 <tokai|noir> "detecting OS... HAIKU" ... "configure: error: no video driver development files found" so far
19:51:32 <tokai|noir> time to install the sdl2 files
19:52:04 <peter1138> It's worth testing it.
19:52:09 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh approved pull request #7325: Fix: Runway too short for large aircraft advice should not depend on plane crashes setting. https://git.io/fhxzN
19:52:24 <tokai|noir> glx: sdl is actually installed already it seems
19:52:46 <LordAro> tokai|noir: detection requires pkg-config, if that's not installed
19:53:15 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain opened pull request #7326: Remove MorphOS / AmigaOS / BeOS support and libtimidity support https://git.io/fhxzA
19:53:28 <Beerbelott> I do not get PR #7325
19:53:33 <tokai|noir> LordAro: should be available too.
19:53:42 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain commented on pull request #7086: Change #6173: Update SDL driver to use SDL 2.0 https://git.io/fhxzx
19:54:03 <Beerbelott> In #7302, I see crash chance being 3276 for too small runways
19:54:06 <LordAro> tokai|noir: well you're going to have to dig a little deeper then, i'm afraid :>
19:54:47 <tokai|noir> LordAro: right now I'm standing on this 1000 square kilometre large area and don't know where to start digging :D
19:54:53 <peter1138> Beerbelott, that crash chance does *not* depend on the plane crashes setting.
19:54:58 <peter1138> Beerbelott, that's the point :)
19:55:17 *** frosch123 has joined #openttd
19:55:40 <LordAro> tokai|noir: `sh -x ./configure` might give some information
19:55:51 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN approved pull request #6965: Add: Option for population-linear town cargo generation https://git.io/fhxgv
19:55:51 <Beerbelott> But it's still *there*, which is a pb #troll
19:56:04 <Beerbelott> (no, not starting the discussion :p )
19:56:35 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain updated pull request #7326: Remove MorphOS / AmigaOS / BeOS support and libtimidity support https://git.io/fhxzA
19:57:06 <Beerbelott> TrueBrain: No AMigaOS support anymore? That's harsh for its millions of users
19:57:22 <peter1138> LordAro, also, you have a MAJOR problem.
19:57:29 <Beerbelott> OK I stop trolling.
19:57:36 <tokai|noir> LordAro: well.. pkg-config sdl doesn't report a thing.. that's probably why. :)
19:57:47 <peter1138> LordAro, why do you not have n+1 bikes?
19:57:59 <peter1138> Everyone needs more than 1 bike :D
19:58:20 <LordAro> TrueBrain: well that's just blatently wrong
19:58:44 <LordAro> peter1138: i have been looking at bikes recently...
19:58:49 <peter1138> I think 4 isn't enough.
19:58:58 <LordAro> have had my Whyte for 18 months now
19:59:06 <peter1138> road bike, commuter bike, touring bike, hardtail mountain bike, full-sus...
19:59:13 <peter1138> folding bike for trains...
19:59:25 <peter1138> recumbent trike for more fun.
19:59:29 <peter1138> I need a bigger garage.
19:59:46 <LordAro> peter1138: i need a garage
19:59:49 <peter1138> Nobody needs a fat bike though, they're just stupid.
19:59:57 <LordAro> that's the other reason why i haven't got another bike
20:00:03 <tokai|noir> LordAro: looks like their package repository only installs the binaries (.so files), but no includes or whatsoever.
20:00:13 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN merged pull request #7325: Fix: Runway too short for large aircraft advice should not depend on plane crashes setting. https://git.io/fhxz7
20:00:27 <LordAro> tokai|noir: ah, the debian way? is there an sdl-dev or equivalent package?
20:00:47 <tokai|noir> LordAro: just found it.. had to activate some filter settings in the repository app.
20:00:54 <LordAro> peter1138: they look like fun in a foot of snow, not much else though
20:01:28 <peter1138> Now... THAT looks like fun.
20:02:16 <peter1138> Nowhere near enough snow around here (i.e. the UK) for that.
20:02:31 <peter1138> TrueBrain, it is :)
20:02:51 <tokai|noir> Configuring was successful, now to the build (no lzo2 though)
20:03:06 <peter1138> It's like a product image photoshopped onto a snowy background.
20:03:13 <TrueBrain> promising tokai|noir :)
20:03:17 <peter1138> The one in the background is fine.
20:03:55 <tokai|noir> TrueBrain: some byte order define issues so far (__LITTLE_ENDIAN not defined)
20:04:55 <TrueBrain> the docker crosscompiling is saying the same thing :)
20:05:15 <tokai|noir> since __BYTE_ORDER is not defined too it will evaluate 0 == 0... I guess that should work out :)
20:06:04 <peter1138> What OSes use Allegro but not SDL?
20:06:08 <LordAro> sounds like endian_check is not doing its thing?
20:06:38 <peter1138> We don't build that either.
20:07:58 <Beerbelott> How are IDs in tar archives decided?
20:08:00 <peter1138> Ah, languages, hehe
20:08:09 <TrueBrain> yeah ... had to fix the language files :)
20:09:46 <TrueBrain> core/os_abstraction.h:108:0: warning: "IPV6_V6ONLY" redefined
20:09:55 <tokai|noir> still building... looking good so far :)
20:10:07 <TrueBrain> console_cmds.cpp:1566:18: warning: format '%lld' expects argument of type 'long long int', but argument 7 has type 'int64 {aka long int}'
20:10:11 <tokai|noir> seems it has linked successful too.
20:10:14 <TrueBrain> seems some type is not what we expect
20:10:44 <tokai|noir> wasn't there some makefile command to launch openttd directly from the build directory?
20:11:15 <tokai|noir> failed to find graphics set. no crash :)
20:11:37 <peter1138> Hmm, it should let you download a set?
20:11:48 <peter1138> Unless it's a dedicated build.
20:11:48 <TrueBrain> yeah, so people keep saying ... :P
20:12:18 <tokai|noir> I don't remember how this works... usually I copied the original files (TTD) from somewhere :)
20:12:29 <TrueBrain> bin/openttd: ELF 64-bit LSB shared object, x86-64, version 1 (SYSV), dynamically linked, not stripped
20:12:33 <TrueBrain> so Haiku uses ELF ..
20:12:58 <TrueBrain> even the MIDI driver still compiles :o
20:13:21 <tokai|noir> TrueBrain: certainly looks to be in a pretty good state
20:13:42 <TrueBrain> if I can figure out how to get the SDL package
20:13:45 <TrueBrain> I can give you a binary you can try?
20:13:50 <TrueBrain> as if that works, we can even cross-compile :P
20:14:07 *** gelignite has joined #openttd
20:17:24 <Beerbelott> glx: In the case of a wrong heightmap file, it seems the game already checks for load file fail... but only triggers exit after having generated a new random world
20:17:38 <Beerbelott> THus I guess it's all a matter of turning things 'round?
20:18:55 <TrueBrain> tokai|noir: you using 64bit Haiku?
20:19:09 <tokai|noir> TrueBrain: Yep, I think so.
20:19:47 <TrueBrain> how was the sdl dev package called? :D
20:19:48 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh merged pull request #6965: Add: Option for population-linear town cargo generation https://git.io/fpkqa
20:19:55 *** octernion has joined #openttd
20:20:19 <tokai|noir> TrueBrain: sdl-devel or something, let me check (need to get some IRC going on that Haiku, so I don't have to switch constantly ;) )
20:21:04 <tokai|noir> TrueBrain: libsdl_devel
20:21:21 <TrueBrain> hmm, I need to manually download that .. hmm
20:23:36 <TrueBrain> okay, the pkgconfig files are installed, but pkg-config doesn't see them, because they are in a weird folder
20:24:11 <tokai|noir> the hardest part was figuring out where the Haiku webbrowser downloaded that gfx files to ;)
20:24:24 <TrueBrain> nice tokai|noir, really nice :)
20:24:33 <nielsm> were anyone on the task of updating opengfx? :)
20:25:00 <TrueBrain> andy was, but we scared him away
20:25:02 <TrueBrain> planetmaker hopefully?
20:26:23 <tokai|noir> TrueBrain: pretty painless. We just need to fix those missing endian defines (by luck both of them missing will result in the correct code build anyway :) ) and this IPv6 define too.
20:26:38 <TrueBrain> tokai|noir: yeah, I have most of those already locally
20:27:21 <TrueBrain> tokai|noir: now I am trying this cross compiler .. if that works, we can add it to the CI :D
20:27:26 <TrueBrain> or possibly only to releases :P
20:27:51 <nielsm> speaking of ancient OSes, do we know if DOS versions can still be produced?
20:28:02 <TrueBrain> is there a Docker to do that? :P
20:29:13 <nielsm> have anyone written an ssh server for dos?
20:29:14 <Beerbelott> glx: Guess what? That's what the game does for standard savgeame anyway... It generates a new game before closing. I guess the closing only happens because of SM_NONE anyway
20:29:25 <TrueBrain> hmm .. pkg-config is not going to help me, as the information is plain wrong .. boooeeee
20:34:02 <TrueBrain> okay, hacked around that .. let me create a binary now ..
20:36:39 <tokai|noir> TrueBrain: the check online stuff doesn't seem to work... perhaps that IPv8 define thingy broke something, else it seems to work well so far.
20:36:42 <TrueBrain> but the README is annoying :P
20:37:09 <TrueBrain> tokai|noir: I found another odd entry that might break the online stuff ..
20:37:22 <TrueBrain> bah, linking errors
20:39:26 <TrueBrain> lol .... 'libsdl_devel' doesn't contain the so ..
20:40:58 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh commented on pull request #7250: K-d tree data structure for spatial lookups https://git.io/fhx2k
20:41:13 <TrueBrain> undefined reference to `_Unwind_Resume'
20:41:51 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh commented on pull request #7250: K-d tree data structure for spatial lookups https://git.io/fhx2I
20:42:30 <TrueBrain> tokai|noir: src/network/core/os_abstraction.h, around line 66, remove __HAIKU__ there; that should fix the online stuff
20:42:46 <TrueBrain> (well, the "!defined(__HAIKU__)")
20:43:03 <TrueBrain> and around line 87, remove the whole ifdef around IPv6
20:43:57 <tokai|noir> TrueBrain: ok, give me a minute.. need to find some sort of editor first :D
20:44:38 <TrueBrain> Xaroth: cat is not an editor :(
20:44:49 <TrueBrain> such attempt; such failure
20:45:13 <DorpsGek_II> [OpenTTD/OpenTTD] M3Henry updated pull request #7324: Codechange: [C++11] Use override specifer for overriding member declarations https://git.io/fhxq0
20:45:15 <TrueBrain> what is this linker issue with _Unwind .. hmm
20:45:25 <tokai|noir> Nah... there's this "StyledEdit", lets try that... it's a GUI based OS after all :)
20:45:53 <Xaroth> cat can be used to edit.
20:46:11 <TrueBrain> do you alias cat to vim?
20:46:34 <TrueBrain> that is not EDITING weirdo
20:46:39 <nielsm> you could maybe use a combo of cat and dd to edit
20:46:58 <TrueBrain> so yeah, the official answer for tokai|noir is: dd :)
20:47:39 <TrueBrain> undefined reference to `_Unwind_Resume' <- anyone any idea?
20:48:09 <LordAro> missing exception support?
20:48:20 <tokai|noir> TrueBrain: I wonder if Haiku defines the BEOS stuff too (similar like MorphOS would also define the AMIGAOS defines) or if the Haike developers did a clean break
20:48:46 <TrueBrain> tokai|noir: I am running on my patch that removes BeOS :P
20:49:18 <LordAro> i've definitely seen it before though...
20:49:20 <tokai|noir> TrueBrain: but maybe my version only works because it included the BeOS stuff? :)
20:49:33 <TrueBrain> ah, no .. seems the crosscompiler is not perfect
20:50:13 <LordAro> TrueBrain: interesting
20:50:57 <TrueBrain> LordAro: if I add -lgcc_s, it works fine
20:51:05 <tokai|noir> TrueBrain: right... trying to figure out your line numbers (they seem to be off quite a bit compared to what I have here ;)
20:51:15 <TrueBrain> tokai|noir: oops, that is possible
20:51:18 <LordAro> TrueBrain: weird that the compiler doesn't add that by default
20:51:28 <LordAro> it usually does for libgccish things
20:52:44 <LordAro> Samu: at least, it needs rereviewing
20:52:54 <DorpsGek_II> [OpenTTD/OpenTTD] SamuXarick commented on pull request #7193: Fix #6468: Load correct version of AI as specified during the time of its save. https://git.io/fhx2l
20:53:18 <tokai|noir> TrueBrain: ok, did it correctly already. :) Building right now.
20:53:38 <TrueBrain> tokai|noir: and while building, you can test my binary :D
20:54:08 <tokai|noir> TrueBrain: I'm not so quick here with siwthcing constantly... and still trying to figure out things on Haiku as we go :D
20:55:28 <TrueBrain> seems int64 is not a long long int
20:56:13 <LordAro> that seems undesirable
20:56:15 <Samu> isn't there a way to filter which PRs currently have conflicts? :(
20:56:36 <LordAro> TrueBrain: i've thought that int64 needs properly aliasing to int64_t & friends for a while
20:57:01 <TrueBrain> tokai|noir: booooeeee
20:57:12 <Beerbelott> What does MarkWholeScreenDirty do?
20:57:15 <TrueBrain> LordAro: int64 should be 8 bytes, right?
20:57:26 <LordAro> TrueBrain: i would certainly hope so
20:57:36 <TrueBrain> just seems that it is a "long int"
20:57:48 <TrueBrain> wait .. we have code for that
20:58:23 <LordAro> since "long" is defined as "at least as big as int", and "int" is defined as "at least 16 bits" in the standard, iirc
20:58:35 <planetmaker> Beerbelott, forces redraw of everything
20:58:54 <TrueBrain> we have defines for printf
20:59:11 <Samu> I remember you had worked on a DoCommand patch for it
21:01:40 <TrueBrain> okay, this is tricky .. SQIntegers are long long int ..
21:02:40 <LordAro> i think MinGW had a similar issue?
21:02:49 <peter1138> Beerbelott, it ... marks the whole screen dirty, so it will be redrawn.
21:03:32 <TrueBrain> HAIKU already had an ugly hack for int32
21:03:36 <TrueBrain> it now also has it for int64 :P
21:03:42 <glx> mingw has weird printf warnings
21:03:56 <peter1138> "what's a 64 bit integer!?"
21:05:45 <TrueBrain> euhm ... threaD_pthread.cpp, line 82
21:05:59 <TrueBrain> static void *, no return
21:06:01 <tokai|noir> hmm... is there a trick to join this channel? Always sends my haiku user to ##openttd because of a missing invite :)
21:06:04 <frosch123> TrueBrain: c++11 adds <cstdint>, which has int64_t and uint64_t
21:06:12 <frosch123> so we should be able to remove all the custom defines
21:06:35 <LordAro> tokai|noir: shouldn't be, it's not invite only
21:06:57 <TrueBrain> okay, that fixes all the warnings
21:07:10 <LordAro> frosch123: that's older than c++11
21:07:44 <TrueBrain> some warnings I think we can ignore
21:08:22 <tokai|noir> LordAro: :verne.freenode.net 470 tokai-haiku #openttd ##openttd :Forwarding to another channel
21:08:49 <glx> yeah we are not on freenode
21:08:49 <tokai|noir> LordAro: argh... I see... :D
21:08:59 <TrueBrain> and #openttd there is still locked :P
21:11:52 <TrueBrain> okay, so I guess I can make haiku in a Docker, so we can produce binaries for it :P Not because it will have many downloads, but because it is funny :D
21:13:36 *** tokai-haiku has joined #openttd
21:15:50 *** octernion has joined #openttd
21:15:52 <glx> 60fps ? how is it possible ?
21:16:10 <TrueBrain> what did you break .....
21:16:54 <nielsm> 320 fps fully zoomed in
21:17:06 <LordAro> nielsm: did you turn full animation off?
21:17:15 <tokai-haiku> TrueBrain: seems to work. But no zlib, so it can't load the title screen game nor do the network stuff (it tells me) :)
21:17:31 <TrueBrain> yeah, I was lazy :P
21:17:38 <TrueBrain> but that it starts, is .. wauw
21:17:45 <nielsm> LordAro: default blitter
21:17:54 <TrueBrain> I will work out this Docker builder a bit more, and we will see where it takes us
21:18:10 <TrueBrain> tokai-haiku: if you happen to have time on your hands, if you can find out why the network acts up, would be awesome :D
21:18:27 <tokai-haiku> TrueBrain: also getting the "error running _netstat" ... so there's definitelly some more fixes to do here and there, but certainly a good start, IMHO :)
21:19:17 <TrueBrain> we just need someone that stays on it :P
21:19:31 *** Wormnest has joined #openttd
21:19:37 <TrueBrain> right; tomorrow we try DJGPP (DOS)
21:19:50 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh updated pull request #7250: K-d tree data structure for spatial lookups https://git.io/fhd4b
21:20:01 <tokai-haiku> OpenTTD really should run on DOS; IMHO ;)
21:20:18 <tokai-haiku> I mean the original TTD was a DOS game originally, AFAIR, no? :)
21:20:40 <TrueBrain> Platform(s) MS-DOS, Mac, PlayStation, Sega Saturn, Android, iOS
21:21:01 <LordAro> the latter 2 are a completely different game
21:21:09 <tokai-haiku> I remember in the early 90s when I was visiting friends who could play the game on their parents PCs... I always wanted to play it too (we never got a port for Amiga back then ;)
21:21:10 <LordAro> well, same graphics, i guess, but
21:21:20 <nielsm> yeah TT was written in 386 assembly
21:21:43 <TrueBrain> but but but .. it worked for RCT
21:21:51 * tokai-haiku played A-Train instead (not a bad game either for its time, IMHO ;)
21:25:58 <Beerbelott> I got a dilemma: should I rather make StartNewHeightMapGameWithoutGUI have a different signature than StartNewGameWithoutGUI, or include heightmap.h into video/dedicated_v.cpp to have access to functions in order to test file content validity?
21:26:34 <Beerbelott> the StartNew*GameWithoutGUI functions are in genworld_gui.cpp
21:26:56 <glx> I don't see issue in signature change :)
21:33:13 <nielsm> (never mind it took several minutes to load the save)
21:37:01 <tokai-haiku> TrueBrain: seems it doesn't define the old __BEOS__ define at least. Should make it easier to compare and keep things clean.
21:42:45 <Heiki> when and where in the actual game is “STR_CONFIG_SETTING_TOWN_CARGOGENMODE_ORIGINAL” used?
21:42:47 <Beerbelott> glx: is it that bad to get access to GetHeightmapDimensions by including heightmap.h into video/dedicated_v.cpp?
21:44:23 <peter1138> Heiki, in the ... game settings window.
21:44:35 <peter1138> Heiki, like all STR_CONFIG_SETTINGS :-)
21:44:41 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh commented on pull request #7250: K-d tree data structure for spatial lookups https://git.io/fhxaR
21:45:42 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN updated pull request #7234: Feature: Game setting to define how industries with neutral stations accept and supply cargo from/to surrounding stations. https://git.io/fh5qg
21:48:05 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN commented on pull request #7250: K-d tree data structure for spatial lookups https://git.io/fhxag
21:49:08 <nielsm> peter1138, any opinion on the other part of that discussion point? :)
21:50:27 <tokai-haiku> TrueBrain: checked the diff of #7261... darn... those AmigaOS/MorphOS changes were really all over the place, huh? Still a bit sad to see some of the code one wrote go to >NIL: (or translation for everyone else: > /dev/null) :D
21:51:54 <DorpsGek_II> [OpenTTD/OpenTTD] M3Henry updated pull request #7165: [core] Implement SmallVector using std::vector https://git.io/fhSz0
21:52:35 <peter1138> nielsm, if it works as is with the asserts, then using 2 x min and 2 x max is wasted cycles ;)
21:53:10 <Heiki> peter1138: never mind, it was just the Web Translator giving me a string newer than the game I’m currently running
21:53:17 <m3henry> I don't think there's anything more to be done that wouldn't be best served in a separate PR now
21:53:51 <nielsm> peter1138 yeah my argument is somewhat that you should know what you're doing if you want to use the tree :P
21:54:51 <peter1138> I'm happy with that.
21:56:23 <peter1138> m3henry, it fails on the commit checker though.
21:56:30 <peter1138> "Replaced SmallVector::Include() with include() "
21:56:34 <m3henry> I ssume it was borked
21:56:43 <m3henry> Can't see why it's failing
21:58:19 <DorpsGek_II> [OpenTTD/OpenTTD] M3Henry updated pull request #7165: [core] Implement SmallVector using std::vector https://git.io/fhSz0
22:02:01 <LordAro> probably want to prioritise reviewing m3henry's PRs, given the amount of changes in them
22:02:12 <LordAro> and how every other PR is going to need rebasing after they're merged
22:02:26 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh updated pull request #7250: K-d tree data structure for spatial lookups https://git.io/fhd4b
22:04:02 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh updated pull request #7250: K-d tree data structure for spatial lookups https://git.io/fhd4b
22:05:44 <m3henry> note: ‘std::make_unique’ is defined in header ‘<memory>’; did you forget to ‘#include <memory>’?
22:06:09 <m3henry> GCC, you are drunk, go home
22:06:42 <LordAro> i'd suggest that C++14 isn't unreasonable
22:06:52 <LordAro> especially given it's basically "fixes for C++11"
22:07:06 <peter1138> How do we cope with C++11 parts that are included by objc code?
22:07:21 <m3henry> there's obj-C in this?
22:07:29 <peter1138> Mac OS X video drivers.
22:07:47 <peter1138> The virtual -> override PR throws up warnings on OS X.
22:08:27 <m3henry> sounds like the way google does it
22:08:36 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN commented on pull request #7165: [core] Implement SmallVector using std::vector https://git.io/fhxVf
22:10:13 <peter1138> Yeah, it's probably fine doing that. override is only a compiler hint and doesn't affect output. At least, it shouldn't :p
22:10:18 <DorpsGek_II> [OpenTTD/OpenTTD] M3Henry commented on pull request #7165: [core] Implement SmallVector using std::vector https://git.io/fhxVJ
22:11:19 <peter1138> Hmm, failed on MacOS, Win32 and Win64 :(
22:11:23 <glx> Beerbelott: it's probably an option, but indeed using StartNewHeightMapGameWithoutGUI() means generating twice and, you're right, it's not good :)
22:12:31 <m3henry> I swear I fixed that yesterday...
22:13:13 <glx> oh the usual os specific files ;)
22:13:50 <m3henry> "uniscribe" -> universal strings ?
22:14:18 <nielsm> microsoft unicode text shaping library
22:14:28 <nielsm> replacement for ICU on windows
22:16:36 <DorpsGek_II> [OpenTTD/OpenTTD] M3Henry updated pull request #7165: [core] Implement SmallVector using std::vector https://git.io/fhSz0
22:23:55 <Samu> I remember you had worked on a DoCommand patch for it, what happened to it?
22:27:33 <Samu> well, I think my PR is currently... pointless to remain open
22:28:08 <Samu> emergency save fix was implemented, and the other crash it attempts to fix, is wrong
22:28:20 <peter1138> Oh yes, I was working on a solution to fix that.
22:28:21 <Samu> and then there's your approach
22:28:51 <peter1138> Although it was overkill, as it modified network protocol, but... I don't think an AI/GS commands go through the network.
22:29:23 <glx> their commands are sent to clients
22:29:55 <peter1138> Yes but clients won't be executing the callback.
22:30:30 <Samu> it was a crash on server side
22:30:57 <Samu> clients dont run ai instances
22:31:16 <glx> hmm no, the callback is only for the server, to tell the script the command has been executed
22:36:23 <peter1138> Nothing for the moment, I think.
22:36:55 <Beerbelott> glx: Just tested ur solution and mine
22:37:13 <glx> yours should be faster ;)
22:37:34 <Beerbelott> It generates the map twice indeed in your + seems ugly to go through a 'proxy' function not made for a check to do one
22:37:57 <Beerbelott> we could always make a second argument to it 'bool check' but that's sliding on the pathof oddities
22:38:30 <glx> can be an optional arg defaulting to false
22:38:37 <Beerbelott> just checked mine and yeah it's faster since no extra loop (not that I know precisely whet happens, but I notice the shorter + faster output)
22:39:00 <Beerbelott> calling directly a function from heightmap.cpp is cleaner/faster and looks prettier
22:39:20 <glx> do you set _settings_newgame.game_creation.generation_seed ?
22:39:23 <Beerbelott> I do not see why we should push to much the other way, just to reuse a function not intiially made for checks
22:40:44 <Beerbelott> I'll push my version
22:41:04 <Beerbelott> Feel free to grab a copy of the PR changes and swap functions around, u'll see
22:50:48 <DorpsGek_II> [OpenTTD/OpenTTD] SamuXarick updated pull request #7184: Change: Distribute cargo to multiple stations or industries https://git.io/fh9lr
23:03:26 *** octernion has joined #openttd
23:07:09 <DorpsGek_II> [OpenTTD/OpenTTD] SamuXarick updated pull request #7190: Fix #7188: AI instance crash when reloading AI in a server, and an AI… https://git.io/fh9jW
23:13:46 <DorpsGek_II> [OpenTTD/OpenTTD] SamuXarick opened pull request #7327: Cleanup: Remove repeated network error message on connection timeout. https://git.io/fhxwc
23:14:29 <Heiki> 20190304T201654+0200 Heiki> trying to translate “Invalid industry production callback” into Finnish, I think I need some wine now
23:15:08 <Heiki> apparently one litre of wine was more than enough for about 100 strings
23:15:59 <nielsm> and yea that's an example of why I think eints could really use a translation memory/"dictionary" feature
23:16:11 <nielsm> to keep track of what you do about specialty words like callback
23:16:34 <nielsm> translation memory is the term used in subtitling systems
23:17:04 <glx> sometimes the suggested similar strings are good enough, but I often need to search for how I translated some terms before :)
23:17:56 <Eddi|zuHause> well, it depends a bit on whether you want an automatted solution or a manually curated one
23:36:56 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN commented on pull request #7301: Feature: When filtering purchase list by cargo type, make buy button perform a refit if required. https://git.io/fhxwp
continue to next day ⏵
