IRC logs for #openttd on OFTC at 2013-08-07
⏴ go to previous day
00:25:38 *** roboboy has joined #openttd
01:18:55 *** Aristide has joined #openttd
01:32:16 *** amiller has joined #openttd
01:55:36 *** fjb is now known as Guest2553
02:06:34 *** montalvo has joined #openttd
03:09:46 <montalvo> does anyone have tips for going back to a really big built up city and somehow sneaking in more train stations around it?
03:10:08 <montalvo> i'm quite new to the game and didn't think ahead to put down loads of train stations around the main city i've focused on, and now i'm finding it hard to expand around it
04:56:18 *** Eddi|zuHause has joined #openttd
05:27:18 *** Supercheese has joined #openttd
05:52:39 *** tokai|noir has joined #openttd
05:52:39 *** ChanServ sets mode: +v tokai|noir
06:25:21 *** Twofish has joined #openttd
06:32:30 *** MatrixCL has joined #openttd
06:47:51 *** sla_ro|master has joined #openttd
07:07:13 *** Alberth has joined #openttd
07:07:13 *** ChanServ sets mode: +o Alberth
07:27:46 *** roboboy has joined #openttd
07:32:58 *** oskari89 has joined #openttd
08:01:23 *** DorpsGek has joined #openttd
08:01:23 *** ChanServ sets mode: +o DorpsGek
08:11:05 *** oskari892 has joined #openttd
08:48:19 *** TheMask96 has joined #openttd
09:01:17 *** Ristovski has joined #openttd
09:09:44 *** insulfrog has joined #openttd
09:10:41 *** oskari89 has joined #openttd
09:19:09 *** Midnightmyth has joined #openttd
09:43:47 <Eddi|zuHause> i know The Ultimate Challenge for openttdcoop -> play future games without "firstred_twoway_eol"
09:45:11 <Eddi|zuHause> and i postulate that (almost) all the constructions are still possible, with some slight modifications
09:55:19 *** LordAro has joined #openttd
09:56:06 <Alberth> I have a different keyboard, still getting used to it
09:56:30 <LordAro> so you missed the tab key?
09:57:13 <Alberth> auto completion failed, as we also have lobsters here
09:57:49 <LordAro> when i type "lo" it autocompletes to lobsters at this end
09:58:00 <LordAro> i'd say that's your client, rather than your keyboard
10:00:32 <Eddi|zuHause> my client by default starts with the last nick that talked
10:01:32 <Alberth> that also fails sometimes, when you have several people discussing :)
10:03:14 <LordAro> ah, i can also go with 'last spoke' also
10:04:32 <LordAro> phew, that was scary - my fglrx module build failed when i updated the kernel, luckily triggering it manually seems to have done the trick
10:07:56 <V453000> what is the point that Eddi :)
10:08:07 <V453000> ofc most things are possible, just not as reliably
10:10:48 *** oskari892 has joined #openttd
10:28:35 <Eddi|zuHause> V453000: the point is to reforge the way you are thinking, when suddenly the rules change :p
10:29:01 <V453000> I dont see how counting penalty values reforges anything
10:31:50 <Eddi|zuHause> whatever... it's not like enlarging your horizon ever helped anybody...
10:32:44 <V453000> wtf do you mean by that
10:36:04 *** sla_ro|master has joined #openttd
10:38:33 <V453000> that isnt much Eddi :)
10:55:34 *** sla_ro|master has joined #openttd
11:00:50 *** insulfrog has left #openttd
11:10:56 *** oskari89 has joined #openttd
11:15:10 *** sla_ro|master has joined #openttd
11:19:16 *** Aristide has joined #openttd
12:11:33 *** oskari892 has joined #openttd
12:13:42 *** sla_ro|master has joined #openttd
12:15:23 <roboboy> does anyone know if it's possible using php to include the servers directory listing into another page without using http?
12:18:39 <Alberth> you can just access the file system for a directory listing?
12:19:02 <roboboy> but then I have to format it :P
12:19:14 <roboboy> I would rather the server do it for me
12:20:34 <Alberth> and the formatter is only accessible through http?
12:20:40 <roboboy> and I have to get the timestamps and the type of directory entries
12:21:02 <roboboy> well I got told the page didn't exist as there is no index file
12:21:20 <roboboy> I presume it only works through http based on that
12:22:05 <Alberth> so you want to access a routine not through http, which can only be accessed through http?
12:22:58 <roboboy> It looks like that is the case :(
12:23:44 <Alberth> the only other option is to duplicate/copy the formatter routine into your own code
12:25:20 <roboboy> it sounds like that is the case
12:26:18 <Alberth> hmm, or embed the server directory listing as a sub-page in your own page? :p
12:26:43 <Alberth> (no idea if that's possible, I don't know enough of html)
12:29:47 <roboboy> I think I'll replicate the servers functionality
12:37:42 *** SamanthaD has joined #openttd
12:38:15 <SamanthaD> hey Alberth and krinn!
12:39:37 <SamanthaD> Does anyone know what's up with the Finnish Trainset?
12:39:55 <krinn> it's not finish ? (ok not my best one)
12:40:02 <SamanthaD> it's like... the steam engines cost ~20k but you get to the diesels and the electrics and they're running ~500k
12:47:43 <Alberth> maybe coal is very cheap in Finland?
12:50:27 *** sla_ro|master has joined #openttd
12:53:34 <SamanthaD> it's complaining that git.openttd.org/info/refs not found
12:55:38 * SamanthaD has decided to use git instead of svn for tracking OpenTTD
12:56:49 <Xaroth|Work> you can use the github repo
12:56:56 <Xaroth|Work> and the github client
12:57:23 <SamanthaD> I'd rather use the client I've got on my computer
12:57:30 <SamanthaD> why, is the github repo better for some reason?
12:57:37 <Xaroth|Work> no, but the github client is
12:58:28 <Alberth> then someone needs to spend more time on the local applications, imho
12:59:44 <SamanthaD> lots of people complain about git being confusing in its interface
12:59:47 <SamanthaD> and honestly, it kinda is
13:00:48 <SamanthaD> I'm only using git because I want to maintain my own branch locally
13:00:49 <Alberth> it's the main reason I use hg instead
13:01:26 <Alberth> hg and git can basically do the same things, but hg has a much easier ui
13:01:56 <SamanthaD> I don't have any experience with Hg but if you folks say it's worth using instead of Git... I suppose I'll give it a shot
13:02:03 <krinn> must be why i find it easy, i have no ui :)
13:02:41 <Alberth> hg is more svn-like, with commands to select what you want. git has less commands, and puts a lot of things in options to the commands
13:03:28 <SamanthaD> and I can easily create a local branch with Hg, right?
13:03:33 *** TrueBrain has joined #openttd
13:03:35 <Xaroth|Work> the amount of commands hg has is confusing as hell
13:03:59 <Xaroth|Work> git, especially with the github client means I don't have to think too much
13:04:11 <Xaroth|Work> plus, it has a sane 'undo' button
13:04:19 <Xaroth|Work> i -really- like the undo button
13:04:23 <Xaroth|Work> as i derp around a lot
13:04:41 <Alberth> Xaroth|Work: don't compare one VCS with a gui with another one without gui
13:05:00 <Xaroth|Work> Alberth: I compare my experience with said VCS
13:05:07 <Xaroth|Work> if HG doesn't have a sane gui, that's their fault
13:05:11 <Xaroth|Work> git does, so I use it
13:05:26 <LordAro> i'm told good things about tortoise[svn|hg|git]
13:05:46 <Xaroth|Work> plain git, like hg, and svn, are command line client
13:05:51 <Xaroth|Work> s/client/clients/
13:05:54 <Alberth> SamanthaD: I have a clone to track upstream, and make local clones for experiments/patches
13:06:13 <SamanthaD> Alberth: instead of branches?
13:06:28 *** sla_ro|master has joined #openttd
13:07:03 <Alberth> you can have branches, but if you do that in the same repo, you cannot easily compare one branch with another one, at the file system
13:07:18 <Xaroth|Work> the gui clients are just to make things easier (if possible), and visualise things easier
13:07:20 <SamanthaD> also, I don't like GUIs with my development tools. I had a bad experience with them in middle school where I got overdependent on my IDE
13:07:38 <Alberth> I found just making more clones easier, you can quickly make new copies and discard repos you messed up
13:08:04 <Alberth> SamanthaD: I hate guis for not being scriptable :)
13:08:28 <SamanthaD> Alberth: Mouse pointer scripting?! ;)
13:08:29 <Alberth> note that clones at the same file system share the history
13:08:51 <Xaroth|Work> SamanthaD: using them and being dependant on them are very far apart
13:09:00 <Alberth> SamanthaD: concurrently over 5 desktops ;)
13:09:50 <SamanthaD> Alberth: That's okay! X11's network interface has you covered!
13:10:41 <Alberth> for some reason, command-line tools seem a lot easier :)
13:11:05 <krinn> Alberth, yeah just put the one need in a bash script, and voila, no need to remember or learn hg
13:11:10 *** oskari89 has joined #openttd
13:12:26 <SamanthaD> I dunno... I suppose as long as you know how to use the CLI you can use the GUI too
13:12:31 <Xaroth|Work> my deploy scripts all use the CLI; but I really can't be arsed having to type (or run the right script) stuff while developing on my desktop, when there's a perfectly useful GUI for that.
13:12:43 <SamanthaD> it's just that when I was little I'd learn to use the GUI and then when the GUI didn't do what I needed to I'd be flat on my butt and helpless
13:13:04 <Xaroth|Work> SamanthaD: there's google for those situations :P
13:13:15 <Alberth> SamanthaD: thg (tortoisehg) is very nice gui tool for examining a hg repo
13:13:35 <Xaroth|Work> tortoise* by default offers nice interfaces
13:13:56 <SamanthaD> and it's part of Debian!
13:14:09 <SamanthaD> praise the great sky compiler!
13:14:10 <Alberth> what more can you desire! :)
13:14:21 <blathijs> amiller: Using TortoiseSVN can be a challenge, even (especiallly?) if you know commandline svn...
13:14:27 <SamanthaD> Alberth: A Debian backport in a year? ;)
13:14:42 <blathijs> s/amiller/SamanthaD/
13:14:59 <Xaroth|Work> tsvn isn't that bad
13:15:10 <Xaroth|Work> but the cli offers some stuff that's hard to find in tsvn
13:15:12 <Alberth> SamanthaD: switch to fedora :)
13:15:37 <SamanthaD> Alberth: I tried that a few months ago. It crashed a lot.
13:15:40 <blathijs> Xaroth|Work: I wasn't saying it's bad, but sometimes they changed terminology in subtle ways, so you have to change the way of thinking at some points :-)
13:15:53 <Xaroth|Work> hard to find sometimes
13:16:21 <Xaroth|Work> thg and tgit have the same issue
13:16:25 <Alberth> SamanthaD: yeah, I agree it has some unstable edges every now and then
13:16:25 <SamanthaD> I'll just stick to the CLI and maybe a program to graphically display commits
13:17:24 <Alberth> searching through the commit history (hg annotate)
13:17:31 <SamanthaD> They have a QT applet that apparently does just that already though
13:18:08 <SamanthaD> hgview, according to apt-cache
13:19:25 <Alberth> that's the package name?
13:20:10 <SamanthaD> that is the package name
13:20:34 <SamanthaD> apt has hgview, hgview-common hgview-curses
13:20:51 <SamanthaD> according to apt-cache, hgview is the qt4 interface so... maybe hgview-qt4?
13:22:25 <SamanthaD> you folks are right... the mercurial CLI *is* a lot simpler than git!
13:24:53 *** Aristide has joined #openttd
13:25:52 <SamanthaD> yay! I know enough to use mercurial as an exceedingly complex wget!
13:25:55 <SamanthaD> I feel hackerish now :3
13:26:11 <LordAro> "mercurial as an exceedingly complex wget!" lol :D
13:27:00 <SamanthaD> exceedingly complicated, bandwidth inefficient wget!
13:28:32 <Alberth> it pulls the entire project history, instead of just the current version
13:28:50 <SamanthaD> Alberth: That's why it's inefficient ;)
13:28:50 *** Aristide is now known as Citelis18
13:29:21 <Xaroth|Work> I survive by that list :P
13:30:23 <SamanthaD> I'm bookmarking that for sure!
13:30:46 <krinn> Alberth, Latest version available: 1.7.1-r1 (love gentoo)
13:30:53 <Citelis18> Service Temporarily Unavailable
13:30:53 <Citelis18> The server is temporarily unable to service your request due to maintenance downtime or capacity problems. Please try again later.
13:31:11 <LordAro> Citelis18: github? worked for me, try again :)
13:31:34 <SamanthaD> the redhat bugtracker is bugged!
13:31:44 <Citelis18> The proxy server received an invalid response from an upstream server. | The proxy server could not handle the request GET /show_bug.cgi. | Reason: Error reading from remote server
13:31:46 <LordAro> i get a "proxy error"
13:32:07 <LordAro> fix your distro, Alberth :p
13:32:43 <krinn> KDE > fat bottom gorilla
13:32:59 <LordAro> Cinnamon > ... actually i'm not sure whether i like it yet
13:33:13 <SamanthaD> fat bottom gorilla > Gnome3
13:33:25 * krinn don't cares it still use gnome2
13:34:25 <Citelis18> Xaroth|Work: Fake (work)
13:34:52 <krinn> didn't test it yet SamanthaD i don't like swap GUI/distro... and my latest try (gnome3) has calm me down for a year
13:35:04 <Xaroth|Work> Citelis18: original.
13:35:04 <Citelis18> My KDE desktop is GOOD !
13:35:13 <Citelis18> Xaroth|Work: What is original ?
13:35:21 <Xaroth|Work> your poor attempt at a joke
13:35:25 <Xaroth|Work> it sais |Work, not |Working
13:36:19 <SamanthaD> krinn: Ah! Maybe give it a try in a year or so. It's basically just a Gnome2 fork but somehow they managed to break stuff along the way. I haven't used it in a year or so though and never seriously.
13:37:40 <krinn> SamanthaD, i'm not really into getting new things, that gnome2 is rock stable, it's enough for me, if i need a tools, i won't use the crappy version provide by the manager anyway
13:39:07 <SamanthaD> well, carry on then :3
13:39:39 <krinn> well, as long as i can maintain it without pain i will stick with it
13:39:55 <SamanthaD> I haven't strayed from XFCE for the better part of a decade other than a few forays to try out the "new shiny thing(tm)"
13:40:27 <SamanthaD> they haven't changed their UI significantly since then
13:40:34 <SamanthaD> just been progressively more polished
13:40:46 <krinn> and that's good no? if they change the UI it's not XFCE anymore
13:41:02 <krinn> and you are using XFCE because you love the UI
13:41:29 <krinn> that's the gnome3 stupid concept : change gnome to not be gnome.
13:41:35 <SamanthaD> Yup! It's kinda like a cross between Gnome2 and KDE3
13:42:23 <krinn> worst of the both world ? (kidding)
13:54:33 *** Aristide has joined #openttd
13:56:17 *** Aristide has joined #openttd
14:05:57 *** sla_ro|master has joined #openttd
14:55:00 *** Aristide has joined #openttd
15:05:40 *** amiller has joined #openttd
15:56:52 *** amiller has joined #openttd
15:59:13 *** TheMask96 has joined #openttd
16:04:49 *** HerzogDeXtEr has joined #openttd
16:06:46 *** frosch123 has joined #openttd
16:36:24 *** sla_ro|master has joined #openttd
16:46:15 *** sla_ro|master2 has joined #openttd
16:54:25 <SamanthaD> Is there a convenient way for me to make mercurial update to a given SVN revision number?
16:56:02 *** Progman has joined #openttd
16:58:35 *** ntoskrnl has joined #openttd
16:58:37 <NGC3982> I guess having the feeling of living in the Groundhog Day makes for a bad day.
16:59:21 <frosch123> SamanthaD: hg log -k r12345 --template '{rev}' | xargs hg up -r
16:59:31 <frosch123> convenient enough? :p
16:59:36 <SamanthaD> frosch123: Very! Thank you!
17:00:20 <frosch123> actually the keyword might not be specific enough
17:00:35 <frosch123> maybe you need '(svn r12345)' or so
17:00:53 <frosch123> else it might hit on stuff like -Fix (r12345)
17:02:42 * SamanthaD took a moment to work out what the command did
17:03:05 <frosch123> yeah, i learned it from alberth :)
17:03:07 *** MadaraUchiha has joined #openttd
17:03:29 <SamanthaD> frosch123: he seems like a clever guy!
17:03:32 *** sla_ro|master2 has quit IRC
17:03:37 <MadaraUchiha> I've read that the Difficulty Settings menu was removed from its place in the latest version.
17:03:53 <MadaraUchiha> How can I access it (or the options it used to have, if it doesn't exist anymore?)
17:04:12 <frosch123> difficulty and news settings were moved to advanced settings
17:04:24 <frosch123> also mind the filter options at the top of the advanced settings
17:04:44 <SamanthaD> I don't think anyone ever played unmodified difficulties anyway...
17:04:50 <frosch123> they defautl to "basic" settings", which hide a lot of the rarely needed options
17:05:34 <SamanthaD> MardaraUchiha: But do note the filter! There's lots of difficulty-changing options that "basic" likes to hide!
17:09:32 <Alberth> I really explained --template '{rev}' ? :o
17:10:24 <Alberth> well, ok, it seems to cover your needs :)
17:14:14 <Alberth> ah, right, that seems more likely :)
17:16:25 <frosch123> NGC3982: it's not only a sc2 player
17:16:25 <NGC3982> The anime that should have ended with Zabuza.
17:18:43 <NGC3982> Wait what. I'm wait out of context.
17:21:29 <MadaraUchiha> @NGC3982 How uptodate are you?
17:22:44 *** Aristide has joined #openttd
17:25:27 <NGC3982> About what, the anime or the games?
17:25:36 *** sla_ro|master has joined #openttd
17:26:23 <NGC3982> I think i ended Naruto around the Konoha attack thingy. I remember watching it as it was broadcasted, so i guesd it was a long time ago.
17:32:05 *** megakacktus has joined #openttd
17:32:43 <SamanthaD> okay, one more stupid question: I need to find the last revision with a certian file in it (in mercurial)
17:33:11 <SamanthaD> specifically, I need to figure out what's going on with src/table/settings.h
17:33:42 <Alberth> it got moved to src/table/*.ini
17:34:02 <Alberth> hmm, not sure what's in settings.h
17:34:05 <SamanthaD> Alberth: Just a flat out move?
17:34:16 <Alberth> nah, that would be too easy :p
17:34:32 <Alberth> where is that settings.h file?
17:34:56 <SamanthaD> anyway... my up-to-date version shows that there are two files. src/table/settings.h.postamble and a *.preamble
17:35:58 <frosch123> well, i guess you can say that settings.h was moved into objs/settings/table/ :p
17:36:05 <frosch123> it's now a generated file
17:36:40 <Alberth> right, I was looking for that :)
17:37:13 <Alberth> what entry do you need?
17:37:43 <SamanthaD> I just have a patch that modified it and I was trying to find the revision that moved it so I could update to the one right before that, merge, then go forward
17:37:45 <DorpsGek> Commit by frosch :: r25699 trunk/src/lang/english.txt (2013-08-07 17:37:41 UTC)
17:37:46 <DorpsGek> -Fix-ish: Hopefully make the settings type filter dropdown less confusing.
17:39:24 <SamanthaD> ah... it adds a bunch of settings
17:40:15 <Alberth> perhaps the first revision of a .ini file (which is svn r22172)
17:43:20 <SamanthaD> why the heck is settings.h.postamble just an empty file?! O.o
17:45:18 <DorpsGek> Commit by translators :: r25700 trunk/src/lang/slovenian.txt (2013-08-07 17:45:14 UTC)
17:45:19 <DorpsGek> -Update from WebTranslator v3.0:
17:45:20 <DorpsGek> slovenian - 21 changes by matej1245
17:46:10 <NGC3982> You know you have a pirated version of Android when you get the error message: "You are low on battery. Please insert SIM card".
17:46:14 <Alberth> consistency with the other .postamble files
17:46:34 <NGC3982> There isn't even hardware for a SIM card in the device.
17:46:38 <SamanthaD> NGC3982: I dunno... I wouldn't count on it ;)
17:46:39 <Alberth> I cannot really find where the file disappeared
17:47:02 <SamanthaD> Alberth: Lemmie dig around in this a bit, I might be able to figure it out
17:47:13 <megakacktus> How exactly is the file list generated / stored? I'm having trouble setting up a filterable GUIlist :-/
17:49:49 <Alberth> SamanthaD: r22174 adds the new system, but without using it in the build process, so after that the switch was made
17:50:04 <Alberth> megakacktus: doesn't the code build a list?
17:50:11 <frosch123> megakacktus: maybe keep the filelist generation as it is, but transfer it into a GuiList
17:50:24 <SamanthaD> Alberth: Thank you, I'll keep that in mind.
17:50:24 <frosch123> filelist generation is also used for console interface etc
17:51:02 <frosch123> most lists use a Guilist<> to store/sort/filter the stuff
17:57:04 <Alberth> SamanthaD: ah, r22188 removes the file
18:02:35 * SamanthaD is glad she's doing this doing this by hand
18:02:39 <Alberth> Luckily I have never had to deal with that so far :p
18:03:13 <SamanthaD> there's a post 22174 update to the patch made by a third party on the forums but it diffs settings.ini
18:03:21 <SamanthaD> and... since that's apparently a temp file now...
18:04:08 <Alberth> you'd need to find where the real switch is made
18:04:50 <SamanthaD> I used her patch before
18:04:55 <SamanthaD> I found it unreliable
18:05:00 <SamanthaD> now maybe I know why >.<
18:06:24 <Alberth> oh, it's also in r22188
18:09:38 <megakacktus> frosch123: How would I transfer it into a GuiList? with a typedef?
18:10:15 <frosch123> the guilist would be a member of your window
18:10:24 <frosch123> then you have some BuildFileList ufnction tor so
18:10:39 <frosch123> which queries the filelist using the existing functions, and adds all items to the guilist member
18:10:45 <frosch123> then that thing is filtered and sorted
18:16:16 <frosch123> megakacktus: maybe take a look at the NetworkGameWindow
18:16:23 <frosch123> i think it's kind of the easiest list with a filter
18:16:44 <frosch123> the other lists with filters have some special stuff
18:19:37 <megakacktus> ok... I'll see what I can do :)
18:19:42 <frosch123> V453000: watching again?
18:20:16 <V453000> wanted stephano to wreak all ass though :(
18:20:59 <frosch123> yeah, it's unusual that protoss is winning :p
18:21:23 <frosch123> but i think naniwas reaction to immediately place a second pylon for powering the gate was awesome
18:22:34 <frosch123> anyway, all ro8 from last season are eliminated :p
18:22:48 <frosch123> it's a game of pure luck :p
18:23:34 <frosch123> (ok, prejuding happy here)
18:30:19 <megakacktus> dumb question: what is CDECL?
18:30:58 <frosch123> one of about infinitely many 32bit calling conventions
18:31:41 <megakacktus> what does it stand for?
18:32:29 <frosch123> megakacktus: enjoy 64bit, and ignore that stuff
18:32:39 <frosch123> it's just needed to make it compile on 32bit
18:34:33 <SamanthaD> quick question if anyone knows off the top of their head... in the translation files under the STR_CONFIG_SETTING there's a bunch of new entries that end in _HELPTEXT
18:34:50 <SamanthaD> anyway... the patch I'm applying is older than those
18:34:58 <frosch123> SamanthaD: opent the advanced settings gui, and watch the bottom part closely :)
18:35:22 <SamanthaD> right... my question is do I need to do anything special or should I just add a new entry in the translation file?
18:35:36 <frosch123> you need to reference it from the .ini file
18:35:46 <SamanthaD> which is autogenerated
18:36:01 <frosch123> the .ini is not generated
18:36:12 <frosch123> .ini and .txt are source
18:37:16 <SamanthaD> I thought we just established that esrc/table/settings.ini is autogenerated from settings.h.preamble...
18:37:40 <frosch123> objs/.../settings.h is generated
18:37:43 <SamanthaD> worst that can happen is I don't get a helptext
18:37:59 <frosch123> "objs" is generated, "src" is source
18:38:01 <SamanthaD> so, which .ini should I be editing?
18:38:49 <frosch123> most liekly settings.ini
18:39:18 *** oskari89 has joined #openttd
18:39:26 <SamanthaD> I can't believe I didn't see that staring me in the face >.<
18:39:46 <SamanthaD> I'm going to go take a five minute break
18:39:50 <SamanthaD> I'm obviously going crosseyed
18:41:55 *** andythenorth has joined #openttd
18:44:41 <andythenorth> what's happened?
18:44:57 <Rubidium> andythenorth: you left, and then you joined again
18:45:10 <Rubidium> andythenorth: nothing in between as the world revolves around you ;)
18:45:12 <andythenorth> frick, that reminds me
18:45:18 <andythenorth> I have to submit that bug report :(
18:45:40 <andythenorth> have to make a savegame and everything
18:46:53 <frosch123> and all of that only to cause work for others :p
18:47:15 <andythenorth> the world is a sad place
18:47:19 <andythenorth> but...firstworldproblem
18:47:37 <frosch123> can we somehow buy some island in the southern hemisphere together?
18:48:15 <andythenorth> what would we do there?
18:49:05 <frosch123> isn't that the common dream to not do anything, and just vegatate staring straight into the sun?
18:49:15 <andythenorth> vegetables can do that
18:54:23 <andythenorth> would we write code?
18:54:36 *** NeuhNeuh has joined #openttd
18:57:01 *** Progman has joined #openttd
19:08:44 <megakacktus> so this is turning out to be much more complex than I original expected :(
19:12:16 <frosch123> ottd is big and complex :p
19:12:42 <frosch123> all custom code, few libraries
19:13:32 <glx> but at least a very big library ;)
19:18:31 <andythenorth> so I had an idea about FIRS providing info to GS
19:18:40 <andythenorth> dunno if it's useful or stupid
19:19:25 <andythenorth> no idea how Squirrel consumes data
19:23:10 <frosch123> andythenorth: static info or dynamic info?
19:23:27 *** Aristide has joined #openttd
19:25:17 <frosch123> or actually: 1) constants specific per industrytype, 2) specific per industrytype but depending on grf parameters, 3) dynamically depending on industry instance and changing over time
19:25:53 <megakacktus> I'm trying to loop through _fios_items, which is declared from a SmallVector<T,S> template...
19:26:06 <megakacktus> what should the type be that I'm pointing to?
19:26:40 <frosch123> use it like a std::vector, just capitalise the member functions, and use S* as iterator
19:27:30 <frosch123> nevermind, member functions are actualy vastly different
19:30:29 <frosch123> andythenorth: anyway, squirrel prefers to consume generic data as json
19:30:54 <andythenorth> frosch123: constants
19:31:00 <andythenorth> this would be outside of openttd
19:31:14 <andythenorth> it would be compile time stuff, inserted via good old copy+paste
19:31:29 <frosch123> inserted into what?
19:32:00 <frosch123> wouldn't you rather add something to firs?
19:32:09 <frosch123> so it is bundled with the grf, and fits to it?
19:32:29 <andythenorth> 'would rather' yes probably
19:32:55 <frosch123> well, hardcoding industry ids into a script library would be a stupid solution
19:33:00 <frosch123> that most likely just breaks
19:33:10 <andythenorth> this was a hack idea to reduce work for GS authors having to personally maintain their script against FIRS
19:33:14 <andythenorth> it's not optimal :)
19:33:21 <frosch123> detecting industries on input/output cargo combinations might not be unique enough
19:33:40 <andythenorth> we've had a groundhog day conversation about that kind of thing too many times :)
19:33:55 <andythenorth> I want to add some kind of metadata to cargos
19:34:01 <andythenorth> and maybe industries
19:34:12 <andythenorth> maybe we should just extend newgrf spec :P
19:34:26 <frosch123> (1) would be easy, it would just be adding a json string via a14, which scripts can read
19:34:46 <frosch123> (2) and (3) are way more tricky, since you somehow need to isnert parameters into the json
19:34:57 <andythenorth> this was actually wrt cargos, which is probably simpler
19:35:02 <andythenorth> but the industry case is valid too
19:35:13 <frosch123> what's the problem with cargos?
19:35:25 <frosch123> scripts can identify them via cargo labels, what else do they need?
19:35:49 <frosch123> or do their meaning change with firs parameters?
19:35:58 <andythenorth> far as I can tell from the forum thread (waiting for a reply), a bit more info about 'intention' might be useful
19:36:07 <andythenorth> like, is this a good cargo for town growth?
19:36:22 <andythenorth> this is all guesswork, no real use case yet
19:36:30 <andythenorth> but guessing is fun
19:36:41 <andythenorth> use cases are for paid work :P
19:37:12 <frosch123> i rather thought about allowing newgrfs to tell scripts how the production mechanics of industries work
19:37:27 <andythenorth> that is interesting
19:37:30 <andythenorth> and complicated :)
19:37:30 <frosch123> i have no idea what would be interesting about cargos
19:38:15 <andythenorth> I want to play that GS though
19:39:15 <frosch123> andythenorth: well, simplified like: cargo affects output few/lot, on its own/only with other cargos, unlimited/limited to 100 per month
19:39:39 <andythenorth> yeah something like that
19:40:01 <andythenorth> I think there is a need for something between 'totally generic' and 'GS author has to know all details of newgrf'
19:40:23 <andythenorth> cargo labels and railtype labels have worked well imho
19:40:33 <andythenorth> despite some arguments
19:40:38 <frosch123> well "know all details of newgrf" will fail, considering how quickly firs economies change :p
19:40:54 <andythenorth> and that's now by design, not by mistake :)
19:42:59 <andythenorth> 'cargo can make electricity' :P
19:43:30 <frosch123> that either depends on the industry accepting it, or the script can decide from the cargolabel itself
19:43:51 <frosch123> there are not so many cargos overall
19:44:08 <frosch123> scripts can easily detect the 5 they are interested in
19:48:45 <andythenorth> ok so maybe industries are the significant aspect
19:49:33 <frosch123> yeah, but they depend on grf parameters, and maybe even more
19:49:50 <frosch123> so, we need some way to insert dynamic stuff into json strings with grfs :p
19:52:07 <frosch123> should we create some examples how the result information could look like?
19:52:18 <frosch123> to get an idea about what dynamic parts would be there
19:52:25 <andythenorth> I am only +0.5 on the usefulness so far
19:52:31 <andythenorth> but usefulness is over-rated :)
19:52:41 <andythenorth> is it a neat thing to do? o_O
19:53:14 <andythenorth> frosch123: you wrote a GS, what did you want to know about the industries?
19:53:21 *** Aristide has joined #openttd
19:53:21 <frosch123> well, if we have some examples we can give them to krinn and other gs authors, and make them flame us how useless the stuff is :p
19:53:49 <andythenorth> static: 'this industry builds in towns'
19:53:51 <frosch123> andythenorth: i made a automatic classification of cargos and industries into secondary and primary
19:54:10 <andythenorth> using what criteria?
19:54:19 <frosch123> but it cannot distinguish secondary from tertairy, nor rare from often
19:54:26 <frosch123> so the goals turn out very different
19:54:33 <frosch123> you can see the same with nocargoal
19:54:45 <frosch123> goals are very different to achieve
19:54:58 <frosch123> i even have a bug report for sv, that it is unbalanced in multiplayer :)
19:55:20 <frosch123> anyway, industries have a raw/processing flag
19:55:26 <frosch123> which constrols funding/prospecting
19:55:40 <frosch123> that allows to classify industries into raw and processing industries
19:55:43 <andythenorth> limited information content though
19:55:46 <frosch123> and then i can look which industries produce what cargo
19:55:53 <andythenorth> some info there, but not enough
19:56:03 <frosch123> i.e. if cargo is only produced by raw, or only by processing cargos
19:56:11 <frosch123> cargos produced by both, are dropped
19:56:28 <frosch123> if you take a look at the debug output of sv, it prints the cargo classification at startup
19:56:35 <andythenorth> but you had to figure all that out for yourself?
19:56:41 <andythenorth> you can't just query the industry
19:57:12 <frosch123> i can query the input/output cargos of the industrytypes, like the chain graph does
19:57:19 <frosch123> and i can query funding/prospecting
19:57:25 <andythenorth> so either that should be a GS library for reuse, or every author has to figure it out?
19:57:36 <andythenorth> we have not many GS yet, and the barrier to starting seems high
19:58:15 <frosch123> we can try to advertise it to zuu or krinn, so they include it into some superlib for gs :p
19:58:20 <andythenorth> so industry is primary / secondary / tertiary seems valid
19:58:24 <frosch123> but i don't feel like maintaining a gs library myself
19:58:46 <andythenorth> maintaing API info in FIRS I would be happy to do
19:59:07 <andythenorth> would also benefit from this info?
19:59:27 <frosch123> i guess the info would be the same
19:59:38 <frosch123> unless newgrf want to hide info from ais, and give gs extra tipps :p
20:00:04 <andythenorth> so more detailed industry type info seems valid
20:00:08 <frosch123> but i would rather prefer if newgrf/gs/ai team up against humans :p
20:00:11 <andythenorth> industry can have production boosted?
20:00:34 <andythenorth> supplying this industry should boost town growth?
20:01:38 <frosch123> andythenorth: "input cargo 2 affects output cargo a lot, but only together with other cargos, and only up to a soft maximum of 100 units per month" ?
20:01:51 <andythenorth> how to express that in json?
20:02:10 <andythenorth> FIRS uses production ratios
20:02:14 <andythenorth> they might be valid
20:02:21 <frosch123> i guess a map from keywords to integers
20:02:42 <andythenorth> I am not bothered about making this totally generic; there are not many industry sets, and there are definitely _some_ similarities between ECS and FIRS
20:03:26 <frosch123> i don't think it should contain more than 4 things per cargo
20:03:32 <andythenorth> cargo 2: affects output cargos [1 | 2]; amount [high]; limit [100]
20:03:33 <frosch123> else it is way too complicated for scripts
20:04:14 <andythenorth> or express it in terms of what is needed to produce output cargos
20:08:35 *** Aristide has joined #openttd
20:08:53 <frosch123> "maximum_monthly_supply" for hard stockpile limits
20:09:19 <frosch123> "recommended_monthly_supply" for production boosters that have some maximum effect at some point
20:09:31 <frosch123> "direct_output" for stuff that is converted into output at a fixed rate
20:09:48 <frosch123> "shortterm_output_multiplier" for stuff that increases output temporarily
20:10:02 <frosch123> "longeterm_output_multiplier" for stuff that increases output permanently
20:10:23 <frosch123> "minimum_transport" for stuff that must be transported from the industry, for it to be happy
20:10:38 <frosch123> one would have to look up the weirdnesses of ecs and pbi :p
20:11:38 <frosch123> outputcargo might also require some "reservoir_left" for pbi expiring mines
20:11:46 <andythenorth> 'recommended' is a nice idea
20:11:55 <andythenorth> good point about PBI
20:12:11 <frosch123> so, basically a bunch of keywords with either string, integer or float as value
20:13:04 <frosch123> ideally only a few keywords, potentially estimating most industries of most sets approxiamtely
20:13:49 <andythenorth> is there a way to do this without requiring all the tools (nml, grfcodec, renum) to be updated? :P
20:17:14 <frosch123> oh, i remember there were steel mills in some set which required both coal and iore to produce anything
20:17:40 <frosch123> so, maybe "minimum_supply_ratio"
20:17:58 <frosch123> well, basically we need an overview of all industry mechanics there are :p
20:18:05 <V453000> are you guys making new specs for industries or what :D
20:18:33 <frosch123> V453000: we are trying to figure out a way, how newgrf can tell gs/ai how their industries behave
20:19:19 <andythenorth> imho, idealy do this in a way where any group of industry author / GS author could agree to go and do something completely different to what we thought of
20:19:36 <andythenorth> but meanwhile supporting the sets we know do exist, which aren't many
20:19:44 <frosch123> andythenorth: anyway, some industry categorisattion like "city", "rural", .. would be entirely different
20:19:51 <V453000> I might code my industries this year so expect anything :)
20:19:56 <frosch123> might be rahter a list of arbitrary keywords
20:20:09 <frosch123> though considering bananas keywords, keyword are crap
20:20:17 <andythenorth> keywords are crap
20:20:48 <andythenorth> make them a bit more costly: 4 letter labels or something
20:20:54 <andythenorth> and you're only allowed 1 per industry?
20:21:09 <andythenorth> then authors have to think harder
20:21:25 <V453000> please make a label OMFG and I will use that
20:21:26 <frosch123> no, 1 per industry would result in many different ones
20:22:09 <frosch123> and usually newgrf authors do it the vehicle way and just set a flag "this vehicle is too complicated for ais" :p
20:23:30 <frosch123> and considering banans tags, they would likely add keywords like "awesome" or their name, and even misspell them
20:27:32 <frosch123> boring games today :p
20:29:09 *** HerzogDeXtEr1 has joined #openttd
20:31:27 <andythenorth> I would definitely add an 'awesome' flag to all industries
20:36:16 <krinn> you want a GS to handle FIRS ?
20:36:49 <krinn> there's no way a GS could handle a newGRF if the newGRF don't do something to hint the GS
20:37:54 <andythenorth> unrelated, but can GS change the production multiplier yet?
20:38:31 <krinn> there's simple no way to speak with a newGRF
20:38:46 <andythenorth> this isn't the newgrf
20:39:00 <andythenorth> production multiplier is inherent to industry ;)
20:39:08 <frosch123> would it work with firs?
20:39:33 <andythenorth> not currently, but it could
20:39:36 <krinn> well, the API cannot control that for "classic" one
20:39:49 <andythenorth> the FIRS production boost could easily use the production multiplier
20:40:00 <andythenorth> only reason it doesn't is because yexo wrote code that I don't understand :)
20:40:08 <andythenorth> and I'm too lazy to restart it from scratch
20:41:24 <frosch123> well, there is no need to, as long as there is no concept :p
20:41:49 <andythenorth> having GS be able to adjust production multiplier would be neat I think
20:41:50 <frosch123> newgrf providing info to scripts via json might work
20:42:07 <frosch123> scripts giving instructions to newgrf would definite require some special library
20:42:17 <andythenorth> not the way I envisage it
20:42:38 <andythenorth> GS would monkey patch random / monthly production change
20:42:52 <andythenorth> it's up to newgrf author how that is handled :P
20:42:58 <andythenorth> I would allow it for FIRS
20:43:17 <andythenorth> "your entire map just went into recession, oops"
20:43:40 <andythenorth> "workers are on strike at this industry, bad luck"
20:43:58 <andythenorth> "you delivered 250 pax / month to this town, now the industry has new workers, well done"
20:44:30 <krinn> well, that's not in the API, nothing from GS->Industry only Industry->GS and not really a lot of function for that, just the basic
20:44:36 <frosch123> that sounds more like scripts setting an additional production/delivery multiplier
20:44:48 <andythenorth> my suggested method might be bad
20:44:55 <frosch123> script sets multiplier to 0.5 -> gui says "50% of production lost"
20:44:58 <andythenorth> but it has the advantage of being simple and following existing spec
20:45:08 <andythenorth> and my idea works with default industry...
20:45:24 <andythenorth> other ideas may also be good :)
20:45:45 <frosch123> well, i don't like messing with existing methods
20:45:55 <frosch123> add a new clean one instead
20:46:06 <frosch123> which is cleary meant for script influence
20:46:14 <frosch123> no convoluted thing to cover all
20:46:44 <frosch123> it would also simplify thing for ais :p
20:46:45 <andythenorth> no nasty patch :)
20:47:08 <frosch123> ais and humans should be able to tell whether an industry is messed up by the grf or by the gs
20:47:16 <frosch123> esp. when the newgrf specifies its behaviour via json
20:47:46 <frosch123> it would be silly if industries could specify their behaviour, but gs would then change it completely again :p
20:48:37 <krinn> frosch123, i'm not sure i get it, but looks as silly as a town production been change by the GS
21:07:19 *** NeuhNeuh has joined #openttd
21:11:02 *** gynterk is now known as gynter
21:41:53 *** MatrixCL2 has joined #openttd
22:21:57 <SamanthaD> Did OpenTTD just get a new dependency? ./configure is complaining that I don't have any video driver development files found
22:26:41 <LordAro> did you manage to uninstall them?
22:27:49 <SamanthaD> weird... I seem to have done so
22:27:56 <SamanthaD> I wonder how I managed to do that?! O.o
22:28:09 <SamanthaD> gremlins are in my computer, apparently
22:28:32 <SamanthaD> OH! I see... libjpeg62-dev conflicts with them
22:28:54 <SamanthaD> OpenTTD > whatever I needed libjpeg62-dev for...
22:29:35 <SamanthaD> several hours of merging...
22:29:38 <SamanthaD> let's see if this compiles!
22:31:24 * SamanthaD wishes the commit log was more verbose
continue to next day ⏵
