IRC logs for #openttd on OFTC at 2025-02-06
⏴ go to previous day
01:23:01 *** Wormnest has quit IRC (Ping timeout: 480 seconds)
03:24:09 <ahyangyi> andythenorth: `def vehicle(self): return self.vehicle or self`
03:31:06 *** Wormnest has joined #openttd
04:00:08 *** debdog has quit IRC (Ping timeout: 480 seconds)
04:01:30 *** D-HUND is now known as debdog
04:45:38 <DorpsGek> - Update: Translations from eints (by translators)
05:32:39 <andythenorth> gwyd4016: Doesn’t align with openttd 🙂
05:33:15 <andythenorth> Buy meni doesn’t have models, it has vehicles 🙂
06:19:13 <andythenorth> maybe it has `Engines`
06:21:39 *** keikoz has quit IRC (Ping timeout: 480 seconds)
06:28:22 <gwyd4016> Perhaps it has trainsets?
06:28:30 <gwyd4016> I mean items do have model life
06:50:50 <andythenorth> the actual game has `Engines`, if I've read the buy menu file correctly
06:51:16 <andythenorth> I thought it had `GroundVehicleSpec` or something, but I misremembered that
06:53:19 <andythenorth> wonder what var 40 and 41 do with articulated vehicles
06:53:36 <andythenorth> presumably they count each articulated part
07:10:32 <andythenorth> Vehicle::Iterate eh
07:28:45 <andythenorth> trying to understand what nml does
07:29:05 <andythenorth> `position_in_consist` `The position of the current vehicle-part from the start of the vehicle. The engine will get value 0, the first wagon (or second engine) gets value 1, etc.`
07:29:16 <andythenorth> so that does account for articulated parts?
07:31:02 <andythenorth> trying to understand `PositionHelper`
07:34:34 <andythenorth> so other train vars have `t -> Train`
07:34:37 <andythenorth> wonder what RVs do
07:36:58 <andythenorth> `GVSF_ARTICULATED_PART` eh
07:37:54 <andythenorth> `IsEngineCountable` looked useful, but isn't what I needed
07:38:45 <andythenorth> do does `Vehicle::Iterate()` just yield 'next in the chain'?
07:40:05 <andythenorth> hmm so everything is a Vehicle
07:40:24 <andythenorth> Vehicles can have a pointer or something to a following vehicle?
07:41:00 <andythenorth> that following vehicle is bool flagged with GVSF_ARTICULATED_PART
07:41:30 <andythenorth> is first vehicle flagged with GVSF_FRONT?
10:05:46 *** mindlesstux has joined #openttd
10:15:53 <andythenorth> I could implement an equivalent of Next() in my python compile
10:16:47 <andythenorth> so instead of trying to store a list of `[<vehicle>, <vehicle>, <vehicle>]` in a non-existent structure, I just do `vehicle.Next()` until it returns None
10:17:15 <andythenorth> and to build the buy menu I walk all `<vehicle>` instances and drop any that are trailing parts
10:17:41 <andythenorth> I'd need something to get `vehicle.Front()`
10:18:00 <andythenorth> I guess I walk all the `<vehicles>` and just recurse the `Next()` chains backwards
10:20:50 <andythenorth> no I can store a reference to the front vehicle
10:20:59 <andythenorth> and then copy all properties into each vehicle
10:21:26 <andythenorth> ach, no the purchase menu doesn't do articulated
10:22:03 <andythenorth> ok, I build `purchase_foo` versions of every attribute and property
10:24:45 <andythenorth> object context is hard to infer at compile time, when it's determined per-instance by run-time properties
10:32:29 <andythenorth> is there a pattern name for OO, where insted of having different objects, or subclasses, for behaviour
10:32:54 <andythenorth> instead you have one class, and most behaviour is conditional inside it, depending on run-time context?
10:33:41 <andythenorth> so for e.g. book.page_count() you'd have "if hardback then 200 elif paperback then 300"
10:34:10 <andythenorth> and then if you want to ship on kindle, you modify all the methods and release
10:34:46 <peter1138> class hardback => book; class softback => book, class ebook => book;
10:36:10 <andythenorth> what about a combined edition?
10:36:15 <andythenorth> so 3 books in 1 book?
10:37:19 <Timberwolf> andythenorth: "god object" is the one I've usually heard
10:49:48 <peter1138> So then you have a "Book", then you have an "Issue" of a book.
10:50:09 <peter1138> And the Issue contains the physical properties. The Book contains the words...
10:51:20 <andythenorth> I could stop having this problem if I just closed my eyes to the grf spec 😛
10:52:02 <andythenorth> trying to refactor Horse and Hog to clean up historical weird evolution
10:52:15 <andythenorth> and make them more conformant to OpenTTD, so they're not inventing their own terms
10:52:29 <andythenorth> or using game/grf spec terms incorrectly
10:54:05 *** reldred has joined #openttd
10:54:05 <reldred> you're not trying to hammer a square peg into a round hole are you
10:54:32 <andythenorth> very close to the metaphor I'd use
10:55:08 <andythenorth> I'm trying to put the star through the circle hole
10:55:16 <andythenorth> I'd sit and shape sort for hours
10:55:58 <andythenorth> yeah that video I've seen, it's crack for my brain
10:56:11 <andythenorth> and that it's totally performed, not real
10:56:33 <peter1138> It's performed well.
10:57:51 <andythenorth> my brain is basically both the people in the video
10:57:55 <andythenorth> until one side wins
10:58:19 <andythenorth> usually the winning side puts everything in the square hole
10:58:32 <reldred> You’re the square peg, the round hole, and the guy with the hammer all at the same time
10:58:46 <andythenorth> nah, that's people in my family
10:58:47 <peter1138> andythenorth wants to be pegged?
10:59:03 <andythenorth> this is a channel about croquet now?
11:01:20 <peter1138> Isn't pog sone aussie term?
11:03:34 <reldred> It’s a military term, usually pogue but I’ve seen it spelt pog.
11:04:09 <andythenorth> hmm did OpenTTD put everything in the square hole?
11:04:12 <andythenorth> I think it did 😛
11:05:39 <peter1138> Ah, I had a phase of being awake when Aussie Twitch streamers were live, and many of them seems to use "pog" or "poggers" or somesuch. Didn't see it much anywhere else.
11:05:41 <andythenorth> oh, probably fair to actually blame TTD?
11:05:48 <andythenorth> poggers is thing
11:05:57 <andythenorth> cooked and cold are thing my kids seem to have imported
11:06:03 <andythenorth> appear to be Aussie
11:06:10 <peter1138> andythenorth, I recommend blaming everything on Patchman, as they're not around to defend it.
11:06:39 <andythenorth> where did articulated vehicles come from?
11:07:35 <peter1138> Cooked is a very unlewd (at least, I think it is...) way of saying screwed or shafted or bu*****d... It's amazing.
11:07:49 <peter1138> TTDPatch invented the, yes.
11:08:20 <reldred> Cooked also implies brain damage, drug use.
11:08:35 <reldred> Or just being left in the sun too long
11:08:43 <reldred> You know these things are all variable.
11:08:52 <andythenorth> cold currently means cool
11:09:26 <reldred> peter1138: Wasn’t just Australian streamers
11:09:49 <reldred> It was mostly a twitch thing
11:10:05 <andythenorth> I thought it was just a red pill thing
11:10:13 <andythenorth> I default to 'probably just red pill'
11:12:10 <andythenorth> ok so TTDP had to do 'fake OO using boolean flags' I assume?
11:15:01 <peter1138> Nothing about NewGRF is OO, nor is it intended to be.
11:15:47 <andythenorth> square hole time
11:15:48 <peter1138> It's possible that NML tries to be OO, but that's nothing to do with the spec.
11:17:32 <andythenorth> the interface for authors isn't very OO
11:18:25 <peter1138> I didn't say it is, I don't know it well enough to say.
11:20:22 <andythenorth> ah, I'm not really stuck so much on the nml generator
11:21:44 <peter1138> Bah, would be nice if I could automaticaly get the last enum value, although then things like Invalid = 0xFF would break.
11:21:49 <andythenorth> for vehicles, we're just running nfo act 0 / 3 / 2/ 1 in a specific vehicle, and we behave accordingly
11:22:20 <andythenorth> but in the docs I have to loop over all the vehicles that can be purchased by the player
11:22:33 <andythenorth> and there's no obvious name for what they are
11:22:37 <andythenorth> they're not 'vehicles'
11:22:53 <peter1138> Maybe add something like `End` to every enum and use that. But I'd rather not have End be a valid value of the enum. Hmm.
11:23:26 <andythenorth> what peter1138[d] said
11:23:42 <peter1138> Or "orientated" if you grew up not learning things properly in the 80s.
11:23:59 <andythenorth> hmm...the docs are a different domain, I can just ignore OpenTTD's implementation?
11:24:06 <andythenorth> that tends to cause concept drift though
11:24:33 <peter1138> But then it is orientation, not oriention.
11:26:11 <andythenorth> possibly the docs can just loop over 'models'
11:26:28 <andythenorth> implicitly 'vehicle_models', but that's a bit redundant maybe
11:28:59 <andythenorth> I keep thinking 'buyables' but that's horrible
11:29:14 <andythenorth> although I just found my printer has a 'printables' feature 😦
11:35:40 <peter1138> Hmm, nearly half-way to replacing DECLARE_ENUM_AS_BIT_SET.
11:37:37 <andythenorth> yeah ok so vehicle grfs primarily compose vehicles
11:37:48 <andythenorth> and trains also being a vehicle in the game
11:38:00 <andythenorth> and articulated parts also being a vehicle in the game
11:38:11 <andythenorth> I will have to put through the square hole
11:39:38 <andythenorth> and what you get in the buy menu are vehicle models
11:54:24 <_glx_> In buy menu you only get heads of articulated vehicles because no callback
11:57:01 <andythenorth> It’s one of the practical reasons my compile nested vehicles as ‘consist’ and ‘units’
11:57:24 <andythenorth> Then purchase is handled by consist
12:24:34 <andythenorth> It’s kind of a quirky implementation detail eh 🙂
12:57:18 <andythenorth> Maybe nml should have abstracted articulated vehicles better 🙂
13:04:18 <andythenorth> Can’t think of anyone for grf to do it
13:32:55 *** bohaska has joined #openttd
13:32:55 <bohaska> how hard would it be to implement a sandbox feature where you can directly set the growth rate of a town?
13:33:32 *** HerzogDeXtEr has joined #openttd
13:41:58 <LordAro> probably not that hard
13:56:31 *** emperorjake has joined #openttd
13:56:31 <emperorjake> JGRPP already allows enabling/disabling growth for individual towns
14:04:31 <bohaska> yea, but it doesn't let you set it directly
14:05:01 <bohaska> like you find people doing model railroad who wait for the town to grow until doing what they need to do
14:06:56 <xarick> May I suggest some default settings changes ? 🙂
14:08:12 <xarick> 1. Scoring year: Never
14:08:33 <xarick> 2. Limit airport placement based on noise level: On
14:09:49 <xarick> 3. Company stations can serve industries with attached neutral stations: Off
14:10:49 <xarick> yeah, I guess that's it, the others are too personal
14:20:06 <xarick> this doesn't exist... @see \ref ScriptCalendarTime
14:41:27 <_glx_> It does, it's an anchor in scriptdate doc
14:41:51 *** belajalilija has joined #openttd
14:41:52 <belajalilija> Since we’re making demands
14:42:22 <belajalilija> Could i request an option to bring the town growth button over from scenario editor into the main game
14:49:12 <andythenorth> Shame GS never took off
15:01:46 <xarick> compiling open ttd takes so much time I get distracted and start watching youtube videos
15:02:00 <xarick> then i forget what I was doing when i go back to it
15:03:07 <peter1138> Oh, Order is full of bitstuffed members :(
15:10:42 <xarick> GetWagonAge doesn't work as intended
15:10:57 <Rubidium> peter1138: yes, it's a hot mess
15:11:24 <xarick> if I'm reading this correctly, v->age is only increased for primary vehicles
15:12:17 <xarick> need to actually test to make sure
15:13:42 <peter1138> * Get the current age of a second (or third, etc.) engine in a train vehicle.
15:13:47 <peter1138> It's probably misnamed.
15:14:53 <peter1138> Yes, wagons don't age, only engines.
15:15:33 <peter1138> The conditional is about retiring engines.
15:15:40 <talltyler> We’re not making demands. Those default setting changes are not supported by survey data. 🙂
15:15:54 <peter1138> talltyler, that's because nobody changes the defaults ;-)
15:16:23 <peter1138> xarick, you are not reading this correctly, basically.
15:16:57 <peter1138> Right, `order->SetType(x)` gets around the first annoyance...
15:19:50 <peter1138> OrderNonStopFlags is not a "flags"...
15:22:33 <peter1138> Eh, it kinda is in some places. Hmm.
15:24:26 <peter1138> Sometimes we use `!=`, sometimes we use `&`. Fun.
15:24:35 <xarick> regression tests a train with 0 attached wagons... 😐
15:25:36 <xarick> oh nevermind, i found some more tests around wagons
15:29:19 *** Extrems has joined #openttd
15:30:20 <talltyler> I’ve seen quite a few places where we misuse flags but it still works, by accident…
15:30:44 <talltyler> Depot flags have some issues with this, I think
15:35:10 * Rubidium wonders what flags depots have
15:36:29 <xarick> oh, they age, i thought they wouldn't...
15:44:30 <LordAro> Rubidium: i hope you an peter1138 have a spreadsheet somewhere so you're not stepping on each others toes :p
15:46:53 *** Wormnest has joined #openttd
15:49:49 <talltyler> Right, not flags for depots but for depot orders: `OrderDepotActionFlags` and `OrderDepotAction` (why are there two containing the same values? I never figured it out when working on depot unbunching 😛 )
16:01:15 <xarick> wagons age confirmed, because i fail to read code
16:04:47 <xarick> GetWagonEconomyAge mouthful name
16:07:56 <xarick> nobody really cares about wagon ages
16:08:07 <xarick> funny to see it supported in AIs
16:10:40 <xarick> unless they have proper breakdown decay support
16:14:05 <peter1138> I have a few commits to PR, but didn't reach network-crypto yet.
16:14:48 <xarick> btw some AIs check station construction date to help determine the route is old enough to check profitability
16:15:13 <xarick> there is no economy_date_of_construction
16:17:55 <xarick> nobody thinks of the scripts!
16:17:57 <peter1138> `build_date` changes if you modify the station anyway.
16:18:48 <xarick> but the AIs probably don't change the station during the initial period
16:18:56 *** Flygon has quit IRC (Read error: Connection reset by peer)
16:21:55 <peter1138> OrderUnloadFlag is fun.
16:22:04 <peter1138> A bit for Unload, and another bit for NoUnload.
16:22:18 <xarick> peter, plz consider my unbunch PR too
16:22:32 <xarick> since you're touching that part
16:26:53 <peter1138> No, it's inappropriate to make such changes at the same time as refactoring.
16:32:11 <peter1138> CommandCost CmdModifyOrder(DoCommandFlag flags, VehicleID veh, VehicleOrderID sel_ord, ModifyOrderFlags mof, uint16_t data)
16:32:23 <peter1138> Well, that's... not great.
16:43:15 <peter1138> Rubidium, hmm, I have a patch to automatically mask the underlying storage to valid values.
16:44:17 <peter1138> I wonder if I make it automatically use `x::End` to do that, instead of requiring a separate limit.
16:45:49 <peter1138> Although it might not be desirable to always mask.
16:46:37 <peter1138> (For now the mask defaults to all bits, so unless it's explicitly reduced it gets optimised away.)
16:51:11 <peter1138> `using IndustryControlFlags = EnumBitSet<IndustryControlFlag, uint8_t, IndustryControlFlag::End>;
16:59:45 <peter1138> Hmm, looking for a GS that uses Industry::SetControlFlags
17:05:15 <andythenorth> I have one for FIRS but I’m out
17:06:03 <andythenorth> It’s buildabke in the FIRS v5-release-track branch
17:20:07 <ahyangyi> ahyangyi: After using the one weird trick the compliation time goes back to about 100 seconds, nice.
17:35:12 <peter1138> Do I need to worry about constexpr performance...
17:37:13 <peter1138> (But also, what if "End" is a valid value)
17:38:32 <ahyangyi> peter1138: Make compilation faster? 😛
17:38:54 <peter1138> Easy, get a faster computer.
17:42:43 <peter1138> A quick scan of our enums suggests that End should be fine.
17:43:31 <peter1138> But whether it is wise to have such automatically functionality.
17:47:05 <johnfranklin> “The reason why you need to learn LaTeX is that your hundred-page long dissertation cannot be easily dealt with Microsoft Word” “Get a faster computer”
17:49:04 <wensimehrp> heh lateh is slow already
17:54:51 <wensimehrp> Remove all constexprs and you'll get faster compilation.
17:54:51 <wensimehrp> At a consequence of longer loading time and slower game speed.
18:06:22 <andythenorth> peter1138: do you still need one?
18:30:42 <xarick> Do I sound like I'm contradicting myself?
18:37:41 <xarick> how do auto named saves look like when time is frozen
18:43:52 <andythenorth> how does the OO work for vehicles in src?
18:44:25 <andythenorth> is there some kind of VehicleSpec that's instantiated?
18:44:36 <andythenorth> oh, I could just look, silly me 😛
18:48:47 *** tokai|noir has joined #openttd
18:48:47 *** ChanServ sets mode: +v tokai|noir
18:55:39 *** tokai has quit IRC (Ping timeout: 480 seconds)
18:56:30 <peter1138> There's an Engine, which contains the common information, and inside that there's a union which contains the vehicle type (train, roadveh, etc) specific details. So it's not very OO.
19:00:15 <_glx_> it's as OO as C could be 🙂
19:03:47 <peter1138[d]> And there's no need for it to be OO.
19:06:58 <LordAro> is this naming thing still going?
19:11:28 <andythenorth> it's a curiosity rabbithole now
19:11:42 <andythenorth> not An Important Code Question
19:12:17 <LordAro> "now" implies that it was ever anything else :p
19:16:30 <andythenorth> engine_base.h is correct place?
19:29:47 <michi_cc> Isn't the new meta composition over inheritance anyway? 😎
19:36:37 <andythenorth> composition is the winning tactic in CSS
19:36:41 <andythenorth> but CSS is not programming 😛
19:41:11 <andythenorth> I just read what composition is in a python context (rather than Java)
19:42:02 <andythenorth> slightly confused that it even has a name as an approach 😛
19:42:09 <andythenorth> isn't it just 'programming'? 😛
19:47:38 *** XeryusTC has quit IRC (Ping timeout: 480 seconds)
19:48:18 *** Hazzard has quit IRC (Ping timeout: 480 seconds)
19:48:18 *** tneo has quit IRC (Ping timeout: 480 seconds)
19:48:21 *** planetmaker has quit IRC (Ping timeout: 480 seconds)
19:48:49 *** Hirundo has quit IRC (Ping timeout: 480 seconds)
19:49:19 *** Osai has quit IRC (Ping timeout: 480 seconds)
19:49:21 *** Ammler has quit IRC (Ping timeout: 480 seconds)
19:50:09 *** Webster has quit IRC (Ping timeout: 480 seconds)
19:50:21 *** V453000 has quit IRC (Ping timeout: 480 seconds)
19:51:18 *** fonsinchen has quit IRC (Ping timeout: 480 seconds)
19:51:19 *** Terkhen has quit IRC (Ping timeout: 480 seconds)
19:51:19 *** SmatZ has quit IRC (Ping timeout: 480 seconds)
19:51:21 *** Guest6616 has quit IRC (Ping timeout: 480 seconds)
19:51:24 *** avdg has quit IRC (Ping timeout: 480 seconds)
19:51:26 *** Yexo has quit IRC (Ping timeout: 480 seconds)
19:51:27 <andythenorth> going to use ModelDef and UnitDef in Horse
19:51:36 <andythenorth> hoping to get LineDef in somehow 😛
19:51:52 *** Smedles has joined #openttd
19:53:33 <andythenorth> maybe I just name all the internals after Doom WAD format
20:12:03 <dihedral> openttdcoop went dark ... :-/
20:15:24 <_glx_> oh indeed, even the parts that used to be accessible
20:15:49 <peter1138> andythenorth, gs which calls the industry control would be nice.
20:15:49 <dihedral> heart warming words, peter1138
20:15:59 <peter1138> Or I could try modifying the regression GS if I knew how...
20:16:08 <peter1138> I mean, it's not the first time is it.
20:16:18 <andythenorth> the FIRS one depends on FIRS 5 unreleased
20:16:23 <andythenorth> it has a grf check
20:16:36 <peter1138> Yes, it's unreleased so I don't have it.
20:17:01 <andythenorth> or you could just write a GS 🙂
20:18:20 <andythenorth> anyway this runs at GS init
20:18:27 <andythenorth> GSIndustry.SetControlFlags(industry, VulcanIndustryControl.DEFAULT_CONTROL_FLAGS);
20:18:39 <peter1138> Yeah, That's probably ... Vulcan? Hmm
20:18:43 <andythenorth> ` foreach (industry, _ in industry_list) {
20:18:43 <andythenorth> // set all flags as of June 2023, unset them later per-industry for specific behaviour
20:18:43 <andythenorth> GSIndustry.SetControlFlags(industry, VulcanIndustryControl.DEFAULT_CONTROL_FLAGS);
20:19:06 <andythenorth> if you load it with the FIRS grf it should do it and you should see the log
20:20:03 <peter1138> I rebased, so waiting on game build anway.
20:20:25 <andythenorth> you might need to turn logging down
20:21:38 <Rubidium> :O peter has a regression GS. Nice!
20:24:04 <xarick> that last merge is scary
20:24:16 <xarick> airports being either hangar or stations
20:24:25 <xarick> and having to pick one choice
20:31:50 <peter1138> Trying to validate if the mask is being applied.
20:32:24 <peter1138> It's definitely appled before the command is despatched, but I need it applied in the command as well.
20:32:59 <peter1138> Because it's not over the network there's no buffer serialisation happening to verify that too.
20:33:38 <peter1138> I wonder if a copy constructor should mask, but then there might be a lot of copying considering we don't pass by reference. Hmm.
20:34:56 <Rubidium> in theory it's already correct
20:35:19 <peter1138> Yes, unless you force it by passing a non-enum value.
20:36:22 <Rubidium> well... I wouldn't have gone on this side quest
20:38:50 <peter1138> Basically this the first place I've needed to validate the EnumBitSet contains valid values.
20:39:04 <peter1138> It comes from script land so it might not.
20:39:34 <peter1138> Not sure if the command can be issued remotely.
20:40:20 <Rubidium> then maybe an 'IsValid' on the enum bitset, and check that in the squirrel API and in the command?
20:40:24 <peter1138> But even so, EnumBitSet didn't need the functionality to test if only valid bits are set, yet.
20:40:25 <Rubidium> (or just in the command)
20:40:35 <peter1138> Rubidium, that is precisely what I am doing :-)
20:41:22 <Rubidium> well, you're talking about copy constructors. I'm talking about if (!enumMask.IsValid()) return CommandError();
20:42:18 <peter1138> Oh, no I'd already decided that was a bad idea.
20:43:28 <peter1138> It's masked by the command caller already, so the fact that the command will now issue an error instead of silently masking again doesn't really matter.
20:47:38 *** Hazzard has joined #openttd
20:48:15 *** Hirundo has joined #openttd
20:49:16 *** planetmaker has joined #openttd
20:49:28 <peter1138> `if (!ctlflags.IsValid()) return CMD_ERROR;`
20:49:37 *** planetmaker is now known as Guest8289
20:49:54 <peter1138> That's... not quite how I thought it would mask :-)
20:50:31 *** ^Spike^ has joined #openttd
20:51:06 *** ^Spike^ is now known as Guest8291
20:51:16 *** V453000 has joined #openttd
20:51:16 *** Terkhen has joined #openttd
20:51:16 *** ChanServ sets mode: +o Terkhen
20:51:46 *** XeryusTC has joined #openttd
20:52:16 *** fonsinchen has joined #openttd
20:54:16 *** Artea has quit IRC (Remote host closed the connection)
20:54:16 *** FLHerne has quit IRC (Quit: There's a real world out here!)
20:54:19 *** FLHerne has joined #openttd
20:54:30 *** ialokin has joined #openttd
20:56:50 <peter1138> I hear M4s are quick, maybe I should get one.
21:01:24 <andythenorth> who told you that?
21:09:49 <peter1138> Has the world ended yet?
21:16:42 <frosch123> Damn, my classmate from school is hitting all news sites with their rejection of rust stuff in the kernel
21:17:46 <frosch123> I assumed he peeked fame with sueing vmware
21:24:29 <andythenorth> peter1138: does it end faster on M4?
21:24:55 <peter1138> Let's see how HasCargoClass is used.
21:25:27 <andythenorth> I think I read that last year, in the CargoClass pitchfork event
21:27:02 <peter1138> Basically ScriptCargo::HasCargoClass makes no mention of CargoClass being a mask.
21:27:34 <peter1138> Hmm, probably best not to subtly change the API just for lols.
21:28:49 <LordAro> frosch123: i kinda empathise
21:29:18 <peter1138> We're clearly of the same frame of mind, we've not accepted any Rust integration.
21:29:45 <LordAro> don't think anyone has asked to either :p
21:30:49 <peter1138> Probably too afriad.
21:31:50 *** Wormnest has quit IRC (Ping timeout: 480 seconds)
21:39:00 <andythenorth> we have a rust grf compiler 😛
21:39:33 <peter1138> Everytime I see Rust code it looks nothing like code, just... syntax.
21:40:27 <peter1138> I guess I've not actually gone out of my way to look at normal Rust :p
21:44:53 <andythenorth> I assume the ; characters are optional 😛
21:48:34 <andythenorth> I should refactor my refactor
21:49:51 <peter1138> You always have to define what a vector contains.
21:50:55 <frosch123> Which has a bad reputation
21:51:56 <peter1138> So a pointer is a Box.
21:52:25 <andythenorth> maybe I should refactor Horse to rust
21:53:03 <peter1138> Yeah, some form of smart pointer. This rust-by-example page doesn't give much in the way of examples for them.
21:53:14 <frosch123> Anyway, I like rust a lot. Though I was surprised it made it into Linux instead of zig
21:54:41 <peter1138> Hmm, then there's Rc.
21:59:45 <frosch123> Recounting like shared_ptr, but single-threaded
22:00:53 <frosch123> Just don't look at macros. They are scary
22:04:47 <andythenorth> _immediately contemplates nml generator using macros and println
22:09:58 <peter1138> That's basically how it already is.
22:13:29 <gwyd4016> Why generate nml when you can generate nfo
22:15:19 <peter1138> Why generate nfo when you generate binary data.
22:19:58 *** nielsm has quit IRC (Ping timeout: 480 seconds)
22:23:46 *** keikoz has quit IRC (Ping timeout: 480 seconds)
22:26:05 <peter1138> Crap, we left it too long.
22:27:33 <xarick> how many of the [Scripts] going to be merged?
22:28:42 <xarick> I'd like to try unbunch on my ships
22:42:35 <gwyd4016> LC-ZorgviaGitHub: Flat earth mentioned
22:46:28 <LordAro> definitely got a point about other water types
22:50:17 <talltyler> It’s been a while because I haven’t finished it 🙂
22:50:22 <talltyler> Not lack of reviews
22:53:23 <peter1138> Ah, that's why Firefox was stalling.
22:53:44 <peter1138> I left a browser tab open in developer mode, and the page kept loading something in the background. After a few hours the persistent logs get a bit much for it.
22:54:35 <talltyler> (No changes, just a rebase and fixing annotations)
22:55:07 <peter1138> The bit that's weird about that PR is we've gone from "all maps have water around the edge" to "you can optionally not have water" back to "all maps should have water around the edge" :)
22:55:56 <talltyler> Time is ~~fake~~ a loop, etc
22:56:00 <peter1138> How about we generate infinite procedural terrain but keep the map size limit ;-)
23:00:24 *** HerzogDeXtEr has quit IRC (Read error: Connection reset by peer)
23:14:05 <peter1138> > Feeder stations to increase % transported; do real life companies do this?
continue to next day ⏵
