IRC logs for #openttd on OFTC at 2025-11-19
⏴ go to previous day
00:00:43 *** jinks has quit IRC (Remote host closed the connection)
00:06:25 <reldred> eheheh, you can definitely get some giant swamps on the right heightmap with the right settings 😄
00:27:12 *** toktik is now known as Guest31541
00:29:56 *** Guest31541 has quit IRC (Remote host closed the connection)
01:38:07 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
01:38:09 *** WormnestAndroid has joined #openttd
02:21:22 <acs121> reldred: These look awesome
02:51:20 *** Wormnest has quit IRC (Quit: Leaving)
03:49:30 *** gnu_jj_ has joined #openttd
03:53:14 *** gnu_jj has quit IRC (Ping timeout: 480 seconds)
04:18:08 *** Zathras_11 has joined #openttd
04:21:29 *** Zathras has quit IRC (Ping timeout: 480 seconds)
05:25:59 *** tokai|noir has joined #openttd
05:25:59 *** ChanServ sets mode: +v tokai|noir
05:32:44 *** tokai has quit IRC (Ping timeout: 480 seconds)
06:41:33 *** ChanServ sets mode: +v tokai
06:48:15 *** tokai|noir has quit IRC (Ping timeout: 480 seconds)
08:52:02 <mmtunligit> is there a way to see a crash log type rundown of what the game is doing when its not crashing? im trying to figure stuff out and i think it could be helpful if i could see whats happening
08:59:48 <LordAro> i feel like there's a key combination in debug mode that triggers it
08:59:56 <LordAro> or are you describing gdb?
09:05:02 <peter1138> mmtunligit, pause/breakpoint in your debugger, and you can see the stack trace. Using a debug build is slower but gives a better result.
09:05:22 <peter1138> LordAro, why is it snowing?
09:06:01 <peter1138> I added a better description to #14801. because "See above." is kinda brief.
09:07:25 *** johnfranklin has joined #openttd
09:10:52 <johnfranklin> is dibridge down?
09:23:24 <peter1138> 7 * 9 02 04 FF 01 01 00 00 00 00
09:23:36 <peter1138> Spriteset with no sprites. Mad scientists.
09:27:31 <reldred> all I had for lunch was a bag of twisties and a bottle of iced coffee at around half past three
09:27:52 <reldred> I'm not sure if I'm actually hungry or not and its 7:30 now
09:29:45 <reldred> talltyler: here's an example of what my jgrpp games look like. zero percent rainforest coverage, desert rocks, big fat tropic band around water sources
09:30:32 <peter1138> I thought that road was a trail of rocks.
09:31:01 <reldred> yeah it is a bit light
09:34:21 <reldred> now I was gonna have another crack this evening at porting swamps to JGRPP and adding my usual nonsense additional settings to it but I half feel like just having a lay down and a snooze
10:04:56 *** johnfranklin has quit IRC (Remote host closed the connection)
10:05:19 *** johnfranklin has joined #openttd
10:07:43 <mmtunligit> could I get some help setting up the debugger? all the documentation i can find seems to assume im building in a way that im not. my IDE is Kate and it comes with a debugger adapter protocol. i have my executable set as {PATH}/openttd/build/openttd and the working directory as {PATH}/openttd/ but every time i hit start debugging i get the error "DAP backend: Child process set up failed: execve: No
10:07:43 <mmtunligit> such file or directory"
10:08:46 <LordAro> sounds like you don't acutally have gdb (or whatever kate is trying to execute) installed
10:11:16 <mmtunligit> running sudo apt-get install gdb says i already have the latest version
10:14:15 <LordAro> i could be misreading the error message - are you *very* certain it's the correct path to openttd executable (is {PATH} something you've done, or part of the config?), and does it work if you run it manually? (paste it into a terminal to be sure)
10:14:42 <LordAro> (and does your path contain any special characters like `~`)
10:16:39 <mmtunligit> path is the correct path to that stuff, /home/{user}/ottdsrc and it works in terminal
10:22:20 <LordAro> on the other hand, maybe you're not supposed to link to openttd there at all
10:22:57 <LordAro> might be worth copying bits of their dotfiles (gdb bit, most relevant)
10:23:05 <LordAro> but otherwise, i suspect you're on your own
10:23:57 <mmtunligit> oks o also the same site ive found, great
10:26:21 <kuhnovic> talltyler: About your hole-filling endevours (ghehehe) : rather than ignoring small map size, wouldn't it make sense to apply some formula like "a sea of X tiles can accomodate a river length of <insert math here> tiles"? Just an idea that popped up in my head.
10:27:24 *** johnfranklin has quit IRC (Ping timeout: 480 seconds)
10:28:02 <mmtunligit> it was defaulting to some nonstandard implementation of gdb for whatever reason
10:31:01 *** johnfranklin has joined #openttd
10:40:19 <talltyler> kuhnovic: I considered it, but you could easily have a case where a small river drains into a small sea, then a huge wide river comes along, but has no idea there’s already a river using that sea. So you fill it and leave the small river disconnected (or worse, the big river finds the small river and terminates there!)
10:41:02 <reldred> Pluugging small oceans also has a positive effect on grfs like FIRS too
10:41:32 <kuhnovic> Hmm yeah, there's always these annoying corner cases
10:42:49 <reldred> but it also reveals that we need more swamps (and lakes), not just where rivers end, maybe chance to spawn them wherever a river finds another river?
10:46:36 <andythenorth> build the lakes, run the rivers backwards
10:46:55 <andythenorth> although I think Xarick found that it's impossible to build rivers backwards
10:47:04 <kuhnovic> I don't believe that
10:47:26 <andythenorth> something about rivers can't flow uphill I think
10:48:18 <kuhnovic> If you take a neighboring tile and check if that flows down a hill towards the original tile, you're still using the same checks 😛
10:48:45 <kuhnovic> I had a look at it a while ago. It's definitely not a trivial thing to do but certainly not impossible..
10:51:38 <andythenorth> maybe we need river regions 😛
10:51:44 <andythenorth> can't explain why I think that helps
10:52:14 <andythenorth> but if a river tried to pathfind uphill, not using a random walk, but using a vector towards a tile or region
10:52:22 *** johnfranklin has quit IRC (Remote host closed the connection)
10:52:29 <andythenorth> and then stopping at some unreachable retry threshold
10:52:40 <andythenorth> and not building the river if the path is lolz short
10:53:25 <andythenorth> we could then make that turtles
10:53:43 <andythenorth> so the river pathfinds more randomly through a tile region, towards an exit
10:53:55 <andythenorth> then selects a new adjacent target region
10:54:03 <reldred> I dunno, I think rivers are working pretty well now in talltyler's most recent hole filling PR, just needs a tweak to reintroduce lakes and wetlands since plugging bullshit holes of ocean tends to prevent his gorgeous new wetlands from spawning.
10:54:18 <reldred> and then tropic needs some love
10:54:37 <reldred> but I think it should be tgrivial to introduce some of the changes I made in jgrpp.
10:54:41 <andythenorth> possibly I've got too much Carcassone in mind
10:55:46 <talltyler> If anyone redoes the river algorithm from scratch, they might as well do Rainfall Lite. There’s got to be a way to approximate the results for cheaper.
10:56:02 <reldred> Rainfall was great but yeah it cooked CPU's
10:56:12 <reldred> but that was also like 10 years ago?
10:56:57 <andythenorth> Rainfall seems logical, because erosion
10:57:21 <andythenorth> but the world isn't made up of rhombus tiles that can have one or more corners raised
10:57:29 <andythenorth> unless I missed it
11:02:46 <andythenorth> RRG had a PDF manual 🙂 Fancy
11:04:56 <reldred> I wouldn't be surprised if they were working on some school/uni project tbh, the amount of documentation certainly tracks for it.
11:05:13 <andythenorth> some of the results look nice
11:05:24 <reldred> It was sublime back in the day
11:05:31 <andythenorth> I am kind of -1 to trying to generate the whole landscape as close-to-realism
11:05:39 <reldred> I used to generate maps in Rainfall then load the saves in JGRPP
11:05:41 <andythenorth> I'd rather we had the option to generate game boards
11:06:52 <reldred> I'm not exactly a model railroad style of player
11:07:07 <andythenorth> just 2 board cards, and then move one of them to continue the game
11:07:10 <andythenorth> when the road runs out
11:07:15 <reldred> but I also dislike the boardgame style of play
11:08:29 <andythenorth> and obviously the game has an excellent name
11:10:07 <talltyler> What does a board game style map in OpenTTD look like, compared to what we have now?
11:10:44 <reldred> sleepy time, someone merge 14797 while i'm asleep plzkthnx
11:12:15 <andythenorth> talltyler: I don't know, I'm taking this line as an antidote to 'realism'
11:12:28 <andythenorth> I've played realistic heightmaps, they're not wildly satisfying to me
11:12:41 <andythenorth> meanwhile, no idea why this is relevant
11:14:36 <andythenorth> I think my point is something like
11:14:36 <andythenorth> * decomposing the map into regions is more useful
11:14:36 <andythenorth> * maybe they're irregular polygons
11:14:36 <andythenorth> * Zephyris had some concepts
11:14:36 <andythenorth> * within a region, we could adjust the terrain generation for flavour (or pre-compose specific behaviours from heightmaps or scripts)
11:15:20 *** emperorjake has joined #openttd
11:15:20 <emperorjake> so, kind of like Minecraft style biomes?
11:15:29 <andythenorth> I watched some minecraft videos
11:15:43 <reldred> andythenorth: That would get us towards recreating TTD’s original alpine and tropic landscape gen
11:15:52 <andythenorth> and the waveform collapse stuff from the bad north author and others, which tends towards something like voroni partitions I think
11:16:06 <andythenorth> currently the TGP approach is to try and scale some waveforms across the whole map, afaict, using basic perlin
11:16:07 <reldred> Came up in conversation earlier today or yesterday actually
11:16:15 <kaji_kaede> andythenorth: ...Was actually thinking of working on something like this down the line. Would be nice if more competent people did it before me.
11:16:27 <emperorjake> I liked the half desert/snow maps from TTD, but they got a bit silly with bigger map sizes
11:17:07 <reldred> emperorjake: Yeah was fine at 256x256 but with larger maps the regions need to be smaller and more numerous than simply cutting the map in half
11:17:34 <reldred> Mountainous region, swampy region, hilly region, flat regions, yadda yadda
11:17:52 <andythenorth> there are a countable number of terrain types
11:17:59 <emperorjake> speaking of mapgen, I played with SimCity 3000's recently and it's actually really good, the sliders work really well
11:18:18 <andythenorth> is it tile based, or polygons?
11:30:14 <locosage> talltyler: I've linked an article on fast terragen here many times already, it does look a lot like rainfall
11:30:42 <locosage> citymania server #8 uses it, though I've never quite finished it so rivers are a bit off
11:57:42 <xarick> i just found out regression_stationlist actually calls SwapList
12:09:52 <peter1138> 10:46 < andythenorth> although I think Xarick found that it's impossible to build rivers backwards
12:11:30 <peter1138> andythenorth, pretty sure I had that Hot Wheels thing.
12:19:55 <xarick> i somehow ended in multimap.hpp file
12:20:11 <xarick> a whole new world to discover
13:21:19 <certator> hi there, what do you think about possible feature where server website can be opened with some command? For example, player types !web command to chat, and OpenTTD client opens server URL with default browser. To reduce fishing risks, list of allowed URLs (or URL prefixes) can be served somewhere on https://www.openttd.org/, or openttd.github.io, and domain owner can be periodically verified
13:21:19 <certator> (maybe with github actions).
13:26:54 <peter1138> Typing commands into chat -- no.
13:27:40 <peter1138> Having a clickable URL appear in the server section of the Online Players window -- more reasonable.
13:29:55 <locosage> clickable url in the chat window would be massively more useful
13:30:00 <locosage> or, at least, copyable
13:32:18 <locosage> also, in all script strings like story window, popups, etc
13:34:40 <talltyler> Good point, Kuhnovic 🙂
13:45:55 <andythenorth> peter1138: my brother had it, but the sticker for the mud was on backwards 🙁 Bothered me for years 😛
13:47:33 <mmtunligit> that is the sam eimage twice, thank you clipboard
13:49:33 <talltyler> (who added it to the milestone)
13:53:27 <LordAro> doesn't mean no one else can look at it
13:55:56 <LordAro> i don't even own anything with a joystick that i could test it with :p
13:56:59 <talltyler> Oh, I thought you were volunteering. Sorry. 🙂
13:57:09 <talltyler> I can test on Windows.
13:57:14 <andythenorth> I would need to build it and try it with XBox controllers
14:13:47 *** Flygon has quit IRC (Remote host closed the connection)
14:34:41 <_glx_> I know it's easier for authors to be able to use the labels directly without using `[cargo|rail|road|tram]type()`, but allowing that was a bad idea
14:52:58 <Borg> okey. so Check Signal state is even more usefull :) u can nicely emulate pre-signals with it..
14:53:38 <Borg> had to make extra platform.. but penalized.. but also due to extra signal before platform.. train could go to red now.. fixed :)
14:56:49 <locosage> hah, was thinking on label syntax in grf-py recently
14:57:02 <locosage> guess I shouldn't allow bare labels 😅
15:00:20 <xarick> why use SpiralTileSequence?
15:00:58 <xarick> TileArea is kinda faster
15:04:08 <xarick> also you can break the loop by checking `if (counter == VALID_TILE_GOAL) break;`
15:06:28 *** tabytac has quit IRC (Remote host closed the connection)
15:06:28 *** acs121 has quit IRC (Remote host closed the connection)
15:06:28 *** DorpsGek_vi has quit IRC (Remote host closed the connection)
15:06:28 *** jack_88271 has quit IRC (Remote host closed the connection)
15:06:28 *** _glx_ has quit IRC (Remote host closed the connection)
15:06:28 *** andythenorth has quit IRC (Remote host closed the connection)
15:06:28 *** mmtunligit has quit IRC (Remote host closed the connection)
15:06:28 *** kuhnovic has quit IRC (Remote host closed the connection)
15:06:28 *** vondpc has quit IRC (Remote host closed the connection)
15:06:28 *** goddess_ishtar has quit IRC (Remote host closed the connection)
15:06:28 *** jfkuayue has quit IRC (Remote host closed the connection)
15:06:28 *** xarick has quit IRC (Remote host closed the connection)
15:06:28 *** kaibaneddy has quit IRC (Remote host closed the connection)
15:06:28 *** efessel has quit IRC (Remote host closed the connection)
15:06:28 *** audigex has quit IRC (Remote host closed the connection)
15:06:28 *** zanooda2000 has quit IRC (Remote host closed the connection)
15:06:28 *** locosage has quit IRC (Remote host closed the connection)
15:06:28 *** _zephyris has quit IRC (Remote host closed the connection)
15:06:28 *** olionkey has quit IRC (Remote host closed the connection)
15:06:28 *** emperorjake has quit IRC (Remote host closed the connection)
15:06:28 *** certator has quit IRC (Remote host closed the connection)
15:06:28 *** talltyler has quit IRC (Remote host closed the connection)
15:06:28 *** gchromodynamics has quit IRC (Remote host closed the connection)
15:06:28 *** bigyihsuan has quit IRC (Remote host closed the connection)
15:06:28 *** yiffgirl has quit IRC (Remote host closed the connection)
15:06:28 *** wensimehrp has quit IRC (Remote host closed the connection)
15:06:28 *** muxydugoulp has quit IRC (Remote host closed the connection)
15:06:28 *** voss749 has quit IRC (Remote host closed the connection)
15:06:28 *** michi_cc[d] has quit IRC (Remote host closed the connection)
15:06:28 *** belajalilija has quit IRC (Remote host closed the connection)
15:06:28 *** sosmakaroni has quit IRC (Remote host closed the connection)
15:06:28 *** merni has quit IRC (Remote host closed the connection)
15:06:28 *** rito12_51026 has quit IRC (Remote host closed the connection)
15:06:28 *** reldred has quit IRC (Remote host closed the connection)
15:06:28 *** _jgr_ has quit IRC (Remote host closed the connection)
15:06:28 *** robinsonmanane has quit IRC (Remote host closed the connection)
15:06:28 *** kaji_kaede has quit IRC (Remote host closed the connection)
15:06:49 *** DorpsGek_vi has joined #openttd
15:09:39 <xarick> I'm finally succeeding
15:10:32 <peter1138> Hmm, people using non-integer interface scaling with pixel font...
15:11:21 <xarick> how trying to do the same for the original openttd code...
15:11:21 <peter1138> mmtunligit, looking promising, but I hope that extra list isn't visible all the time, cos it's using a lot of space :)
15:12:33 *** talltyler has joined #openttd
15:12:33 <talltyler> Early break added, town generation is fast enough that I'm not interested in measuring SpiralTileSequence vs TileArea. I doubt there's an appreciable difference.
15:12:36 <peter1138> talltyler, will someone™ come along and make up some test that proves that generating 4096x4096 maps with a 65535 towns is now slow?
15:12:57 <talltyler> I ignore those tests as a general rule.
15:19:07 <Borg> cudos for _jgr_ for routing restrictions :>
15:19:56 <Rubidium> it might actually be faster, if town generation is now stopped more often due to location instead of having to go through a few hundred attempts to create a town name when the number of town name's relatively low
15:35:12 <peter1138> 44 towns in master, 27 towns with the PR.
15:40:34 <xarick> the buckets system is so cumbersome
15:43:50 <talltyler> I guess checking for flat tiles is not a strict requirement, but it wouldn't help towns in the mountains anymore.
15:44:33 <peter1138> Towns in mountains are basically guaranteed to be on slopes.
15:45:28 <peter1138> I had a similar idea a while back, my patch checked for elevation deviation, so it still allowed towns where the area is "approximately flat"
15:46:49 <peter1138> There's also a conflict where there is code that prefers to place towns near shore... which this PR then rejects.
15:47:28 <peter1138> (Might even be "near water" so rivers as well, but would need to check)
15:49:09 <xarick> thoughts on 14753? merge?
15:52:37 <xarick> I need it merged to be able to validate iterators
15:52:54 <peter1138> talltyler, funny thing is, until I tested it, I was ready to yolo that change :o
15:56:51 <talltyler> Elevation deviation might be a good idea. A deviation > 1 fails that tile.
16:03:05 <peter1138> I also did something that just terraformed around the town, a bit like industries, but that is also quite drastic.
16:11:34 <peter1138> Hmm, do I want a 14.3" tablet...
16:29:22 <peter1138> LordAro, if "04:00:00" parsed as a TimeSpan is 4 hours, what does "24:00:00" give?
16:38:27 *** Wormnest has joined #openttd
17:02:09 <xarick> if this->bucket_list is nullptr, how come this->bucket_list_iter points to end?
17:02:28 <Rubidium> obviously 24 days, 0 hours and 0 minutes? :D
17:10:41 <peter1138> Rubidium, actually correct, yes.
17:11:18 <peter1138> 23:00:00 would be 23 hours, but as 24 is longer than a day, it instead parses it as a decimal days.hours value.
17:11:32 <peter1138> So for 24 hours, it actually wants 1 day... "1.00:00:00"
17:11:54 <talltyler> peter1138: I've adopted your elevation deviation idea, and also moved this check into the "build near water" search loop, to avoid it failing unnecessarily.
17:25:16 *** audigex has joined #openttd
17:25:16 <audigex> Peter, further to our conversation about some kind of NO_PURCHASE_TEXT type of flag for badges... if you make that change, could we also have a NO_PURCHASE_SPRITE option please? Sounds counter intuitive I know, why add a sprite if it's not going to be shown
17:25:44 <audigex> But actually I just realised it would be nice to be able to have sprites on the filter even if I don't want to display the sprite next to the purchase sprite
17:26:12 <audigex> (This isn't the best example since I *would* want the operator sprite shown... but just to demonstrate)
17:36:02 <peter1138> Eh, that doesn't make much sense. Players can filter on the badge even though the badge column is not displayed.
17:36:55 <peter1138> Showing the sprite or not is a player choice.
17:40:22 <audigex> I’m thinking more of a “default off, player can enable” option
17:44:18 *** gelignite has joined #openttd
17:45:29 <audigex> So yeah maybe more “SPRITE_DEFAULT_HIDDEN” than “NO_PURCHASE_SPRITE”
17:45:29 <audigex> It would give newGRF devs more freedom to add badges with sprites for higher level of detail without overwhelming users who don’t want sprites for excessive realism like route availability or types of diesel that don’t actually affect gameplay (random examples off the top of my head). They’d be available but hidden by default
17:47:33 *** tabytac has joined #openttd
17:47:33 <tabytac> that sounds quite reasonable to me
18:00:11 *** andythenorth has joined #openttd
18:00:11 <andythenorth> peter1138: does Doom run on it?
18:09:27 * Borg wonders why he hears about DooM here so often ;)
18:17:35 <andythenorth> it has a lot of railtypes
18:17:41 <andythenorth> so it's of great interest
18:34:10 <andythenorth> seems a lot longer
18:35:18 <andythenorth> maybe I should get the small sea PR too
18:36:30 <andythenorth> I do wonder if we should report it as marsh
18:37:16 <talltyler> We could, but scope creep 🙂
18:38:40 <talltyler> Yeah, if it were a new water class we could do a new sprite, maybe even have trees in it
18:39:55 <andythenorth> badge the whole terrain
18:40:01 <andythenorth> do we have the RAM? 😛
18:40:40 <andythenorth> "applying badges made my 4096x4096 map slower"
18:42:53 <andythenorth> I want to badge industry tiles, although why, and for what? 😛
19:21:13 *** Zathras_4 has joined #openttd
19:24:39 *** mmtunligit has joined #openttd
19:24:39 <mmtunligit> is there a way to set up a watchdog on a debugger to interrupt the program when a specific symbol's value gets changed? "saved" seems to get intialized somewhere between clicking on a button in the pickerwindow and the OnClick method gettign called and i have no idea why or where id go to find that. also useful would be a way to step back through the program instead of just forward
19:24:59 <mmtunligit> or just look backward in general
19:30:32 *** michi_cc[d] has joined #openttd
19:30:32 <michi_cc[d]> "A" debugger can do that, yes 🙂 Not familiar with gbd though, so I don't know how it is called there.
19:30:41 <michi_cc[d]> Ah, see, too slow... 🤣
19:31:01 <michi_cc[d]> But I was suggesting to look for either watchpoint or data breakpoint
19:31:12 <LordAro> i find gdb watchpoints extremely finicky
19:31:16 <LordAro> never seem to work when i want them to
19:33:00 <michi_cc[d]> The other way would be to use grep/your IDE to find all lines writing to the variable in question and just putting breakpoints on all of them.
19:33:44 <mmtunligit> thats the thing, i could swear ive already done that, though im guessing theres one somewhere i dont know about
19:34:46 <mmtunligit> also i do not know how to use grep
19:35:01 <mmtunligit> im aware it can get very complicated very quickly
19:55:39 *** Borg has quit IRC (Quit: leaving)
20:03:16 <mmtunligit> why is ther so little documentation on this
20:04:04 <mmtunligit> if you have anything more than a simple one file program there is just no help to figure out what to put in the expression
20:06:29 <_jgr_> What are you trying to do exactly?
20:08:24 <mmtunligit> set a watchpoint for the "saved" varible declared in picker_gui.h ive made it a map instead of a set for the thing im working on and it gets initalized somewhere weird that i cant find
20:09:17 <_jgr_> a std::map is not a single address like an int or whatever, most likely you're not going to be able to break on it using a watchpoint
20:10:12 <mmtunligit> then i guess the first element of the map, which is a string
20:11:51 <_jgr_> You may be able to catch the address of the root node changing, as that'll be somewhere in the top-level struct
20:12:35 <_jgr_> However you'll likely have much easier time just inserting a call to Debug at all the code sites which modify the saved map
20:13:12 <mmtunligit> i could swear ive checked them all is the thing
20:13:21 <mmtunligit> ill try it again though
20:13:36 <_jgr_> What exactly do you mean by "gets initialised" anyway?
20:17:21 <mmtunligit> so when i first open the pickerwindow in a new save it is a map with zero elements, and it *should* stay that way until i add a group, at which point it should be a map with 1 element. however, pressing any button in the pickerwindow seems to add an element to it, but only the first time, after that it works as expected
20:18:48 <audigex> mmtunligit: It's pretty easy if you aren't doing fancy stuff with it
20:18:48 <audigex> grep -in "text" filename
20:18:48 <audigex> grep -in "text" folder/
20:18:48 <audigex> I always whack -in flags because it's easy to remember as "find in" and they're useful flags for case-insensitive and line numbers, so may as well just use them every time
20:18:48 <audigex> add -r for recursive in folders, -E for regex instead of just text (and I have to look that one up every time, but how often are you using regex?)
20:19:19 <audigex> You can get 90% of what you need done with either
20:19:19 <audigex> grep -in "text" filename
20:19:19 <audigex> grep -rin "text" folder/
20:19:25 * andythenorth in alternate timeline: all vehicle grfs included rail / road / tram types for all their infra, but could mark them as "only if no other grf defines this type"
20:19:34 <andythenorth> which is sort of what Horse now does
20:19:49 <andythenorth> in a way that involves red boxes telling the user they did bad thing
20:19:57 <_jgr_> mmtunligit: And this is with the code for PR 14744?
20:20:09 <mmtunligit> no i dont have a PR for this yet
20:20:35 <mmtunligit> it is getting set with whatever sel_group is set to which is good info
20:20:48 <mmtunligit> which is a new variable ive added
20:23:03 <_jgr_> Might be worth pushing up a branch with this in to your github?
20:24:03 *** squirejames has joined #openttd
20:25:08 <andythenorth> yes one of those
20:25:11 <andythenorth> Horse has that now
20:25:14 <squirejames> Implement it and watch Reddit go incandescent with rage
20:25:32 <andythenorth> Reddit don't seem to use JP+ Tracks.....yet
20:25:39 <andythenorth> too busy with zbase and CZTR
20:33:57 <_jgr_> I see various `saved[callbacks.sel_group]`, these will insert `callbacks.sel_group` into `saved`, in case this is what might be tripping you up
20:34:22 <mmtunligit> i thought id got all of those but i guess not, thank you
20:37:15 <mmtunligit> id put all those in when i was trying to figure how to get the right datatype at all, i had two options and having a function in between dots seemed like it wouldnt work for whatever reason
20:39:10 <_jgr_> mmtunligit: It's fine for the cases where you want to make it on demand, if you just want to get it if it exists, I'd use `.find()`
20:41:03 <mmtunligit> no i need to use .at() for all of those places
20:41:36 <xarick> i just noticed @param target but the param is called new_list
20:42:30 <_jgr_> mmtunligit: This will throw if it doesn't exist
20:42:42 <mmtunligit> yeah i need to deal with that somehow
20:43:40 <mmtunligit> maybe i do need find? thatss give me the end iterator though wich id need to deal with
20:44:21 <_jgr_> Yes, that's the usual way
20:46:58 <mmtunligit> thank you very much for the help, hopefully ill be able to sort this bit out tomorrow
20:50:39 *** johnfranklin has joined #openttd
20:50:47 *** johnfranklin has quit IRC ()
21:15:57 <xarick> visual studio 18.0.1 released
21:49:09 <xarick> how do I add the Release there?
22:45:29 *** keikoz has quit IRC (Ping timeout: 480 seconds)
23:18:16 <xarick> PR#14772 should pass now
23:20:39 <xarick> I'm off to bed... will know the result tomorrow
23:25:36 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
23:57:59 *** greeter has quit IRC (resistance.oftc.net charm.oftc.net)
23:57:59 *** felix has quit IRC (resistance.oftc.net charm.oftc.net)
23:57:59 *** Extrems has quit IRC (resistance.oftc.net charm.oftc.net)
23:57:59 *** dwfreed has quit IRC (resistance.oftc.net charm.oftc.net)
23:58:05 *** Extrems has joined #openttd
23:59:58 *** dwfreed has joined #openttd
continue to next day ⏵
