⏴ go to previous day
00:06:30 *** HerzogDeXtEr has quit IRC (Read error: Connection reset by peer)
00:38:23 *** Flygon has quit IRC (Read error: Connection reset by peer)
02:44:59 *** gelignite is now known as Guest7015
02:45:02 *** gelignite has joined #openttd
02:52:19 *** Guest7015 has quit IRC (Ping timeout: 480 seconds)
03:11:16 *** D-HUND has joined #openttd
03:14:40 *** debdog has quit IRC (Ping timeout: 480 seconds)
03:43:47 *** fygxcfgvhb has joined #openttd
03:45:24 *** fygxcfgvhb has quit IRC (Remote host closed the connection)
04:42:00 <DorpsGek> [OpenTTD/OpenTTD] eints-sync[bot] pushed 1 commits to master https://github.com/OpenTTD/OpenTTD/commit/492a5284ea5b69a7f639b8105caa1042cb1e87e3
04:42:01 <DorpsGek> - Update: Translations from eints (by translators)
04:59:09 *** gelignite has quit IRC (Quit: Stay safe!)
05:11:49 *** keikoz has joined #openttd
08:03:11 *** Flygon has joined #openttd
08:19:07 *** nielsm has joined #openttd
08:38:28 *** Wolf01 has joined #openttd
09:09:11 <peter1138> Well.
09:14:35 <peter1138> > 97 files changed, 1057 insertions(+), 1056 deletions(-)
09:14:43 <andythenorth> probably
09:15:22 <peter1138> Hmm, badge dropdown filter lists: multiple select or single?
09:16:15 <andythenorth> do we have a multiselect?
09:17:24 <peter1138> Yes.
09:18:16 <andythenorth> probably that then
09:18:38 <peter1138> Station list cargo filter does it.
09:18:43 <peter1138> Also it appears to be broken.
09:20:55 <peter1138> Lack of StrongTypes :(
09:22:09 <peter1138> I'm trying a new NewGRF.
09:22:36 <andythenorth> StronkTypes
09:23:56 <xarick> hi
09:24:10 <peter1138> Oh, perhaps this one has no badges.
09:25:15 <peter1138> I see. My design is flawed :D
09:27:22 <peter1138> Or, I suppose, D:
09:30:33 <DorpsGek> [OpenTTD/OpenTTD] PeterN opened pull request #13382: Fix 4c8f1b0f81: First entry of station cargo filter list was broken. https://github.com/OpenTTD/OpenTTD/pull/13382
09:31:30 <peter1138> Right, I can't be lazy and use the GUIBadgeClass list for the badge filter, because that only lists badge classes with icons, and we should be able to filter by badges with only names.
09:35:28 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 approved pull request #13382: Fix 4c8f1b0f81: First entry of station cargo filter list was broken. https://github.com/OpenTTD/OpenTTD/pull/13382#pullrequestreview-2574183919
10:09:21 <DorpsGek> [OpenTTD/OpenTTD] PeterN merged pull request #13382: Fix 4c8f1b0f81: First entry of station cargo filter list was broken. https://github.com/OpenTTD/OpenTTD/pull/13382
10:18:20 <DorpsGek> [OpenTTD/OpenTTD] PeterN updated pull request #13381: Codechange: Introduce and use EnumBitSet. https://github.com/OpenTTD/OpenTTD/pull/13381
10:22:19 <peter1138> Hmm, I guess for badges the extra filtering options can be added later.
10:22:30 <peter1138> But I should really sort out the configuration stuff. :S
10:22:44 <peter1138> Like, why is it only configurable for building vehicles?
10:23:47 <peter1138> If I removed the drop down list and added a separate configuration window, then it could have a feature-selector.
10:24:24 <peter1138> And it could also be used later to configure what filters players wnat.
10:39:43 <peter1138> What icon does a battery vehicle get?
10:40:02 <peter1138> Wrong chat again, hah.
10:40:57 <peter1138> Do different vehicle types need different icons?
10:41:00 <Rubidium> oh darn... not going with enums for IDs (first) will mean big-bang conversions need to be done due to unions. Anything in a union can't be strong-type, so you need to go with std::variant. But that can't work with duplicate types (like SignID and TownID are both uint16_t). So anything that is connected to eachother in enums needs to go in at once which might be a nuisance
10:41:01 <peter1138> (That's doable in the spec)
10:42:19 <Rubidium> ... it also means the conversion to std::variant of those unions gets wrapped up into that massive commit
10:42:52 <peter1138> Going with enums seems to be less work, right?
10:43:45 <peter1138> "Technically UB" though?
10:47:44 <peter1138> (That has (almost) never been a problem, of course)
10:48:14 <Rubidium> depends on how you read/interpret CWG 1766: "A value of integral or enumeration type can be explicitly converted to an enumeration type. The value is unchanged if the original value is within the range of the enumeration values (9.7.1 [dcl.enum]). Otherwise, the behavior is undefined." So, if range is from the minimum to the maximum value and anything in between is also considered in range, then it
10:48:20 <Rubidium> wouldn't be UB. If range implies the specified values, then it would be UB.
10:50:36 <Rubidium> the GCC 4.5 era issue is due to the former interpretation, where it optimised out '<= ENUM_END' checks (assuming ENUM_END was the last value). If we define ENUM_BEGIN, ENUM_END and ENUM_INVALID and check '< ENUM_END' we wouldn't even trigger the GCC 4.5 issue
10:52:15 <Rubidium> and if it is the second interpreation, then we have been UB for a very long time already; case and point: Company/Owner (and all the bitmasks)
10:52:56 *** HerzogDeXtEr has joined #openttd
10:52:57 <Rubidium> note that the enum bitmask thing would be UB in both of my interpretations of CWG 1766.
10:58:27 <peter1138> Yeah, I dunno either.
10:59:48 <Rubidium> IMHO we use enums as a stop gap measure to aid the conversion to StrongTypeDefs for IDs
11:04:39 <peter1138> That is also what I was thinking.
11:06:25 <xarick> UB?
11:08:51 <xarick> \
11:09:09 <xarick> ah, undefined behaviour?
11:20:36 <xarick> I'm in a deadlock
11:20:56 <xarick> poor tile mutexes
11:22:07 <Rubidium> that's a good sign!
11:23:19 <Rubidium> now you can finally figure out the interdependencies of those threads messing with things
11:38:33 <xarick> std::shared_lock read_lock_tile(TileMutex(tile));
11:38:41 <xarick> yep, I dont understand locks
11:41:17 <_zephyris> On a random now, how are heightmaps interpreted? Vertex heights, with the constraint that height difference between manhattan neighbours must be 0 or 1?
11:43:33 <_zephyris> Thinking about a heightmap pre-processor which can accurately preview how a heightmap will be interpreted
12:02:03 *** debdog has joined #openttd
12:02:24 *** debdog has quit IRC ()
12:10:57 <xarick> https://cdn.discordapp.com/attachments/1008473233844097104/1333046521825853523/image.png?ex=679777d1&is=67962651&hm=5fd8a460331f217f9282b81c79617cd43a4540ebe8d04d1f70a8fcf41d5d44da&
12:10:57 <xarick> what does that even mean? I didn't touch that
12:18:52 *** D-HUND is now known as debdog
12:23:42 <xarick> so these locks work kinda like an AutoRestoreBackup
12:52:51 <_glx_> Basic scope
12:54:01 <michi_cc> Rubidium: The whole enum business is really tricky trying to understand all the standardeese. CWG 1766 respectively https://eel.is/c++draft/expr.static.cast#9 talks about "range of the enumeration values" and refers to https://eel.is/c++draft/dcl.enum#8 for that.
12:54:01 <michi_cc> Now I'm not totally sure if "the values of the enumeration" in that paragraph really means the same (footnote 82 seems to support that), but if it is indeed the case, then gcc 4.5.2 simply had a bug and nothing more. dcl.enum has "For an enumeration whose underlying type is fixed, the values of the enumeration are the values of the underlying type", and the sentence in CWG 1766 is only for enum
12:54:01 <michi_cc> types without a fixed underlying type.
12:54:54 <michi_cc> Given that, one can certainly read the standard in a way that enums with a fixed underlying type are fine because the values of the enumeration are the full range of the underlying type.
13:17:52 <peter1138> I... have a patch for that.
13:36:58 <xarick> does this look alright, anyone? <https://gist.github.com/SamuXarick/b1ea46bea3a4bd6bf08f8b6a7964e11f>
13:37:02 <xarick> the lock usage
13:38:49 <_glx_> seems correct, but it's a pain to do it at this level
13:40:49 <DorpsGek> [OpenTTD/OpenGFX] jayaddison opened issue #94: gimpscript: GIMP3 compatibility problem(s) https://github.com/OpenTTD/OpenGFX/issues/94
13:42:32 <Rubidium> xarick: how much faster is it with that locking?
13:43:01 <xarick> very poor imo
13:43:07 <xarick> > [RemoveRiverArtifacts mt] 25050 us
13:43:07 <xarick> > [RemoveRiverArtifacts st] 33546 us
13:46:38 <Rubidium> so... you've added ~640 MiB of extra memory usage on a 4kx4k map and made making the map 8 milliseconds faster
13:47:13 <xarick> 640 wow
13:47:47 <Rubidium> and only 1280 MiB for MSVC x64
13:48:23 <xarick> didn't pay attention to mem usage
13:49:51 <peter1138> ... did he create a lock per tile ... ?
13:50:55 <xarick> yes
13:51:13 <peter1138> YOLO.
13:51:57 <xarick> i thought that was the smart way to do it
13:52:56 <xarick> where is this memory coming from?
13:53:17 <xarick> https://cdn.discordapp.com/attachments/1008473233844097104/1333072274521194618/image.png?ex=67978fcd&is=67963e4d&hm=6684210bbd31e819ad3104771c587d56536f5e29693da6c221bda3bc9fef27c1&
13:53:56 <Rubidium> Google
13:55:11 <DorpsGek> [OpenTTD/OpenTTD] PeterN opened pull request #13383: Codechange: Specify underlying type for all enums. https://github.com/OpenTTD/OpenTTD/pull/13383
13:55:19 <Rubidium> it says std::mutex is either 40 or 80 bytes
13:56:18 <Rubidium> wow... biggus diffus ;)
13:59:19 <xarick> https://cdn.discordapp.com/attachments/1008473233844097104/1333073789835018250/image.png?ex=67979136&is=67963fb6&hm=268aabf2ee73337a042c0a2ff9d85a7af0c3ac4c14889e96ab920bb90081f888&
13:59:25 <peter1138> What's so funny about biggus diffus?
13:59:35 <xarick> okay, I see the difference now
13:59:49 <xarick> about 192 MB
13:59:57 <peter1138> I have a vewy gweat fwiend in Wome called biggus diffus.
14:02:47 <xarick> https://cdn.discordapp.com/attachments/1008473233844097104/1333074663001034884/image.png?ex=67979206&is=67964086&hm=f67a821e133ba9fd5fbbee3c4c27d398f8fa050be38a7a2a0b953edf195f08be&
14:02:54 <xarick> 8 bytes it says
14:08:12 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 commented on pull request #13383: Codechange: Specify underlying type for all enums. https://github.com/OpenTTD/OpenTTD/pull/13383#pullrequestreview-2574291266
14:10:04 <DorpsGek> [OpenTTD/OpenTTD] PeterN commented on pull request #13383: Codechange: Specify underlying type for all enums. https://github.com/OpenTTD/OpenTTD/pull/13383#pullrequestreview-2574293499
14:10:36 <DorpsGek> [OpenTTD/OpenTTD] PeterN commented on pull request #13383: Codechange: Specify underlying type for all enums. https://github.com/OpenTTD/OpenTTD/pull/13383#pullrequestreview-2574293642
14:10:42 <DorpsGek> [OpenTTD/OpenTTD] Kuhnovic opened pull request #13384: Change: Don't distinguish between bus and truck stops when removing them https://github.com/OpenTTD/OpenTTD/pull/13384
14:21:33 <DorpsGek> [OpenTTD/OpenGFX] glx22 commented on issue #94: gimpscript: GIMP3 compatibility problem(s) https://github.com/OpenTTD/OpenGFX/issues/94
14:25:05 <peter1138> Seems like API compatibility is not a concern.
14:25:45 <peter1138> Hmm, errors in all builds.
14:27:08 <peter1138> Oh, I guess they are not exported.
14:27:42 <_glx_> xarick: + at least 4 bytes of padding, so 12 bytes
14:28:03 <xarick> it's allocating 16 bytes 😦
14:28:17 <xarick> could be 8 if i were intelligent
14:28:44 <_glx_> no you can't make it less than 16 🙂
14:30:08 <DorpsGek> [OpenTTD/OpenTTD] PeterN updated pull request #13383: Codechange: Specify underlying type for all enums. https://github.com/OpenTTD/OpenTTD/pull/13383
14:30:25 <peter1138> Well, probably still problematic but should pass regression now.
14:39:21 <peter1138> More passes this time.
14:40:02 <DorpsGek> [OpenTTD/OpenGFX] planetmaker commented on issue #94: gimpscript: GIMP3 compatibility problem(s) https://github.com/OpenTTD/OpenGFX/issues/94
14:41:42 <xarick> https://cdn.discordapp.com/attachments/1008473233844097104/1333084459611721788/image.png?ex=67979b26&is=679649a6&hm=496ee6c31807bcf9f8487363d06e27cbbf50b238bb39930841800846e6bc80c8&
14:42:55 <xarick> just saved 64 MB
14:49:38 <xarick> https://cdn.discordapp.com/attachments/1008473233844097104/1333086454775156817/image.png?ex=67979d02&is=67964b82&hm=43317f7b945f8ded7ff4c9219fac50d2d1845d21f2bc4771d12bd4aeb8e4cdf6&
14:49:44 <xarick> yep, fantasticv
14:50:09 <xarick> but it's still a sad increase
14:50:15 <xarick> 302 to 430 😦
14:53:24 *** gelignite has joined #openttd
14:54:00 <xarick> hmm [RemoveRiverArtifacts mt] 16035 us it decreased
15:31:57 <xarick> i dont understand something...
15:33:30 <xarick> ``` for (TileIndex tile = start; tile < end; tile++) {
15:33:30 <xarick> {
15:33:30 <xarick> std::shared_lock read_lock_tile(TileMutex(tile));
15:33:30 <xarick> if (!IsTileType(tile, MP_WATER) || !IsRiver(tile)) continue;
15:33:30 <xarick> }
15:33:32 <xarick> }```
15:33:32 <xarick> only with this piece of code I get an additional 10000 us
15:33:46 <xarick> and the tile mutexes don't even collide
15:34:07 <xarick> what am I doing wrong
15:35:05 <peter1138> Locks.
15:35:34 <Rubidium> they *do* collide, with all the other (unknown to the compiler) parallel functions that you are running. You may say there are none, but the compiler doesn't know that
15:36:15 <Rubidium> and if you want to really slow things down, run it on a machine with multiple CPUs (not multiple cores)
15:38:23 * Rubidium has some work-site experience with someone buying the biggest computer with most CPUs for extra performance, only making the application that ran fine on my then 7 year old light laptop horrendously slow on a new dual CPU 64 core 512 GB machine
15:41:10 <DorpsGek> [OpenTTD/OpenTTD] michelkeijzers opened issue #13385: [Bug]: Dragging train station length does not work https://github.com/OpenTTD/OpenTTD/issues/13385
15:41:58 <peter1138> Is the orientation button confusing? That's the second one.
15:43:03 <DorpsGek> [OpenTTD/OpenTTD] PeterN commented on issue #13385: [Bug]: Dragging train station length does not work https://github.com/OpenTTD/OpenTTD/issues/13385
15:43:30 <DorpsGek> [OpenTTD/OpenTTD] michelkeijzers commented on issue #13385: [Bug]: Dragging train station length does not work https://github.com/OpenTTD/OpenTTD/issues/13385
15:45:44 <DorpsGek> [OpenTTD/OpenTTD] michelkeijzers commented on issue #13385: [Bug]: Dragging train station length does not work https://github.com/OpenTTD/OpenTTD/issues/13385
15:46:12 <DorpsGek> [OpenTTD/OpenTTD] michelkeijzers closed issue #13385: [Bug]: Dragging train station length does not work https://github.com/OpenTTD/OpenTTD/issues/13385
15:49:18 <xarick> dang, I want a 16 core cpu now 🙂
15:49:49 <xarick> 8+8 ryzen cores
15:50:06 <DorpsGek> [OpenTTD/OpenTTD] michelkeijzers commented on issue #13385: [Bug]: Dragging train station length does not work https://github.com/OpenTTD/OpenTTD/issues/13385
15:57:05 <xarick> just the lock itself... 14540 us
15:57:34 <xarick> is it because it's 16 million tile locks?
15:57:42 <xarick> mutexes
15:59:03 <xarick> https://cdn.discordapp.com/attachments/1008473233844097104/1333103923036291236/image.png?ex=6797ad47&is=67965bc7&hm=7112c5053fa8a15d7f88655f138901fb71bc8857147917ff377375cb5a274209&
15:59:34 <xarick> I mean... it's probably that
16:09:23 *** kuka_lie has joined #openttd
16:19:38 <_glx_> peter1138: maybe we also need to add a red blinking border around the selection 🙂
16:24:39 <peter1138> https://www.reddit.com/r/Somerset/comments/1i95ef5/does_anyone_know_why_theres_a_modern_train_at/
16:24:42 <peter1138> Oof
16:28:06 *** Wormnest has joined #openttd
16:42:09 <DorpsGek> [OpenTTD/OpenTTD] PeterN merged pull request #13368: Codechange: Remove ZeroedMemoryAllocator from WindowDesc https://github.com/OpenTTD/OpenTTD/pull/13368
16:45:40 <DorpsGek> [OpenTTD/OpenTTD] PeterN commented on pull request #13308: Codechange: use C++ constructs over MallocT/free https://github.com/OpenTTD/OpenTTD/pull/13308#issuecomment-2614500530
16:47:25 <DorpsGek> [OpenTTD/OpenTTD] PeterN commented on pull request #13315: Codechange: simplify management of temporary Engine in saveload https://github.com/OpenTTD/OpenTTD/pull/13315#issuecomment-2614501301
16:49:39 <DorpsGek> [OpenTTD/OpenTTD] PeterN commented on pull request #13317: Codechange: remove manual memory management in old string loading https://github.com/OpenTTD/OpenTTD/pull/13317#pullrequestreview-2574333387
16:51:32 <DorpsGek> [OpenTTD/OpenTTD] PeterN approved pull request #13324: Codechange: use std::vector to store SpriteCache instances in https://github.com/OpenTTD/OpenTTD/pull/13324#pullrequestreview-2574333736
16:58:53 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 commented on pull request #13317: Codechange: remove manual memory management in old string loading https://github.com/OpenTTD/OpenTTD/pull/13317#pullrequestreview-2574335306
16:59:01 <DorpsGek> [OpenTTD/OpenTTD] PeterN approved pull request #13329: Codechange: [Windows] replace C-style allocation with C++-style allocation https://github.com/OpenTTD/OpenTTD/pull/13329#pullrequestreview-2574335331
17:00:00 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 merged pull request #13329: Codechange: [Windows] replace C-style allocation with C++-style allocation https://github.com/OpenTTD/OpenTTD/pull/13329
17:00:11 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 merged pull request #13324: Codechange: use std::vector to store SpriteCache instances in https://github.com/OpenTTD/OpenTTD/pull/13324
17:01:02 <DorpsGek> [OpenTTD/OpenTTD] PeterN approved pull request #13335: Codefix: EngineID is used when it's not an actual EngineID https://github.com/OpenTTD/OpenTTD/pull/13335#pullrequestreview-2574335688
17:02:15 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 merged pull request #13335: Codefix: EngineID is used when it's not an actual EngineID https://github.com/OpenTTD/OpenTTD/pull/13335
17:02:18 <DorpsGek> [OpenTTD/OpenTTD] PeterN commented on pull request #13353: Codechange: [Script] do not redefine types in the script API https://github.com/OpenTTD/OpenTTD/pull/13353#pullrequestreview-2574335879
17:06:18 <DorpsGek> [OpenTTD/OpenTTD] PeterN approved pull request #13356: Codechange: rename CargoID to CargoType and amend related variables/comments https://github.com/OpenTTD/OpenTTD/pull/13356#pullrequestreview-2574337100
17:06:26 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 commented on pull request #13353: Codechange: [Script] do not redefine types in the script API https://github.com/OpenTTD/OpenTTD/pull/13353#pullrequestreview-2574337129
17:07:15 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 merged pull request #13356: Codechange: rename CargoID to CargoType and amend related variables/comments https://github.com/OpenTTD/OpenTTD/pull/13356
17:16:59 <DorpsGek> [OpenTTD/OpenTTD] LifWirser opened issue #13386: [Bug]: Little endian compile fail producing system error 9 https://github.com/OpenTTD/OpenTTD/issues/13386
17:28:57 <peter1138> I mean...
17:31:19 <DorpsGek> [OpenTTD/OpenTTD] WenSimEHRP commented on issue #13385: [Bug]: Dragging train station length does not work https://github.com/OpenTTD/OpenTTD/issues/13385
17:33:09 <peter1138> Rebased my secret cargo-related branch...
17:41:03 <_glx_> they seem to don't accept they need more RAM
17:43:47 <andythenorth> have they downloaded some?
17:53:31 <talltyler> Secret cargo branch: change cargo limit to 16, “64 was a mistake” 😛
17:53:45 <peter1138> Almost exactly.
17:58:12 <talltyler> I found a new way that people play OpenTTD: change the code trying to create bugs. Sort of like what we do, but on purpose.
18:04:31 <DorpsGek> [OpenTTD/OpenGFX] jayaddison commented on issue #94: gimpscript: GIMP3 compatibility problem(s) https://github.com/OpenTTD/OpenGFX/issues/94
18:16:25 <andythenorth> hmm does cb 161 have any var in scope for 'this variant has child variants'? https://newgrf-specs.tt-wiki.net/wiki/Callbacks#Engine_name_.28161.29
18:16:39 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 commented on issue #13386: [Bug]: Little endian compile fail producing system error 9 https://github.com/OpenTTD/OpenTTD/issues/13386
18:16:59 <andythenorth> then I can just compile the name callback once for all similar variants, not per variant
18:27:04 <DorpsGek> [OpenTTD/OpenTTD] PeterN commented on pull request #13363: Codechange: do not use MallocT for newgrf commons https://github.com/OpenTTD/OpenTTD/pull/13363#pullrequestreview-2574355772
18:42:49 <DorpsGek> [OpenTTD/OpenTTD] PeterN updated pull request #13377: Codechange: Remove ZeroedMemoryAllocator from network socket handlers. https://github.com/OpenTTD/OpenTTD/pull/13377
18:44:16 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 approved pull request #13377: Codechange: Remove ZeroedMemoryAllocator from network socket handlers. https://github.com/OpenTTD/OpenTTD/pull/13377#pullrequestreview-2574358832
18:45:12 <andythenorth> hmm I think if I could simply name callback handling, it would remove about 8000 switches, which are advanced varact2 setting registers
18:45:29 *** akimoto has joined #openttd
18:48:06 *** tokai has joined #openttd
18:48:06 *** ChanServ sets mode: +v tokai
18:52:59 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 commented on pull request #13363: Codechange: do not use MallocT for newgrf commons https://github.com/OpenTTD/OpenTTD/pull/13363#pullrequestreview-2574360494
18:53:01 <andythenorth> bool? has children?
18:54:09 <xarick> `EnforcePrecondition(VEHICLE_INVALID, !::IsValidCargoType(cargo) || ScriptCargo::IsValidCargo(cargo));` this looks suspicious
18:54:54 *** tokai|noir has quit IRC (Ping timeout: 480 seconds)
19:09:26 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain closed issue #13386: [Bug]: Little endian compile fail producing system error 9 https://github.com/OpenTTD/OpenTTD/issues/13386
19:09:36 <truebrain> cool, you can close tickets now as duplicate of another ticket
19:09:38 <truebrain> that is fancy
19:33:29 <andythenorth> I guess I could put the bool in a register, the compile knows about it
20:52:12 <peter1138> Ah, code that compiles with clang but not gcc.
20:52:50 *** gelignite has quit IRC (Quit: Stay safe!)
20:56:42 <andythenorth> oof I could read badges 😛
21:03:38 <peter1138> Could you.
21:14:42 <andythenorth> seems so
21:15:29 <andythenorth> badge for 'intended' variant level of each vehicle
21:15:43 <andythenorth> compare that with 'current' variant level
21:15:46 <andythenorth> switch string
21:15:55 <peter1138> Sounds bad.
21:16:13 <andythenorth> probably cuts out 8000 switches
21:16:40 <andythenorth> but I'd have to write python to set the badges
21:16:45 <belajalilija> replaced vehicle names with badges that conform to IPA standard
21:16:54 <andythenorth> working out the variant tree is already quite...the thing
21:17:47 <andythenorth> I wrote this in 2023, it works, I never want to touch it ever again 😛 https://github.com/andythenorth/iron-horse/blob/main/src/roster.py#L422
21:17:49 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 opened pull request #13387: Codechange: also add prefix increment/decrement for enums when requested https://github.com/OpenTTD/OpenTTD/pull/13387
21:18:55 <andythenorth> oof it's a bit early for naptime, but my eyes 😛
21:28:03 <andythenorth> new action 0 list of name strings, ordered by variant level?
21:28:43 <andythenorth> the text stack stuffing for every vehicle is really a lot of code 🙂
21:30:08 <andythenorth> dropping it cuts 1MB from the grf
21:32:07 <DorpsGek> [OpenTTD/OpenTTD] PeterN commented on pull request #13387: Codechange: also add prefix increment/decrement for enums when requested https://github.com/OpenTTD/OpenTTD/pull/13387#issuecomment-2614599532
21:32:33 <peter1138> Not again.
21:32:51 <peter1138> I have a trait patch for the bitset stuff too.
21:34:55 *** nielsm has quit IRC (Ping timeout: 480 seconds)
21:35:22 <Rubidium> *sigh*
21:35:32 <DorpsGek> [OpenTTD/OpenTTD] PeterN updated pull request #13381: Codechange: Introduce and use EnumBitSet. https://github.com/OpenTTD/OpenTTD/pull/13381
21:35:36 <peter1138> I know.
21:35:54 <peter1138> I probably had another patch at some point that wanted the prefix operators, and then did that a different way...
21:38:37 <andythenorth> ok such sleep 💤
21:39:33 <andythenorth> oh wait
21:39:41 <andythenorth> do I even need the text stack?
21:39:59 <andythenorth> I think it's only being used for `name` callback because I ran out of strings
21:40:17 <andythenorth> but perhaps the limit has been bumped
21:40:50 <andythenorth> and I've reduced consumption, I only need about 2500
21:41:02 <andythenorth> previously it was about 10k
21:49:05 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 updated pull request #13387: Codechange: also add prefix increment/decrement for enums when requested https://github.com/OpenTTD/OpenTTD/pull/13387
21:51:36 <DorpsGek> [OpenTTD/OpenTTD] PeterN opened pull request #13388: Codechange: Pass GRFConfig by reference where feasible. https://github.com/OpenTTD/OpenTTD/pull/13388
21:51:39 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 commented on pull request #13387: Codechange: also add prefix increment/decrement for enums when requested https://github.com/OpenTTD/OpenTTD/pull/13387#issuecomment-2614604927
21:54:14 <peter1138> . o O ( make branches of all my stashes, then push it all to github ... )
21:55:15 <DorpsGek> [OpenTTD/OpenTTD] PeterN updated pull request #13388: Codechange: Pass GRFConfig by reference where feasible. https://github.com/OpenTTD/OpenTTD/pull/13388
21:56:31 <peter1138> Hmm, I still don't know if my string-id-feature thing is valid, as it bumped the strings outside the 0000-FFFF range.
21:56:45 <peter1138> Though it seems to work, so "probably fine."
22:02:14 <peter1138> I wonder if GRFConfig is more easily movable now.
22:02:53 <peter1138> Ah, 280 bytes. That's quite a lot.
22:03:38 *** kuka_lie has quit IRC (Quit: Lost terminal)
22:05:34 <DorpsGek> [OpenTTD/OpenTTD] PeterN updated pull request #10835: Codechange: Use std::vector for GRFConfig lists. https://github.com/OpenTTD/OpenTTD/pull/10835
22:08:57 <peter1138> Hmm, a few `std::find_if` I can update there.
22:16:32 <DorpsGek> [OpenTTD/OpenTTD] 2TallTyler approved pull request #13384: Change: Don't distinguish between bus and truck stops when removing them https://github.com/OpenTTD/OpenTTD/pull/13384#pullrequestreview-2574405896
22:18:40 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 opened pull request #13389: Codechange: allow mapping enums as parameter and return type from scripts https://github.com/OpenTTD/OpenTTD/pull/13389
22:21:14 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 approved pull request #13388: Codechange: Pass GRFConfig by reference where feasible. https://github.com/OpenTTD/OpenTTD/pull/13388#pullrequestreview-2574406544
22:21:26 <DorpsGek> [OpenTTD/OpenTTD] PeterN commented on pull request #13389: Codechange: allow mapping enums as parameter and return type from scripts https://github.com/OpenTTD/OpenTTD/pull/13389#pullrequestreview-2574406583
22:24:52 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 updated pull request #13389: Codechange: allow mapping enums as parameter and return type from scripts https://github.com/OpenTTD/OpenTTD/pull/13389
22:24:57 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 commented on pull request #13389: Codechange: allow mapping enums as parameter and return type from scripts https://github.com/OpenTTD/OpenTTD/pull/13389#pullrequestreview-2574407101
22:28:22 <DorpsGek> [OpenTTD/OpenTTD] PeterN commented on pull request #13388: Codechange: Pass GRFConfig by reference where feasible. https://github.com/OpenTTD/OpenTTD/pull/13388#pullrequestreview-2574407569
22:28:56 <DorpsGek> [OpenTTD/OpenTTD] PeterN approved pull request #13389: Codechange: allow mapping enums as parameter and return type from scripts https://github.com/OpenTTD/OpenTTD/pull/13389#pullrequestreview-2574407636
22:29:26 *** keikoz has quit IRC (Ping timeout: 480 seconds)
22:41:11 <DorpsGek> [OpenTTD/OpenTTD] PeterN merged pull request #13388: Codechange: Pass GRFConfig by reference where feasible. https://github.com/OpenTTD/OpenTTD/pull/13388
22:47:36 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
23:11:00 <DorpsGek> [OpenTTD/OpenTTD] PeterN commented on pull request #13384: Change: Don't distinguish between bus and truck stops when removing them https://github.com/OpenTTD/OpenTTD/pull/13384#issuecomment-2614627790
23:31:03 *** HerzogDeXtEr has quit IRC (Read error: Connection reset by peer)
23:50:57 <DorpsGek> [OpenTTD/OpenTTD] PeterN updated pull request #10835: Codechange: Use std::vector for GRFConfig lists. https://github.com/OpenTTD/OpenTTD/pull/10835
continue to next day ⏵
