IRC logs for #openttd on OFTC at 2021-06-14
⏴ go to previous day
00:13:30 *** tokai|noir has joined #openttd
00:13:30 *** ChanServ sets mode: +v tokai|noir
00:20:16 *** tokai has quit IRC (Ping timeout: 480 seconds)
00:44:14 <DorpsGek> [OpenTTD/OpenTTD] VacuumBreather commented on pull request #9043: Feature: ability to build overlapping bridges on different axes (WIP) https://git.io/JZhau
01:02:43 *** ChanServ sets mode: +v tokai
01:09:15 *** tokai|noir has quit IRC (Ping timeout: 480 seconds)
01:10:43 *** tokai|noir has joined #openttd
01:10:43 *** ChanServ sets mode: +v tokai|noir
01:17:28 *** tokai has quit IRC (Ping timeout: 480 seconds)
02:02:22 *** Wormnest has quit IRC (Quit: Leaving)
02:55:22 *** debdog has quit IRC (Ping timeout: 480 seconds)
03:16:10 *** ChanServ sets mode: +v tokai
03:23:06 *** tokai|noir has quit IRC (Ping timeout: 480 seconds)
05:05:58 *** sla_ro|master has joined #openttd
05:44:07 *** sla_ro|master has quit IRC ()
06:18:23 *** tokai|noir has joined #openttd
06:18:23 *** ChanServ sets mode: +v tokai|noir
06:25:06 *** tokai has quit IRC (Ping timeout: 480 seconds)
06:37:27 <DorpsGek> [OpenTTD/OpenTTD] James103 commented on pull request #9366: Fix 81062163: for (really) old games, station bus/truck station cache was not updated https://git.io/JnebJ
06:48:44 *** esselfe has joined #openttd
07:15:39 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9366: Fix 81062163: for (really) old games, station bus/truck station cache was not updated https://git.io/JnvOC
07:21:02 <_dp_> TrueBrain, btw, that testing is not enough to test _all_ afterload cases
07:21:29 <_dp_> since you always call afterload at least once and there are some cases (i.e. server after new game) where it doesn't call afterload at all
07:21:50 <_dp_> for example that way you won't pick up VRF_EL_ENGINE_ALLOWED_NORMAL_RAIL difference that exists
07:24:01 * Rubidium wonders what new game (server or non-server) comes with road stops
07:24:54 <_dp_> I didn't mean _right_ after the new game, there can be some player actions
07:25:01 <_dp_> point is, server never calls afterload
07:26:24 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9339: Codechange: add and use the ability to store (lists of) structs in savegames https://git.io/JZLZC
07:27:00 <_dp_> for example, with VRF_EL_ENGINE_ALLOWED_NORMAL_RAIL afterload always sets it if elrail is disabled, but newgame doesn't
07:27:49 <Rubidium> does that cause problems?
07:28:02 <_dp_> no, that flag doesn't rly matter if it's disabled
07:29:27 <Rubidium> so, it's not needed for "new" savegames, but it is needed for old savegames that did not have that electric rails. Sounds like afterload is the right place for that
07:30:22 <_dp_> flag difference is alright, I just mentioned it as an example of difference that testing afterload that way won't pick up
07:34:09 <Rubidium> still, it seems kinda pointless to mention. It's like saying, after someone changed something to trains, that you did not test all cases because you didn't test aircraft
07:35:50 <Xaroth> Sounds like complaining for the sake of complaining.
07:36:15 <_dp_> I'm mentioning because in that pr TB says "I did this to validate "Afterload", that we are doing the right thing in all cases."
07:36:33 <_dp_> so it's cool, but that's not _all_ the cases
07:37:04 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9322: Add: store table header for each chunk in savegame https://git.io/JGlmd
07:37:13 <Xaroth> By that definition we might as well spend the next decade testing for all use cases that are impossible to happen, because otherwise we wouldn't be testing "all" the cases.
07:37:44 *** HerzogDeXtEr has joined #openttd
07:37:54 <_dp_> in fact, no afterload-related network desync would be caught like that because it will be comparing the same thing
07:39:09 <_dp_> Xaroth, huh? that's just a remark, chill
07:39:21 <_dp_> I don't even know how to reasonably check those cases
07:39:32 <TrueBrain> Rubidium: fun fact ... I used to drop "debug.h" include whenever I wanted to add a DEBUG() statement somewhere, as I am lazy .. in result, often I forget to clean up the random "debug.h" include .. BUT NO LONGER! It fails to compile, as fmt uses snprintf, which you can no longer after the safeguard :D
07:39:50 <TrueBrain> so my laziness is now punished, rightfully :P
07:40:45 <Xaroth> _dp_: You do realize that I was holding up a mirror, right?
07:42:11 <_dp_> Xaroth, I've no idea wtf are you on about
07:48:21 <Rubidium> in any case, the "cases" that were spoken about was loading an old save game, and a the new save game of the same game state and comparing they are the same, as that's the premise that is set up in the lines before. Not "a new game, and then loading that on another client" cases
07:49:37 *** Ttech has quit IRC (Quit: Este é o fim.)
07:56:32 *** andythenorth has joined #openttd
07:56:35 *** andythenorth has quit IRC ()
07:57:27 <_dp_> Rubidium, whatever, it was just a passing remark on something I think is worth keeping in mind, don't start a drama out of it
07:58:19 *** snail_UES_ has quit IRC (Quit: snail_UES_)
07:58:52 <peter1138> Oh no, nearly work o'clock :(
07:59:32 <Xaroth> peter1138: been work o'clock for me for almost an hour now.... I'm already looking forward to weekend o'clock.
07:59:43 <peter1138> Weekends are too short.
08:00:06 <peter1138> Hmm, 14 hrs 43 mins in VS Code last week. That is odd, used to be 40+ hours.
08:00:31 <peter1138> Oh of course, had several stupid meetings that ended up with horrible work in Excel.
08:00:39 <TrueBrain> I agree, weekends are too short .. this is why I extended them with no job-weeks :P
08:01:57 <peter1138> Couple of years I'll have paid off the mortgage and that may be an option :p
08:03:00 <TrueBrain> ha, I am far away from paying that off :P
08:06:40 <Xaroth> only 30 more years to go \o/
08:10:03 <TrueBrain> 21 and 25 for me :P
08:10:18 <TrueBrain> after 21 the biggest bulk is done, 4 years later the rest
08:10:34 <TrueBrain> does require me to not move out :P
08:34:51 <TrueBrain> "wait, I made this change 3 times already, why do I have to do it again?!" - "owh .... this file is autogenerated ...."
08:35:34 <TrueBrain> way too used to ctrl+click on error lines
08:35:38 <TrueBrain> VSCode made that so smooth
08:38:42 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9322: Add: store table header for each chunk in savegame https://git.io/JGlmd
08:44:21 <TrueBrain> owh, the nightly failed
08:44:48 <TrueBrain> "clang: error: unable to execute command: Bus error: 10"
08:44:57 <TrueBrain> very descriptive too
08:46:43 <TrueBrain> it is the arm64 build
08:46:48 <TrueBrain> so I cannot judge if it was a temporary failure
08:46:55 <TrueBrain> or that the fmt addition put it over the edge
08:49:03 <TrueBrain> guess we need orudge to find out what it is exactly that is causing it :P
08:54:43 <peter1138> I remember having issues with Alpha back in the day. Just being 64 bit was enough to cause issues.
09:00:34 <TrueBrain> okay, also interesting .. I only refactored my PR a bit, basically moved code around between commits
09:00:37 <TrueBrain> and now the test fails :P
09:01:18 <TrueBrain> a DemolishTile is not being done .. lol
09:06:03 <TrueBrain> and this is why "git reflog" is incredibly useful :)
09:07:48 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9322: Add: store table header for each chunk in savegame https://git.io/JGlmd
09:15:37 *** NGC3982 has quit IRC (Quit: leaving freenode)
09:16:43 *** NGC3982 has joined #openttd
09:18:11 *** NGC3982 has joined #openttd
09:49:40 <peter1138> I did once get into the habit of creating a separate branch before rebasing/refactoring
09:55:35 <Rubidium> yeah, that's especially useful if you want to have the same result at the end, but are moving around a lot of things in commits
09:55:55 <Rubidium> although I might have misused tags for that
09:56:42 <peter1138> I tend to consider tags as permanent.
11:10:38 <DorpsGek> [OpenTTD/OpenTTD] PeterN opened pull request #9373: Cleanup: nested_array should not be accessed directly. https://git.io/JnJv0
11:11:16 <TrueBrain> can't we make the variable private or something? :P
11:13:06 <LordAro> TrueBrain: do i need to tell you off for making a comment in IRC rather than the PR? :p
11:13:28 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain approved pull request #9373: Cleanup: nested_array should not be accessed directly. https://git.io/JnJfi
11:14:16 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9373: Cleanup: nested_array should not be accessed directly. https://git.io/JnJfh
11:15:26 *** WormnestAndroid has quit IRC (Remote host closed the connection)
11:16:57 *** WormnestAndroid has joined #openttd
11:17:12 <TrueBrain> now to find a savegame with a GSTR chunk ..
11:27:18 <TrueBrain> lol, GSTR chunk reads a std::string, runs c_str() over it, to store it in a std::string again
11:27:28 <TrueBrain> partial C++ conversions stacking up yo!
11:30:58 <LordAro> there's probably a lot of those
11:31:08 <TrueBrain> "fun fact" .. a GameScript language blob stored in savegames don't have to be valid entries
11:31:16 <TrueBrain> "##### STORY BOOK #####" are also stored
11:32:11 <Rubidium> yup, it's easy to miss those. Though, there should be less c_str()s than before the weekend, so it should be slightly easier to sift through them (which is basically what I've been doing for some of my std::string conversion PRs)
11:32:43 <TrueBrain> well, I am going to help you, and remove .. 3 I think :P
11:32:51 <TrueBrain> also one less SLE_STR
11:33:58 <DorpsGek> [OpenTTD/OpenTTD] PeterN commented on pull request #9373: Cleanup: nested_array should not be accessed directly. https://git.io/JnJm2
11:35:01 <_dp_> it kind of makes sense that gs strings are stored uncompiled, saves from maintaining binary compatibility
11:35:27 <TrueBrain> huh? Who was talking about compiled?
11:37:23 <_dp_> you? that's kind of why it has all the comments and junk
11:38:27 <TrueBrain> I am a bit scared that validate means compiled in your world :s
11:39:54 <TrueBrain> I made the table-stuff fail hard when I try to store a table header without it being a table chunk
11:39:58 <TrueBrain> as I kept doing that wrong
11:40:05 <TrueBrain> I can tell you .. I do it wrong EVERY SINGLE conversion I do :P
11:40:23 <_dp_> TrueBrain, there is no "validation" per se in openttd iirc, raw strings just go into Compile()
11:40:47 <_dp_> sure, you could theoretically add an extra step and some intermediate format but why
11:41:52 <peter1138> Is this another irrelevant factoid, just to bear in mind?
11:42:25 <_dp_> yeah, just a "fun fact" ;)
11:42:59 <peter1138> It doesn't seem to relate to what TrueBrain is doing.
11:43:29 <peter1138> Sometimes a bit of context is useful.
11:44:31 <_dp_> there is no good way, to reply to a specific message in irc, look a few messages higher :P
11:44:59 <peter1138> How many is a few? None of it is context to what you wrote.
11:45:13 <_dp_> TrueBrain> "fun fact" .. a GameScript language blob stored in savegames don't have to be valid entries
11:46:44 <_dp_> also, why do I now feel like a need to pack a car full of lawyers to talk in this chat? :p
11:47:39 <TrueBrain> something something mirror
11:49:17 <TrueBrain> ha, I can now read GSDT correctly too via savegame-reader .. okay, that was easier than I expected
11:50:17 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9322: Add: store table header for each chunk in savegame https://git.io/JGlmd
11:50:17 <TrueBrain> right .. linkgraph left .... oof
11:51:18 <Rubidium> TrueBrain: added bonus for the linkgraph parsing in your new tool: .dot output, or json that dotty understands (if that's possible) ;)
11:51:28 <peter1138> You seem to jump to massive conclusions about what people are talking about, and then never explain the connection, assuming it's obvious. I still don't see the connection here other than gamescript and strings...
11:51:59 <TrueBrain> Rubidium: you do like me digging this rabbit hole nice and deep, don't you? :P
11:52:49 <peter1138> I should add some bacon to my bacon flavoured Tucs.
11:53:52 <Rubidium> TrueBrain: no, I want to warn you for the rabbit hole so you see it and don't break your leg ;)
11:54:42 <Rubidium> as the OpenTTD field you're walking through has a lot of, yet unseen, rabbit holes to fall into or break your leg in (on?)
11:55:47 <TrueBrain> true true, very true :)
11:56:05 <TrueBrain> it is appreciated :)
11:56:11 <_dp_> peter1138, what other connection do you expect? TB commented on gs strings and so did I
12:06:16 <TrueBrain> "Impossible as they have been present in previous game." <- I love the optimism in that statement :)
12:15:42 <TrueBrain> funny, LinkGraph in-memory storage is as such that is it non-trivial to store it in the SaveLoad struct method without keeping some meta-data in a global
12:15:52 <TrueBrain> basically, in-memory is not nested
12:16:08 <TrueBrain> 2 "globals" track edges and nodes based on index, which are not stored in the object itself
12:41:48 <Timberwolf> My retail copy of Transport Tycoon appears to have a behaviour where if a primary industry transports more than about 70% of its cargo, it gets immediately marked for closure.
12:41:54 <Timberwolf> I'm surprised I never noticed this.
12:42:09 <TrueBrain> I am surprised you are playing the original :P
12:42:19 <Timberwolf> Although that said, I used to have a patch floppy so maybe that got fixed from the retail version.
12:43:02 <Timberwolf> I did a little playing around in it, mostly to see what building a network was like without any of the modern signalling options.
12:43:43 *** snail_UES_ has joined #openttd
12:44:14 <Timberwolf> None of your new-fangled "building signals on diagonal track" here!
12:44:26 <esselfe> When I 'save gamename', shutdown the server and later restart with -g gamename.sav, the company passwords disappeared, why?
12:44:49 <TrueBrain> because company passwords are not stored in the savegame
12:45:21 <esselfe> ok I'll advise my players that I had to restart :P
12:46:59 <esselfe> storing/comparing SHA 256 sums in the savegame wouldn't be possible?
12:48:13 <Timberwolf> The other thing I liked from the original - "Tutorial 1" (building a basic road service) has an error, it builds one of the road stops in a place where nothing is in the catchment area.
12:48:24 <milek7> I guess nobody wants to improve passwords because they are bad mechanism anyway
12:48:48 <Timberwolf> I think the TTD one is different entirely, so it got fixed between TTO and TTD.
12:53:01 <TrueBrain> esselfe: it is all a bit nuanced, but it is not trivial to do so
12:53:13 <TrueBrain> so possible, yes, but nobody put in the effort for it
12:53:34 <TrueBrain> (as in, enough effort has been put in it, but nothing that hit master yet)
12:53:54 <TrueBrain> so possible, yes, but nobody carried it to the goal-line
12:53:59 <TrueBrain> is a better way of wording myself :)
12:54:04 <Samu> it's not really a password
12:54:21 <glx> it is a password, but not very safe :)
12:54:48 <TrueBrain> but we ahve the first ticket dating back to 2007 requesting the functionality :P
12:56:36 *** ChanServ sets mode: +v tokai
13:03:19 *** tokai|noir has quit IRC (Ping timeout: 480 seconds)
13:14:14 <peter1138> I guess that just means we're all lazy.
13:36:06 <TrueBrain> I am weirdly surprised linkgraph stuff works at all :P
13:36:32 <TrueBrain> but lets not look into the details of our Pool system ..
13:45:12 *** WormnestAndroid has quit IRC (Remote host closed the connection)
13:45:25 *** WormnestAndroid has joined #openttd
13:45:26 <Eddi|zuHause> i think the main problem with saving passwords was: "we need to do cryptography, but we can't be bothered to do it properly"
13:45:34 <Timberwolf> Client certificates would be a nice way of handling the multiplayer problem, although you have the issue of sharing them between multiple people who want to use a shared company.
13:46:23 <Timberwolf> So you either need players to share them outside the game, or have some kind of "invite code" or "player x wants to join company, allow?" functionality.
13:46:47 <TrueBrain> Timberwolf: you are aware there is a PR and discussion addressing all this? :D
13:47:08 <Timberwolf> Oh cool, PR already?
13:47:21 <TrueBrain> still has a way to go, but it shows a possible approach to this problem
13:48:13 <TrueBrain> AddressSanitizer: allocator is out of memory trying to allocate 0x31223f400 bytes
13:48:17 <TrueBrain> that is not good, is it?
13:48:47 <Eddi|zuHause> sounds a tad large :p
13:49:16 <Eddi|zuHause> that's like what? 12GB?
13:49:32 <TrueBrain> check your DHCP server, I have been told it contains the answer
13:50:26 <TrueBrain> @calc 0x31223f400 / 1024 / 1024 / 1024
13:50:26 <DorpsGek> TrueBrain: 12.28344440460205
13:50:33 <TrueBrain> Rubidium: seems you are not correct, sorry :P
13:50:58 <Rubidium> @calc 0x31223f400 / 1000 / 1000 / 1000
13:51:18 <TrueBrain> lol, you became a harddisk manufacturer?
13:52:06 <Rubidium> if not specifically noted, I assume the proper SI units ;)
13:52:27 <TrueBrain> well, Eddi|zuHause used a weird unit, so that was an assumption on your part :P :P
13:52:39 <TrueBrain> don't we love a good troll :)
13:53:26 <Eddi|zuHause> Rubidium: but isn't GiB specifically referring to the 1024 based unit?
13:54:07 <Rubidium> Eddi|zuHause: yes, and isn't 12.28 closer to 13 than 13.18 is to 12?
13:59:41 <TrueBrain> omg, I have linkgraphs in JSON format!
13:59:43 <TrueBrain> what an utter mess :P
14:00:36 <TrueBrain> not the code, but the JSON output, to be clear
14:01:26 <TrueBrain> job -> linkgraph -> node -> edge
14:01:43 <TrueBrain> and jobs are weird .. it seems to hijack an index from the linkgraph pool
14:01:47 <TrueBrain> not using the correct method of doing so
14:01:51 <TrueBrain> dunno .. bit odd duck
14:03:03 <TrueBrain> but ... this does mean I finally managed to do all chunks, except for map-chunks
14:03:18 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9322: Add: store table header for each chunk in savegame https://git.io/JGlmd
14:03:38 <TrueBrain> glx: btw, I did find a clean way to use CH_READONLY already
14:03:51 <TrueBrain> so I cherry-picked that part from your commit :P
14:09:04 <TrueBrain> right, time for hopefully the final run of tests :D
14:10:49 <glx> oh I should steal your default span constructor
14:12:17 <TrueBrain> yeah, without it std::optional doesn't work :D :D
14:13:20 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9339: Codechange: add and use the ability to store (lists of) structs in savegames https://git.io/JZLZC
14:13:29 <TrueBrain> finally ready for review ^^ :)
14:17:22 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 updated pull request #9371: Assorted C-string -> std::string conversions in network https://git.io/JZxHY
14:22:17 <TrueBrain> needed to make the PR to know the number to update the PR :P
14:24:00 <Rubidium> const TOwn * <- was about to comment on the capitalisation there ;)
14:27:28 <TrueBrain> it confused me more than once
14:27:30 <TrueBrain> but it is correct :P
14:27:33 <TrueBrain> I take suggestions :)
14:28:05 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain opened pull request #9375: Change: make as many savegame chunks a CH_ARRAY as possible https://git.io/JnUze
14:32:49 <TrueBrain> owh, I still have a TODO to fix .. hmm .. well, SO CLOSE :)
14:32:58 <TrueBrain> luckily it is in one of the last commits, so what-ever
14:43:43 <TrueBrain> and of course there is a warning in the firs tPR
14:44:39 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9339: Codechange: add and use the ability to store (lists of) structs in savegames https://git.io/JZLZC
14:48:44 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9375: Change: make as many savegame chunks a CH_ARRAY as possible https://git.io/JnUze
14:48:50 <TrueBrain> 3 PRs, 3 different warnings
14:52:00 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9375: Change: make as many savegame chunks a CH_ARRAY as possible https://git.io/JnUze
14:52:59 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9322: Add: store table header for each chunk in savegame https://git.io/JGlmd
14:54:44 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9322: Add: store table header for each chunk in savegame https://git.io/JnUXT
14:55:05 <TrueBrain> what was especially funny that the warnings from one were fixed by the other
14:57:22 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 commented on pull request #9339: Codechange: add and use the ability to store (lists of) structs in savegames https://git.io/JnU1s
15:01:41 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9339: Codechange: add and use the ability to store (lists of) structs in savegames https://git.io/JnUDm
15:02:00 <TrueBrain> I knew someone would complain about that, but forgot to write that in the description :P
15:11:39 <TrueBrain> lol, the CI is being busy :P
15:12:07 <TrueBrain> maybe I should add a docs/savegame-format.md
15:12:16 <TrueBrain> explaining how this very simple savegame format works
15:12:25 <TrueBrain> LordAro will like that, I am sure :D
15:15:21 <Timberwolf> Well-documented savegame format, people can write offline GS :)
15:15:34 <Timberwolf> "please save and quit now, so the GS can run its next tick"
15:19:21 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9339: Codechange: add and use the ability to store (lists of) structs in savegames https://git.io/JZLZC
15:20:18 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 commented on pull request #9339: Codechange: add and use the ability to store (lists of) structs in savegames https://git.io/JnUd3
15:32:39 <TrueBrain> hmm .. I expected MSVC to complain about one part of the code .. it does not
15:32:54 <TrueBrain> so "friends" now either work or turn out to not be needed :P
15:34:03 <Rubidium> or the builder got updated?
15:34:24 <glx> there's no new VC2017 compiler IIRC
15:41:10 <TrueBrain> ah, the other code wants access to protected, not private information
15:41:13 <TrueBrain> there seems to be the difference
15:56:27 <TrueBrain> euuhh ... master kinda resets a lot of settings
15:56:49 <TrueBrain> 2TallTyler made the mistake and both me and michi_cc didn't spot it ... and my PR is exactly trying to prevent that mistake :D
15:58:03 <TrueBrain> owh, no, I am wrong
15:58:08 <TrueBrain> ugh, this is so annoying to read
15:58:16 <TrueBrain> I am happy my PR removes that problem :P
15:59:52 <TrueBrain> basically, a new setting got added, but I didn't add it in my compat layer in my PR .. so it was resetting all the settings
16:00:03 <TrueBrain> but I completely missed the "SLV" entry in the settings :P
16:03:41 *** crem has quit IRC (Quit: WeeChat 3.0)
16:09:26 *** Progman has joined #openttd
16:09:33 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 commented on pull request #9374: Change: make lists in savegames consistent https://git.io/JnT3y
16:11:30 *** Wormnest has joined #openttd
16:26:24 *** D-HUND has quit IRC (Quit: Initiating getting-the-hell-out-of-here maneuver!)
16:27:39 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 commented on pull request #9375: Change: make as many savegame chunks a CH_ARRAY as possible https://git.io/JnT8e
16:30:45 <DorpsGek> [OpenTTD/OpenTTD] PeterN commented on pull request #9373: Cleanup: nested_array should not be accessed directly. https://git.io/JnT4l
16:41:26 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9374: Change: make lists in savegames consistent https://git.io/JnTun
16:42:52 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9375: Change: make as many savegame chunks a CH_ARRAY as possible https://git.io/JnTuF
16:49:05 <DorpsGek> [OpenTTD/OpenTTD] PeterN merged pull request #9373: Cleanup: nested_array should not be accessed directly. https://git.io/JnJv0
17:49:43 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9339: Codechange: add and use the ability to store (lists of) structs in savegames https://git.io/JZLZC
17:55:14 *** sla_ro|master has joined #openttd
18:01:00 *** Flygon has quit IRC (Quit: A toaster's basically a soldering iron designed to toast bread)
18:05:54 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9375: Change: make as many savegame chunks a CH_ARRAY as possible https://git.io/JnUze
18:09:24 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9322: Add: store table header for each chunk in savegame https://git.io/JGlmd
18:11:55 <TrueBrain> Rubidium: you were joking about it, I think, but for sure it is now possible to render a map with the linkgraphs on it :P
18:16:05 <TrueBrain> just the way it is stored on disk is a bit awkward
18:17:02 <TrueBrain> nodes: 0, 1. edges: [0][0] -> (0 -> 0) [0][1] -> (0 -> 1) .. so far so good. But now: edges: [1][0] -> (1 -> 1), [1][1] -> (1 -> 0)
18:17:13 <TrueBrain> not sure that makes sense .. but the order on disk is not the order of edges
18:31:54 *** gelignite has joined #openttd
18:32:51 <_dp_> wait, all that was to move savegames to sqlite? why?
18:33:11 <_dp_> I mean, I get some external save to sqlite converter but as a savegame format it fells like a regression
18:33:21 <_dp_> as it gets not of the benefits but all the downsides
18:35:46 <_dp_> db stuff it doesn't need as the game still needs to load it fully
18:36:04 <_dp_> and I don't think it will do any good for size and speed which are very important for the save game
18:55:02 <DorpsGek> - Update: Translations from eints (by translators)
19:00:39 <Xaroth> they aren't going to use sqlite, it's all going to be xml.
19:01:59 <Xaroth> then with some epic xsl trickery allow you to create a nice HTML render of your savegame's map
19:05:46 <TrueBrain> meh, turns out to be pretty difficult to find a Python lib that can draw self-loops and isn't ugly
19:22:29 <TrueBrain> ValueError: Some given node positions are not within the data range specified by `origin` and `scale`!
19:25:32 <TrueBrain> so it calculates positions for me, then complains they are not within bounds
19:33:59 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 approved pull request #9339: Codechange: add and use the ability to store (lists of) structs in savegames https://git.io/JnkKe
19:43:18 <glx> next step replace numbers in circle with station name ;)
19:43:34 <TrueBrain> they are not in the savegame :'( :'(
19:43:56 <TrueBrain> as I said, they are not in the savegame :P :P
19:43:57 <glx> else it's magic generation
19:58:09 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain merged pull request #9339: Codechange: add and use the ability to store (lists of) structs in savegames https://git.io/JZLZC
19:59:40 <glx> haha my client replaced the __ in url (and webster seems to do it too)
20:04:33 <glx> youhou conflicts everywhere :)
20:11:52 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9372: Fix: [Network] Determining GetNetworkRevisionString could overflow and underflow its buffer https://git.io/Jnk5w
20:12:14 <TrueBrain> glx: I think it was my client that did it :P
20:19:43 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9371: Assorted C-string -> std::string conversions in network https://git.io/Jnkb2
20:25:53 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9369: Fix: don't show join button in online players window when the company is an AI https://git.io/JnkAH
20:32:14 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9368: Codechange: Remove last FOR_XXX macros https://git.io/JnkhZ
20:32:21 <TrueBrain> poeh ... that was a hard one to review .. can't imagine how it was to write :P
20:38:03 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9368: Codechange: Remove last FOR_XXX macros https://git.io/JnIe4
20:45:41 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 updated pull request #9372: Fix: [Network] Determining GetNetworkRevisionString could overflow and underflow its buffer https://git.io/JZpCf
20:47:11 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain approved pull request #9372: Fix: [Network] Determining GetNetworkRevisionString could overflow and underflow its buffer https://git.io/JnIJQ
20:48:08 *** sla_ro|master has quit IRC ()
20:55:28 *** virtualrandomnumber has joined #openttd
20:55:48 *** virtualrandomnumber has quit IRC ()
21:05:22 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 merged pull request #9372: Fix: [Network] Determining GetNetworkRevisionString could overflow and underflow its buffer https://git.io/JZpCf
21:05:51 *** nielsm has quit IRC (Ping timeout: 480 seconds)
21:15:23 *** Samu has quit IRC (Quit: Leaving)
21:22:47 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 commented on pull request #9371: Assorted C-string -> std::string conversions in network https://git.io/JnIGW
21:23:46 <glx> not bad, with all the huge conflicts to solve in my saveload branch, the only error I had was a missing }
21:27:11 <TrueBrain> LordAro: but I had faith in you!!
21:33:20 <TrueBrain> ps: nightly is still failing
21:34:53 <TrueBrain> for some reason I am going to laugh if it turns out to be the 'static std::string_view abc = "def";' we now have :P
21:35:02 <TrueBrain> I wouldn't be surprised if Apple does something idiotic there :P
21:36:08 *** gelignite has quit IRC (Quit: Stay safe!)
21:36:18 <glx> would be nice to at least find the breaking commit
21:36:45 <TrueBrain> I believe only orudge has an ARM Apple at hand. So I think he is the only one who can find the commit
21:37:24 <TrueBrain> I do not really have an idea how we otherwise can find it .. except for abusing the CI a lot .. which would be really annoying and painful
21:40:15 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 updated pull request #9371: Assorted C-string -> std::string conversions in network https://git.io/JZxHY
21:40:59 <glx> we don't even know if x64_86 osx release builds fine
21:41:10 <TrueBrain> but we can ask others that
21:41:31 <TrueBrain> at least the CI works for it
21:41:34 <TrueBrain> so that is something :D
21:41:39 <glx> I'm trying win64 clang release as a test
21:42:43 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain approved pull request #9371: Assorted C-string -> std::string conversions in network https://git.io/JnI8w
21:44:22 <glx> because OSX is the only platform using clang with RelWithDebInfo
21:44:37 <glx> CI checks clang but only for Debug
21:45:06 <glx> and who knows what happens during LTO
21:49:35 <TrueBrain> sometimes reddit is just fun .. post #1 of user, super proud of his station design .. post #2, a much more efficient version of his station design "based on feedback"
21:49:36 <glx> oh linux generic seems to warn when linking
21:49:44 <TrueBrain> you don't have to open post #1 to know how that conversation went :P
21:50:41 <TrueBrain> glx: yeah, has been there for a while now
21:51:51 <TrueBrain> hmm, no, that is a lie
21:51:54 <TrueBrain> wasn't there 2 days ago, it seems
21:51:59 <TrueBrain> locally I have had it for a while now :P
21:54:14 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 dismissed a review for pull request #9371: Assorted C-string -> std::string conversions in network https://git.io/JnI8w
21:54:17 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 updated pull request #9371: Assorted C-string -> std::string conversions in network https://git.io/JZxHY
21:54:23 <TrueBrain> you want me to approve it again or something?
21:55:00 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain approved pull request #9371: Assorted C-string -> std::string conversions in network https://git.io/JnI0w
21:55:11 <glx> ok no fail with msvc clang
21:55:20 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 commented on pull request #9371: Assorted C-string -> std::string conversions in network https://git.io/JnI0X
21:57:06 <Rubidium> TrueBrain: no, I definitely didn't want you to have to re-approve it...
21:57:24 <TrueBrain> glx: I cannot reproduce that error anymore ... lol
21:57:36 <Rubidium> but as you already mentioned, Apple works in mysterious ways
21:58:35 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
22:03:07 <TrueBrain> ah, GCC only warning
22:10:18 <TrueBrain> compiles take ~5 minutes each
22:10:22 <TrueBrain> so bisecting is .. euh .. taking a while
22:11:38 <DorpsGek> [OpenTTD/OpenTTD] glx22 commented on pull request #9369: Fix: don't show join button in online players window when the company is an AI https://git.io/JnIaC
22:12:32 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9369: Fix: don't show join button in online players window when the company is an AI https://git.io/JnIaD
22:13:03 <TrueBrain> owh, right, I was wondering if we wanted the button disabled
22:14:01 <glx> for me, nobody should join AI company
22:14:53 <TrueBrain> if that is what console does, that is what the GUI should do
22:14:57 <TrueBrain> but he now removes the button
22:15:02 <TrueBrain> but we also disable buttons
22:15:12 <TrueBrain> which means the GUI is a bit less .. jumpy, with buttons
22:15:48 <glx> disabling should be better visually
22:20:37 <glx> /* Disable the chat button when there are players in this company. */
22:20:37 <glx> chat_button->disabled = !has_players;
22:20:49 <glx> double negation failure in the comment ?
22:21:33 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9369: Fix: don't show join button in online players window when the company is an AI https://git.io/JnIot
22:22:59 <glx> too bad we can't add comments on lines not in the diff
22:24:49 <TrueBrain> cd .. && git bisect good && cd build4 && make -j9
22:24:53 *** HerzogDeXtEr has quit IRC (Read error: Connection reset by peer)
22:24:54 <TrueBrain> stupid bisect must be executed from root bla
22:26:35 <TrueBrain> okay, that GCC warning is introduced by the Debug() commit
22:30:14 <TrueBrain> but I think that is just what pushed it over the edge
22:30:19 <TrueBrain> as I had it in my PR for a while longer now
22:30:29 <TrueBrain> I think it just has too many lines of code in a single source-file
22:33:58 <glx> it's not the biggest file we have
22:34:21 <TrueBrain> remember it includes a 1000 line long settings table :P
22:38:04 <glx> maybe this gcc warning has the same cause as apple ld crash
22:39:38 <TrueBrain> GCC gives this warning when it is tracking 50M variables
22:43:35 <TrueBrain> I am now getting the error on much earlier commits
22:43:40 <TrueBrain> this is weirdly inconsistent
22:55:14 *** Progman has quit IRC (Remote host closed the connection)
22:55:27 <TrueBrain> omg, these LTO compiles ...
22:55:34 <TrueBrain> GCC, come on, this is just painful
23:02:19 <TrueBrain> bisecting now claims it is the commit changing SettingDesc::name to a std::string
23:05:53 <TrueBrain> okay, I can reproduce that now somewhat reliable it seems
23:12:07 <TrueBrain> lets try something to see if we can fix it a tiny bit
23:17:31 <TrueBrain> moved a shitton out of the file, still the same error
23:18:31 *** tokai|noir has joined #openttd
23:18:31 *** ChanServ sets mode: +v tokai|noir
23:18:38 <TrueBrain> no clue .. going to give up on this :) Somewhat doubt it is the same issue as MacOS is having
23:18:42 <glx> I think it's not related to particular file
23:19:43 <glx> as it's tracking all variables for improved debug info during optimisation
23:22:55 <TrueBrain> for now, goodnight!
23:25:21 *** tokai has quit IRC (Ping timeout: 480 seconds)
23:30:00 *** ChanServ sets mode: +v tokai
23:36:52 *** tokai|noir has quit IRC (Ping timeout: 480 seconds)
continue to next day ⏵
