⏴ go to previous day
00:07:56  <nielsm> recompiling ottd master in about 30 seconds: https://0x0.st/io4V.png

00:08:24  <nielsm> (debug build)

00:09:52  *** Wolf01 has quit IRC

00:10:06  <nielsm> hm

00:10:16  <LordAro> nielsm: nice.

00:10:29  <nielsm> and less than 2 minutes for a release build

00:10:30  <LordAro> i should time compiling OTTD on our 40 core machine at work :p

00:10:55  <nielsm> that's only ottd itself, not including language packs or the tools

00:11:23  <milek7> why so long for release?

00:11:30  <nielsm> link time codegen

00:11:40  <milek7> with lto?

00:11:48  <LordAro> clearly we should turn that on for linux builds

00:11:50  <nielsm> yes

00:12:04  <LordAro> gotta get dem marginal gains

00:12:43  <nielsm> building the objects in release mode is just as fast as building them in debug mode, it's only the link that adds the extra minute+

00:13:29  <nielsm> and the basesets generating is painful :)

00:13:51  <nielsm> entirely serial, and it's a slow vbscript

00:14:28  <milek7> hm, 38s with --enable-lto

00:14:49  <milek7> same cpu

00:15:11  <LordAro> i'm looking at #6515. I've modified regression to test it, and it seems that engine + wagon - both valid vehicles, but adding an extra wagon means that next one is an invalid vehicle

00:15:56  <LordAro> it's weird

00:16:09  <_dp_> https://i.imgur.com/XOcJnw5.png :p

00:20:08  <nielsm> gn

00:24:43  <glx> baseset stuff is way faster in cmake

00:27:29  <LordAro> Message:   Assertion failed at line 85 of C:/msys64/home/LordAro/OpenTTD/src/fontcache.cpp: IsPrintable(key)

00:27:32  *** nielsm is now known as Guest16488

00:27:45  <LordAro> i do hate it when OTTD crashes when i'm trying to do something else

00:28:11  *** Guest16488 has quit IRC

00:50:18  <LordAro> so... there are tabs in a print statement in the regression test output

00:50:34  <LordAro> if you open the ai debug window soon enough, OTTD crashes when trying to print that character

00:50:43  <LordAro> how has no one found this before?!

00:51:18  <glx> luck ?

00:51:18  <_dp_> LordAro, ppl don't usually debug AI's on debug build?

00:52:04  <LordAro> _dp_: but even release builds have had asserts turned on since 1.9

00:52:13  <glx> anyway tab should be filtered out

00:53:30  <_dp_> LordAro, huh? why? there's some quite slow stuff in asserts

00:53:39  <_dp_> like k-d tree invariant check

00:53:51  <LordAro> because no one thought to do it

00:53:57  <LordAro> probably should be done, tbh

00:54:06  <glx> I guess pressing reload AI should be easier than opening the window very fast

00:54:57  <LordAro> glx: open the window before `startai <foo>` :p

00:55:05  <glx> yeah that too

00:55:09  <_dp_> ah, nvm, kdtree has it's own debug var

00:55:36  <LordAro> hmm, it is inside the FallbackParagraphLayout

01:00:22  <LordAro> in fact, not sure why it's even using the fallback, configure says it found uniscribe...

01:01:26  <glx> just need an AI printing tab ?

01:03:03  <glx> baseset metadata step is so slow

01:04:39  <LordAro> glx: yeah, this is my main.nut: https://pastebin.com/6LUQb947

01:04:55  <LordAro> (except a tab in between '', obviously)

01:05:29  *** tokai|noir has joined #openttd

01:05:29  *** ChanServ sets mode: +v tokai|noir

01:05:31  <glx> yeah I modified a test AI to add a tab

01:05:40  <glx> still waiting for compilation :)

01:06:20  <glx> I really need to add more RAM

01:06:31  <glx> 8GB is clearly not enough

01:07:34  <glx> hmm MSVC not responding

01:12:21  *** tokai has quit IRC

01:18:02  <LordAro> oh i see, uniscribe layouter is returning null

01:18:50  <glx> no crash for me with msvc

01:19:33  <LordAro> glx: perhaps uniscribe is working for you?

01:19:47  <glx> let me check with mingw

01:26:12  <LordAro> i'm reasonably certain that FallbackParagraphLayout::NextLine is the one at fault - it happily passes unprintable characters through to FallbackVisualRun, which then explodes

01:28:39  <glx> no crash with mingw either

01:29:03  <LordAro> you definitely have a tab character in your ai?

01:29:18  <glx> yes

01:29:43  <LordAro> is yours actually using uniscribe though?

01:29:56  <LordAro> mine is supposed to be able to, but it's returning null for some reason

01:30:09  *** WormnestAndroid has quit IRC

01:30:35  *** WormnestAndroid has joined #openttd

01:31:45  *** Wormnest has joined #openttd

01:36:11  <glx> maybe also depends on the font

01:36:57  <LordAro> opengfx, otherwise standard

01:37:12  <glx> so sprite font

01:37:57  <LordAro> yeah

01:39:43  <LordAro> but i can definitely see the issue in the code - '\t' gets passed into DrawString, never gets filtered out before FallbackVisualRun constructor calls SpriteFontCache::MapCharToGlyph on the tab character

01:41:54  <glx> ok without specifying a font in cfg it crashes

01:42:03  <LordAro> :)

01:42:17  <glx> in msvc

01:42:27  <glx> won't retry with mingw, will be the same

01:42:37  <LordAro> ah, so uniscribe must be rejecting the string as well

01:42:47  *** Progman has quit IRC

01:42:55  <glx> with a font it's accepted

01:45:25  <LordAro> i imagine most fonts know what to do with a tab :)

01:51:48  <glx> and it's quite impossible to use breakpoint to test, as layouter is called from many other places

01:52:06  <LordAro> that's what i found too :)

01:52:31  <LordAro> the most luck i had was a conditional breakpoint with a string of the right length

01:52:49  <LordAro> MSVC might be able to do a string literal better

01:52:52  <glx> hmm I could breakpoint on the AI print

01:53:06  <glx> and follow

01:53:50  <LordAro> it's not the ai print, it's rendering it in the window

02:05:07  <glx> indeed uniscribe pass with a font

02:07:45  <glx> with a breakpoint in AI debug window it's doable

02:15:27  <glx> I guess GetLayouter() should handle tab as it already handles RTL stuff

02:30:24  *** spnda has quit IRC

02:45:42  <glx> if I ignore the assert it prints a '?' for the tab

03:11:33  <glx> LordAro: https://github.com/OpenTTD/OpenTTD/compare/master...glx22:tabs <-- I get the same output as without this patch when using a font, and sprite font works

03:41:51  *** gnu_jj has quit IRC

03:42:05  *** gnu_jj has joined #openttd

03:49:14  *** tokai has joined #openttd

03:49:14  *** ChanServ sets mode: +v tokai

03:52:06  *** Lejving_ has quit IRC

03:55:51  *** Wormnest has quit IRC

03:55:53  *** tokai|noir has quit IRC

04:26:31  *** D-HUND has joined #openttd

04:29:56  *** debdog has quit IRC

04:40:39  *** tokai|noir has joined #openttd

04:40:39  *** ChanServ sets mode: +v tokai|noir

04:47:36  *** tokai has quit IRC

04:57:48  *** glx has quit IRC

05:27:59  *** tokai has joined #openttd

05:27:59  *** ChanServ sets mode: +v tokai

05:34:52  *** tokai|noir has quit IRC

06:19:09  *** Arveen has quit IRC

06:19:36  *** Arveen has joined #openttd

08:36:52  *** Lejving has joined #openttd

09:04:32  *** Progman has joined #openttd

09:21:31  *** sla_ro|master has joined #openttd

09:30:13  *** WormnestAndroid has quit IRC

09:30:28  *** WormnestAndroid has joined #openttd

09:50:08  *** nielsm has joined #openttd

10:29:15  *** Wolf01 has joined #openttd

11:00:22  <Wolf01> Eddi|zuHause: did you try csur reborn for C:S?

11:33:04  *** adikt has quit IRC

11:50:53  <Eddi|zuHause> no, what's that?

11:51:15  *** Wormnest has joined #openttd

11:51:31  <Wolf01> Road pack with intersections, splits/joins, bus lanes, asymmetric town roads etc

11:51:49  <Wolf01> https://steamcommunity.com/sharedfiles/filedetails/?id=1959216109

11:52:27  <Wolf01> https://bbs.csur.fun/d/62-csur-steam-workshop-collection-description ok, maybe the english description is better :P

12:11:03  <Eddi|zuHause> that explains why i wouldn't have noticed it :p

12:23:32  *** Samu has joined #openttd

12:23:40  <Samu> hi

12:25:52  <Eddi|zuHause> i'm using the NEXT mod for asymmetric roads

12:43:54  *** Wormnest has quit IRC

12:49:25  <DorpsGek_III> [OpenTTD/OpenTTD] junaidnaseer commented on issue #7772: Showing destination when stopped or waiting https://git.io/Je4C2

12:51:59  <DorpsGek_III> [OpenTTD/OpenTTD] LordAro commented on issue #7772: Showing destination when stopped or waiting https://git.io/Je4C2

12:55:55  <DorpsGek_III> [OpenTTD/OpenTTD] LordAro opened pull request #8002: Doc: [Script] Add a note about how wagon connectivity works for scripts https://git.io/Jvl9G

13:04:58  <DorpsGek_III> [OpenTTD/OpenTTD] junaidnaseer commented on issue #7772: Showing destination when stopped or waiting https://git.io/Je4C2

13:06:20  <DorpsGek_III> [OpenTTD/OpenTTD] LordAro commented on issue #7772: Showing destination when stopped or waiting https://git.io/Je4C2

13:08:04  <DorpsGek_III> [OpenTTD/OpenTTD] aruscher commented on issue #7772: Showing destination when stopped or waiting https://git.io/Je4C2

13:17:27  <DorpsGek_III> [OpenTTD/OpenTTD] junaidnaseer commented on issue #7772: Showing destination when stopped or waiting https://git.io/Je4C2

13:17:44  *** Wormnest has joined #openttd

13:20:31  <DorpsGek_III> [OpenTTD/OpenTTD] aruscher commented on issue #7772: Showing destination when stopped or waiting https://git.io/Je4C2

13:34:18  <DorpsGek_III> [OpenTTD/OpenTTD] LordAro opened pull request #8003: Fix #6399: Create parent directories if they don't already exist https://git.io/Jvl9x

13:34:30  <LordAro> oops, created 2 branches in the main repo, not my fork

13:34:42  <LordAro> one of which is now linked to a PR :)

14:05:14  <TrueBrain> Silly boy

14:05:26  <LordAro> :(

14:06:34  <TrueBrain> Nothing wrong with being silly :)

14:06:46  <TrueBrain> Problem fixes itself if you merge those PRs :p

14:10:03  *** Wormnest has quit IRC

14:19:44  <LordAro> TrueBrain: exactly!

14:21:31  *** andythenorth has joined #openttd

14:21:55  <andythenorth> AWS emailed me to say I'm about to hit limit on free tier

14:21:57  <andythenorth> oof

14:25:02  *** spnda has joined #openttd

14:44:40  <peter1138> Are you?

14:44:52  <peter1138> What are you having served?

14:55:04  *** glx has joined #openttd

14:55:04  *** ChanServ sets mode: +v glx

15:04:09  <DorpsGek_III> [OpenTTD/OpenTTD] James103 commented on pull request #8003: Fix #6399: Create parent directories if they don't already exist https://git.io/JvlQQ

15:04:10  <DorpsGek_III> [OpenTTD/OpenTTD] glx22 commented on pull request #8002: Doc: [Script] Add a note about how wagon connectivity works for scripts https://git.io/JvlQ7

15:05:16  <andythenorth> https://firs-test-1.s3.eu-west-2.amazonaws.com/iron-horse/docs/html/tech_tree_table_red.html

15:05:19  <andythenorth> and other docs

15:05:32  <andythenorth> I think it's the number of uploads I've done, but I didn't look further

15:05:43  <glx> oh a comment from james

15:05:52  <andythenorth> 17,338 Requests	limit is 20,000 Get Requests of Amazon S3

15:06:09  <andythenorth> hmm GET suggests serve, not upload, upload would probs be POST

15:06:27  <andythenorth> I could just....pay? :P

15:09:36  <Eddi|zuHause> technically you can use "GET" to upload

15:10:05  <Eddi|zuHause> you just have to encode the file you want to upload into the URL

15:12:03  <DorpsGek_III> [OpenTTD/OpenTTD] James103 commented on pull request #8003: Fix #6399: Create parent directories if they don't already exist https://git.io/JvlQp

15:16:16  <DorpsGek_III> [OpenTTD/OpenTTD] glx22 commented on pull request #8003: Fix #6399: Create parent directories if they don't already exist https://git.io/Jvl7k

15:16:50  <DorpsGek_III> [OpenTTD/OpenTTD] LordAro commented on pull request #8003: Fix #6399: Create parent directories if they don't already exist https://git.io/Jvl7I

15:19:59  <DorpsGek_III> [OpenTTD/OpenTTD] LordAro commented on pull request #8002: Doc: [Script] Add a note about how wagon connectivity works for scripts https://git.io/Jvl7O

15:21:33  <LordAro> glx: i don't have the ability to test with MSVC (or MacOS), can you try it out and give me a clue?

15:21:48  <glx> I was about to do it

15:21:51  <LordAro> :)

15:21:54  <glx> I cancelled the run btw

15:22:00  <glx> no need to wait an hour :)

15:22:03  <LordAro> i saw :)

15:27:43  *** spnda has quit IRC

15:28:16  <glx> hmm I guess I'll need to try without any openttd.cfg :)

15:35:56  <glx> ok https://pastebin.com/vCgsFJvp

15:42:21  <LordAro> ah

15:42:42  <LordAro> i did make the assumption that closedir acted like free

15:42:43  <glx> I guess it also fails there for macos

15:44:07  *** Flygon has quit IRC

15:53:53  <DorpsGek_III> [OpenTTD/OpenTTD] LordAro updated pull request #8003: Fix #6399: Create parent directories if they don't already exist https://git.io/Jvl9x

15:56:57  <DorpsGek_III> [OpenTTD/OpenTTD] LordAro commented on pull request #8003: Fix #6399: Create parent directories if they don't already exist https://git.io/Jvl7j

16:01:59  <glx> crash is gone for me

16:02:09  *** Wormnest has joined #openttd

16:05:28  <DorpsGek_III> [OpenTTD/OpenTTD] stormcone opened pull request #8004: Fix: [SDL2] support pasting from clipboard on Linux https://git.io/Jvl5G

16:06:16  <DorpsGek_III> [OpenTTD/OpenTTD] glx22 approved pull request #8003: Fix #6399: Create parent directories if they don't already exist https://git.io/Jvl5Z

16:09:02  *** frosch123 has joined #openttd

16:33:18  *** Wormnest has quit IRC

16:37:09  <glx> oh I see why it hanged, it's because message boxes

16:37:41  <glx> crash window is not displayed for null video driver, but we still show message box

16:38:49  <glx> hmm no it's not that

16:39:14  *** adikt has joined #openttd

16:41:42  <glx> ah no crash happens before video driver init, so crashlog window is displayed inconditionaly

16:53:44  *** WormnestAndroid has quit IRC

16:53:50  *** WormnestAndroid has joined #openttd

17:13:53  <DorpsGek_III> [OpenTTD/OpenTTD] JGRennison commented on issue #6472: Cargodist: various link graph performance improvements https://git.io/fhn0p

17:15:40  <_dp_> wish someone reduced cargodist memory usage

17:15:54  <_dp_> especially when the damn thing isn't even enabled

17:17:29  *** gelignite has joined #openttd

17:21:12  <DorpsGek_III> [OpenTTD/OpenTTD] nielsmh closed issue #6472: Cargodist: various link graph performance improvements https://git.io/fhn0p

17:33:07  <DorpsGek_III> [OpenTTD/OpenTTD] junaidnaseer commented on issue #7772: Showing destination when stopped or waiting https://git.io/Je4C2

17:40:50  <DorpsGek_III> [OpenTTD/OpenTTD] glx22 commented on issue #7772: Showing destination when stopped or waiting https://git.io/Je4C2

17:41:14  <andythenorth> hmm the speed / destination display

17:41:39  <andythenorth> should be '[speed | stopped]  + destination'

17:41:53  <andythenorth> solves other issues

17:41:57  <andythenorth> no need for hover crap :)

17:42:34  <andythenorth> oof Horse 99%

17:42:50  <glx> only 99% left then :)

17:43:28  <andythenorth> hmm

17:43:36  <andythenorth> no last 10% is 50% of the work

17:43:54  <andythenorth> @calc 0.9 * 50

17:43:54  <DorpsGek> andythenorth: 45

17:43:58  <andythenorth> 5%

17:44:08  <andythenorth> lol the maths I should be able to in my head :P

17:54:43  *** Arveen has quit IRC

18:13:07  *** Smedles has quit IRC

18:13:45  *** Smedles has joined #openttd

18:45:37  <DorpsGek_III> [OpenTTD/OpenTTD] LordAro requested changes for pull request #8004: Fix: [SDL2] support pasting from clipboard on Linux https://git.io/JvlA3

19:08:11  <DorpsGek_III> [OpenTTD/OpenTTD] glx22 commented on pull request #8004: Fix: [SDL2] support pasting from clipboard on Linux https://git.io/JvlA9

19:21:00  *** adikt has left #openttd

19:21:00  *** adikt has joined #openttd

19:25:08  *** WormnestAndroid has quit IRC

19:29:10  *** WormnestAndroid has joined #openttd

19:34:29  *** Arveen has joined #openttd

19:37:14  *** WormnestAndroid has quit IRC

19:45:50  <DorpsGek_III> [OpenTTD/OpenTTD] DorpsGek pushed 1 commits to master https://git.io/JvlxN

19:45:50  <DorpsGek_III>   - Update: Translations from eints (by translators)

20:13:51  *** sla_ro|master has quit IRC

20:14:11  <LordAro> glx: you going to PR the printable fix?

20:14:18  <LordAro> i can approve & merge that one :)

20:21:07  *** WormnestAndroid has joined #openttd

20:29:37  <DorpsGek_III> [OpenTTD/OpenTTD] stormcone updated pull request #8004: Fix: [SDL2] support pasting from clipboard on Linux https://git.io/Jvl5G

20:38:17  <DorpsGek_III> [OpenTTD/OpenTTD] LordAro approved pull request #8004: Fix: [SDL2] support pasting from clipboard on Linux https://git.io/JvlhR

20:38:31  <DorpsGek_III> [OpenTTD/OpenTTD] LordAro merged pull request #8004: Fix: [SDL2] support pasting from clipboard on Linux https://git.io/Jvl5G

20:56:08  <DorpsGek_III> [OpenTTD/OpenTTD] glx22 opened pull request #8005: Fix: Ignore not printable characters when layouting a string https://git.io/Jvlh7

20:56:54  <DorpsGek_III> [OpenTTD/OpenTTD] LordAro approved pull request #8005: Fix: Ignore not printable characters when layouting a string https://git.io/JvlhN

21:20:26  *** Webster has quit IRC

21:27:28  <andythenorth> come back webster

21:30:43  *** Webster has joined #openttd

21:34:00  <andythenorth> like magic

21:48:03  <andythenorth> oof Horse 100%

21:48:12  <andythenorth> ship, or redesign?

21:50:40  <LordAro> NRT docs :p

21:51:02  <andythenorth> I got stuck

22:03:15  <andythenorth> I started going down this in order from the top https://wiki.openttd.org/NewGRF_Specification_Status

22:04:46  <LordAro> where did you get stuck?

22:05:55  * andythenorth looks

22:06:47  <andythenorth> oh RV var 4A

22:06:55  <andythenorth> I've fucked up nml and I got sad and quit

22:07:05  <andythenorth> it should never have been done this way

22:07:07  <andythenorth> https://github.com/OpenTTD/nml/blob/master/nml/actions/action2var_variables.py#L197

22:07:16  <andythenorth> it's a right mess tbh

22:07:40  <andythenorth> var 4A now has two meanings

22:13:21  <DorpsGek_III> [OpenTTD/OpenTTD] LordAro commented on issue #8001: Ship finds path but reports it is lost https://git.io/JvlVp

22:15:44  <andythenorth> I don't really see a way out except reverting NRT from NML

22:15:46  <andythenorth> and starting it again

22:16:02  <LordAro> would that be the worst thing in the world?

22:16:19  <LordAro> what does the current implementation do, and how is it wrong?

22:16:35  <LordAro> if it's a "can't be used", i have no issues with removing it entirely

22:19:31  <andythenorth> it has two vars resolving to 4A

22:19:36  <andythenorth> which is clearly broken

22:19:48  <andythenorth> current_roadtype and current_tramtype

22:20:10  <andythenorth> the only way I can think of to fix this is to redo NRT splitting trams and RVs entirely

22:21:59  <LordAro> wait, is the issue in NML, or NFO?

22:22:55  *** gelignite has quit IRC

22:23:15  <andythenorth> it's in the implementation

22:23:23  <andythenorth> meaning of var 0x04 varies

22:23:31  <Eddi|zuHause> no, it's in the concept. the implementation is fine

22:23:47  <andythenorth> implementation / spec /s

22:23:54  <andythenorth> what eddi said

22:28:15  <andythenorth> it stems from trams and RVs being same thing, but not same

22:42:08  <andythenorth> I didn't get further with the rest of the spec, this derailed it

22:47:48  *** Samu has quit IRC

22:58:05  <andythenorth> is it fine to just have two var names?

22:59:45  *** WormnestAndroid has quit IRC

23:01:17  *** Samu has joined #openttd

23:01:21  <Samu> hi

23:09:03  *** Montana_ has joined #openttd

23:21:48  *** andythenorth has quit IRC

23:22:06  *** andythenorth has joined #openttd

23:22:27  <andythenorth> I've marked it up as broken here https://wiki.openttd.org/NewGRF_Specification_Status#OpenTTD_1.10

23:22:48  <andythenorth> not sure whether to continue with the rest, or whether NRT just needs reverted

23:23:45  <andythenorth> seems a bit late to remove it from 1.10?

23:33:37  <nielsm> what's the actual behaviour? does it query roadtype or tramtype depending on vehicle type?

23:34:43  *** Samu has quit IRC

23:34:48  * andythenorth looks

23:35:15  <FLHerne> andythenorth: Assuming it does that, I don't really see the problem

23:35:33  <andythenorth> it's a broken spec

23:35:36  <andythenorth> no?

23:35:38  <nielsm> https://github.com/OpenTTD/OpenTTD/blob/master/src/newgrf_engine.cpp#L615-L618

23:35:51  <FLHerne> andythenorth: No

23:36:03  <nielsm> road vehicles get road type, trams get tram type

23:36:13  <nielsm> tramtrack type

23:36:23  <andythenorth> and in the case that it's tram and queries current_roadtype?

23:36:25  <andythenorth> or vice versa?

23:36:29  <andythenorth> the spec returns lies

23:36:51  <nielsm> trams can't query the roadtype, and road vehicles can't query the tram track type

23:36:56  <andythenorth> I did this in a really really stupid way

23:37:23  <andythenorth> if I use current_roadtype in a tram, I'll get the wrong answer

23:37:31  <FLHerne> andythenorth: Then reject `current_roadtype` for trams and vice versa

23:37:45  <andythenorth> in nml

23:37:46  <FLHerne> In nml

23:37:46  <andythenorth> ?

23:37:49  <andythenorth> syntax check?

23:37:55  <nielsm> it's an NML problem, not an ottd/nrt problem

23:38:56  <nielsm> just document it, "trams use current_roadtype to get the current tram track type. deal with it."

23:39:06  <andythenorth> hmm

23:39:21  <andythenorth> easiest would be to agree and move on

23:39:31  *** cHawk has joined #openttd

23:39:32  <andythenorth> I just don't want to be responsible for making nml worse

23:40:01  <FLHerne> andythenorth: It's the same for road_type and tram_type, they're both 0x05?

23:40:12  <FLHerne> It would be hard to make nml worse

23:40:57  <andythenorth> you're right about road_type and tram_type

23:41:02  <andythenorth> that was my fault also

23:41:16  <andythenorth> I can't see any other way to do it because trams are RVs but not RVs

23:41:41  <andythenorth> and I don't know how to implement a syntax checker in nml (or anywhere, let's be honest)

23:45:47  <FLHerne> It's nothing to do with syntax  [/pedantic]

23:46:27  <andythenorth> validation?

23:48:06  *** Wolf01 has quit IRC

23:49:56  * FLHerne tries to decide the right place

23:51:01  <andythenorth> what about authors who need to check the tram type for an RV and vice versa?

23:51:05  <andythenorth> or is that just bollocks?

23:51:27  <FLHerne> Hard to do in action0properties, because you can only tell if it's a tram if misc_flags is set, and that doesn't have to be first

23:51:35  <FLHerne> Why should anyone want to do that? :P

23:52:03  <andythenorth> [shrug emoji]

23:52:08  <andythenorth> I have NFI

continue to next day ⏵