00:04 <+bridge> <gorp_tw> Is there no way to request that the server not keep track of what you were spectating?
00:04 <+bridge> <gorp_tw> I personally find the "sticky" behavior of specing annoying, I would rather fallback to free view every time I unspec
00:05 <+bridge> <gorp_tw> it's especially annoying with dummy since often you forget after a while that dummy was specing something, and then when you spec you can't see your tee so you don't know if you're changing spec for dummy or not and its just sodifghnsfoaihnsfao
00:05 <+bridge> <12944qwerty> You can bind +spectate when you click it and i think it resets
00:05 <+bridge> <12944qwerty> Iirc
00:05 <+bridge> <gorp_tw> click it?
00:06 <+bridge> <gorp_tw> seems like i need a on/off bind for this?
00:06 <+bridge> <12944qwerty> spectate-1
00:06 <+bridge> <12944qwerty> spectate -1
00:06 <+bridge> <12944qwerty> MB thats the bind
00:06 <+bridge> <gorp_tw> so i would keep track of spectating state
00:07 <+bridge> <gorp_tw> and when i press it OFF i would do that
00:07 <+bridge> <gorp_tw> seems kinda hacky
00:07 <+bridge> <gorp_tw> i guess it makes little difference if it always does it
00:07 <+bridge> <12944qwerty> https://cdn.discordapp.com/attachments/293493549758939136/1476717357853577348/IMG_8341.png?ex=69a223b6&is=69a0d236&hm=25d8406e74fa29370ecc56b31244d433ebbe0fda4ece253cc1a713c26cbb7292&
00:07 <+bridge> <12944qwerty> Where i found this lol
00:07 <+bridge> <gorp_tw> i see
00:07 <+bridge> <12944qwerty> Idk I've never tried it so idk if it does what you want
00:08 <+bridge> <gorp_tw> ill try
00:09 <+bridge> <gorp_tw> omg thank god
00:09 <+bridge> <gorp_tw> thanks
02:46 <+bridge> <sans._.> https://cdn.discordapp.com/attachments/293493549758939136/1355599450038468821/tee.png?ex=69a1d8de&is=69a0875e&hm=c15e76c119bb9ef82f4ea447a0b8306f871f071f709e33d8b93b98221f19ca2d&
04:20 <+bridge> <12944qwerty> disco lmao
04:20 <+bridge> <12944qwerty> https://cdn.discordapp.com/attachments/293493549758939136/1476781071898116137/Video_2026-02-26_21-20-19.mp4?ex=69a25f0c&is=69a10d8c&hm=fa16b0213e583b213a3a916db3c0a815ce388fcf27a8b06f835c8595075b4b29&
04:20 <+bridge> <12944qwerty> disco lmao
04:21 <+bridge> <12944qwerty> https://cdn.discordapp.com/attachments/293493549758939136/1476781233899044884/Video_2026-02-26_21-21-06.mp4?ex=69a25f33&is=69a10db3&hm=1b129cda4cd10f8a0f8470760cc959781c050f948ff4b4760906cc7f459de1e4&
04:26 <+bridge> <12944qwerty> ig i'm not finishing tonight. @essigautomat am I not able to use the image texture for unbuffered? I'm looking at the speedup code and it just manually rotates each arrow. so a little confused rn
10:13 <+bridge> <Assa> for the unbuffered code you need to color each tile individually, like you did before. Yes you can't use the buffer image for the unbuffered case :owo:
11:05 <+ChillerDragon> omg aintnoway
11:06 <+ChillerDragon> after all these years i got trolled big time while passing `Var++` as argument to a method and expecting to receive an incremented value on the callsite xd
11:07 <+ChillerDragon> oke ima craft quiz
11:08 <+ChillerDragon> wat dis print
11:08 <+ChillerDragon> `int Six = 6;printf("%d %d\n", Six, Six++);`
11:12 <+bridge> <aegisub> UB
11:52 <+bridge> <aegisub> ok no it should just be unspecified, not undefined
11:52 <+bridge> <aegisub> x++ + ++x is undefined
12:08 <+bridge> <Assa> this guy was against my proposed refactor for a similar situation involving post increments ...
12:37 <+bridge> <chillerdragon> Haha fair
12:54 <+bridge> <fokkonaut> ++x increments and returns, x++ increments after usage, or am i missing smth?
12:56 <+bridge> <chillerdragon> yes exactly
12:56 <+bridge> <chillerdragon> its pretty basic thats why i raged so hard when i realized this was my bug root cause :D
12:58 <+bridge> <fokkonaut> 😄
13:12 <+bridge> <learath2> "returns" is the key issue people have with this, these statements don't really "return" in the normal sense, they guarantee that something will happen before the next sequence point
13:15 <+bridge> <learath2> For this specific case the issue is that the order of evaluation of function arguments is not specified, so there are no sequence points there.
13:16 <+bridge> <learath2> It is UB btw, it's a `value computation` that is `unsequenced relative to` a `side effect` on the same `scalar object`
13:23 <+bridge> <aegisub> ok nice my first gut feeling was corect
13:23 <+bridge> <aegisub> thanks
13:23 <+bridge> <learath2> Maybe if you remember the older C99 wording it is kinda harder to say that it's UB
13:24 <+bridge> <learath2> Maybe if you remember the older C99 wording it is kinda harder to say that it's UB under that
13:24 <+bridge> <learath2> (but it is, between two sequence points the object is being modified once, but is read twice, one of those reads is not for the purpose of determining the value to be stored, thus UB)
14:29 <+bridge> <chillerdragon> @gorp_tw  how good is your lua lsp working?
14:30 <+bridge> <chillerdragon> i feel like mine is full broken but not sure if that is just the state of tooling in lua
14:30 <+bridge> <chillerdragon> i can not autocomplete things defined in other files smh
14:55 <+bridge> <chillerdragon> woah `git init` drastically helped my lsp xd
21:04 <+bridge> <milkeeycat> @robyt3 it's possible to fix windows compilation error with this change https://github.com/MilkeeyCat/ddnet/commit/5920d7adcad23e5d3f0bc2e0e6d1b592cceec8bb
21:05 <+bridge> <milkeeycat> https://sourceware.org/binutils/docs/as/i386_002dOptions.html
21:05 <+bridge> <milkeeycat> > On PE/COFF target this option forces the use of big object file format, which allows more than 32768 sections.
21:09 <+bridge> <robyt3> The testrunner also seems to be crashing though
21:11 <+bridge> <robyt3> (which should not be caused by the intentional crash I added for testing)
21:12 <+bridge> <milkeeycat> I'm not sure where I should look, can you send a link?
21:13 <+bridge> <gorp> @chillerdragon:matrix.org its pretty not working to the point that I disabled it
21:14 <+bridge> <robyt3> It's in the runs for the commit you sent. We don't get the stack trace for crashes unfortunately. Running gdb seems almost impossible due to argument escaping when I tried.
21:49 <+bridge> <12944qwerty> For buffered, how does someone customize the image within the game. The string for thr image for speedup and tune are hardcoded so
22:08 <+bridge> <milkeeycat> welp, even with the backtrace, it's not clearer why it segfaults xd
22:08 <+bridge> <milkeeycat> https://cdn.discordapp.com/attachments/293493549758939136/1477049837227872509/image.png?ex=69a3595b&is=69a207db&hm=f706ec6fcd66fc1182d6280eebfb99946abc0cf3881c4b2794bf1169f12bafd2&
22:32 <+bridge> <robyt3> It seems to crash in one of the score tests (SingleScore)
22:48 <+bridge> <milkeeycat> I think it crashes in gtest initialization code, it probably doesn't even reach our test code