00:07:24 <Eksell> Unstable branch on CRAWL.XTAHUA.COM updated to: 0.22-a0-367-gb72a975 (34)
01:24:38 <Gretell> Unstable branch on crawl.develz.org updated to: 0.22-a0-367-gb72a975 (34)
01:57:01 <Gretell> Windows builds of master branch on crawl.develz.org updated to: 0.22-a0-367-gb72a975
02:53:04 <Gretell> Monster database of master branch on crawl.develz.org updated to: 0.22-a0-367-gb72a975
03:14:27 <Rotatell> Unstable branch on crawl.beRotato.org updated to: 0.22-a0-367-gb72a975 (34)
05:25:50 -!- amalloy is now known as amalloy_
10:24:22 <advil> wow, positivity on the internet
13:10:24 <Henzell> Unstable branch on crawl.akrasiac.org updated to: 0.22-a0-367-gb72a975 (34)
15:57:10 -!- amalloy_ is now known as amalloy
16:02:38 -!- amalloy is now known as amalloy_
16:04:41 <NormalPerson7> aidanh, you know your suggestion about reading the options file into a string? i can do that except i dunno how to read a file to a string in c++, can you tell me/advise me whether crawl code has things like that already built in or whether it's necessary to use a standard C++ method?
17:05:42 <gammafunk> advil: btw, did you ever see https://crawl.develz.org/mantis/view.php?id=11432 ; I was vaguely wondering if any recent ghost file changes you made might cause a duplication
17:05:58 <gammafunk> for this game in the report I think we saw that the player only had one death on that floor with that combo
17:06:08 <advil> yeah, I did see that and couldn't figure out an obvious reason, but it did clearly seem to be a duplicate
17:06:24 <gammafunk> ok
17:06:48 <advil> it's on my list to look at again...it's a bug that I wouldn't easily notice because I'm usually testing this stuff with 8 identical ghosts created for convenience
17:07:16 <gammafunk> heh
17:07:52 <gammafunk> you are probably also aware of https://crawl.develz.org/mantis/view.php?id=10318 ; that webtiles persistent window bug
17:08:01 <gammafunk> sadly I've had that happen multiple times recently
17:08:05 <gammafunk> I guess the fix didn't cover all cases
17:08:08 <advil> yes
17:08:36 <advil> the window thing is a related but distinct bug, I'm pretty sure
17:09:00 <advil> I've tried quite hard to reproduce it and couldn't so far, had multiple computers next to each other trying to mess with the timing of when I click join and so on
17:09:44 <gammafunk> interesting
17:10:17 <advil> more detailed discussion here https://crawl.develz.org/tavern/viewtopic.php?f=8&t=25255&p=328167#p328167
17:10:45 <advil> what server did you encounter it on?
17:12:04 <advil> also, were there a lot of people in chat when it happened?
17:13:00 <advil> I also need to try in chrome, since one reporter confirmed it definitely happens in chrome (I have only tried in firefox)
17:14:25 <gammafunk> advil: cbro and using chrome
17:14:32 <gammafunk> you can actually see it in my vod, maybe I can find it
17:14:36 <gammafunk> no idea if it would be helpful
17:15:10 <advil> if it happens again you could load up developer tools in your stream ;-)
17:15:37 <gammafunk> yeah, and funny you mention this
17:15:49 <gammafunk> but I used developer tools to remove the chat element, I now realize :(
17:15:55 <gammafunk> so it's less useful to you probably
17:16:07 <advil> oh huh
17:16:07 <gammafunk> but it still happens even with that div chat element removed through developer tools
17:16:12 <advil> that's interesting
17:16:16 <gammafunk> maybe it happened again afterwards and I hadn't removed though
17:16:23 <gammafunk> since I usually don't after I reload the page
17:16:26 <gammafunk> let me find
17:17:13 <advil> I expect it's a similar bug (parent scoped variable in a closure where a selector should be used), but even knowing that I haven't been able to deduce exactly where it is analytically, so I really need to figure out how to reproduce it to make progress
17:19:08 <advil> wow, that first bugfix was almost exactly a year ago
17:19:31 <gammafunk> advil: ok, I made this one unlisted (was scheduled to be published), but you can see it happening a few seconds after here: https://youtu.be/xJgKkrL8W4Q?t=969
17:20:01 <gammafunk> it unfortunately does not show the chat spectator count since that div is removed
17:21:17 <gammafunk> running a very recent chrome on ubuntu (chrome gets updated regularly by system update)
17:21:29 <advil> I guess the thing in the LR is twitch chat?
17:21:46 <gammafunk> yes
17:22:00 <gammafunk> danggurl mentions how the window got frozen
17:22:13 <gammafunk> oh yeah and it shows me re-deleting the chat element
17:22:40 <gammafunk> but I am just obs window capturing my spectating webtiles
17:22:44 <gammafunk> I'm actually playing on console
17:22:50 <advil> ah
17:22:51 <gammafunk> which you can see part of in the UR
17:23:12 <advil> you know it should be possible to add a chat hide rc setting
17:23:40 <advil> might be easier :)
17:23:57 <advil> it'd require some server updates but I could lump it in with the mute stuff
17:24:38 <gammafunk> heh, yeah that would be nice
17:25:19 <gammafunk> I usually only have to do this once at start of stream though
17:25:44 <gammafunk> well, unless that bug happens
17:26:01 <gammafunk> hrm, actually the way I do things is I spectate myself twice
17:26:14 <gammafunk> once with a specific window to capture and once so I can see webtiles if I want
17:26:17 <gammafunk> and also see chat
17:26:48 <gammafunk> otoh it's not really critical to see webtiles chat when I stream
17:27:17 <advil> could also imagine having a minimize thing that turns it into a tiny icon or something
17:27:28 <gammafunk> ah, that would probably be better
17:27:46 <advil> probably some players find it distracting and wouldn't mind the feature, independent of streaming
17:28:03 <gammafunk> yeah, if it were per-windows I think it would be better
17:28:07 <gammafunk> *per-window
17:28:11 <advil> yeah
17:40:05 <Silurio_> Hey people, what do you think about making warper a ranged background? would make the spells in there much more usefull
17:43:20 <gammafunk> well there's already a spell-based ranged background, and it's relatively easy to find a launcher (or use throwing) for use with warper spells
17:44:29 <gammafunk> the warper spells are also useful already if you're not using ranged attacks
17:47:42 <Silurio_> Yeah, but Ranged could use some love, and it would give newbies a good idea on how to exploit the power of the spells in the book
17:47:59 <Silurio_> Well, some attention more than love, ranged is pretty damn strong
18:13:05 <gammafunk> ranged has issues that are unrelated to the warper book, but AM is the start where newbies can learn about spells and ranged
20:38:19 <twelwe> ok back
21:13:13 <advil> if two websocket messages (from crawl process to browser) are very closely timed, what would it take for them to be reordered? It seems like that should be impossible without something intercepting them, but I know servers sometimes run reverse proxies for webtiles, would they ever do that?
21:14:05 <advil> my current best theory of this menu bug requires messages being reordered
21:44:34 <gammafunk> advil: well, cbro probably doesn't have a reverse proxy
21:44:46 <gammafunk> I think it's simply listening on 8080 directly
21:44:58 <advil> hm
21:45:23 <gammafunk> I haven't looked too carefully at the client's "comm" library (what it calls its websocket message handling)
21:45:44 <gammafunk> but since that's just a function call based on the type of message, I don't think any change of order would happen through the client itself
21:46:44 <gammafunk> I wonder what would help debug this
21:47:04 <gammafunk> just logging all websocket json recieved by the client to a file?
21:47:18 <gammafunk> I guess if you don't know that the bug is going to trigger, then it's hard for this to be useful
21:47:56 <gammafunk> it seems to happen maybe every other stream or something like that, so it's something that's unlikely to come up frequently unless maybe you have a lot of spectators
21:49:03 <advil> yeah, it definitely seems to require *exact* timing
21:49:34 <advil> your video was fairly helpful in that it gave me a completely different theory than I was following before
21:49:54 <gammafunk> oh? you were able to discern something based on the timing of the commands?
21:49:59 <Cheibriados> 03advil02 07* 0.22-a0-368-g870eee8: Ensure that close_all_menus syncs with webtiles clients 10(6 minutes ago, 2 files, 11+ 0-) 13https://github.com/crawl/crawl/commit/870eee8612f8
21:50:17 <advil> yeah, actually because you didn't notice it for a while so I saw how it interacted with further keys
21:50:25 <gammafunk> ah, yeah
21:50:35 <gammafunk> yeah, I can't actually notice it unless someone in chat tells me
21:50:47 <gammafunk> since I'm just looking at console and not at the window I capture
21:50:52 <advil> I'm not really sure if that will work, but if it doesn't, we'll know more
21:50:59 <gammafunk> which is kind of bad since it can happen for a while if no one tells me
21:53:58 <advil> well, let me know if 870eee8 doesn't work
22:09:55 <Jorgrell> Unstable branch on crawl.jorgrun.rocks updated to: 0.22-a0-368-g870eee8612 (34)
23:59:25 <Eksell> Unstable branch on CRAWL.XTAHUA.COM updated to: 0.22-a0-368-g870eee8 (34)