00:19:24 <Henzell> Edwardauron (L10 DsSu) Crash caused by signal #15: Terminated (D:9)
00:20:07 <Henzell> astronautguy (L9 FeWr) ERROR in 'tileweb.cc' at line 222: Socket write error: Resource temporarily unavailable (D:9)
00:20:07 <Henzell> jomj (L9 DrFE) ERROR in 'tileweb.cc' at line 219: Socket write error: Resource temporarily unavailable (D:8)
00:20:08 <Henzell> Halgar (L16 MiBe) ERROR in 'tileweb.cc' at line 222: Socket write error: Resource temporarily unavailable (IceCv)
00:20:08 <Henzell> HawkPuma (L9 OpTm) ERROR in 'tileweb.cc' at line 219: Socket write error: Resource temporarily unavailable (D:6)
00:20:10 <Henzell> Thalamas (L27 GrFi) ERROR in 'tileweb.cc' at line 222: Socket write error: Resource temporarily unavailable (Coc:7)
00:20:13 <Henzell> MuffinSoup (L27 BaCj) ERROR in 'tileweb.cc' at line 222: Socket write error: Resource temporarily unavailable (Hell)
00:20:16 <Henzell> Tossi (L27 GnVM) ERROR in 'tileweb.cc' at line 222: Socket write error: Resource temporarily unavailable (Zig:7)
00:20:18 <Henzell> Dragonsploder (L14 MiFi) ERROR in 'tileweb.cc' at line 219: Socket write error: Resource temporarily unavailable (Lair:6)
01:29:44 <gammafunk> seen when compiling 0.24 on cao
01:29:45 <gammafunk> ui.cc: In member function 'ui::SizeReq ui::Widget::get_preferred_size(ui::Widget::Direction, int)':
01:29:48 <gammafunk> ui.cc:178:10: warning: declaration of 'min_size' shadows a member of 'this' [-Wshadow]
01:29:51 <gammafunk> ui.cc:179:10: warning: declaration of 'max_size' shadows a member of 'this' [-Wshadow]
01:29:57 <gammafunk> not sure if a commit needs to be cherry-picked
01:31:22 <Henzell> Stable (0.24) branch on crawl.akrasiac.org updated to: 0.24-b1-30-g840c9ef
01:45:26 <Henzell> Webtiles server stopped.
01:45:32 <Henzell> Webtiles server started.
01:49:39 <gammafunk> advil: fyi, I set up 0.24 on cao
03:21:02 <Rotatell> Unstable branch on crawl.beRotato.org updated to: 0.25-a0-42-g6c54e25 (34)
03:31:30 <Kelbell> Fork (bcrawl) on crawl.kelbi.org updated to: 0.23-a0-2634-gb4598b3726
03:34:45 <Kelbell> Fork (bcadrencrawl) on crawl.kelbi.org updated to: 0.22.1-165-g67e1fcc289
04:06:46 -!- Tiobot is now known as Guest33089
08:15:05 <Cheibriados> New branch created: pull/1191 (1 commit) 13https://github.com/crawl/crawl/pull/1191
08:15:05 <Cheibriados> 03kitchen-ace02 07https://github.com/crawl/crawl/pull/1191 * 0.25-a0-43-g5d03901: Warn when exploding portal projectiles would hit you 10(8 hours ago, 1 file, 7+ 0-) 13https://github.com/crawl/crawl/commit/5d03901e2657
11:24:00 <Lantell> Stable (0.23) branch on underhound.eu updated to: 0.23.1-89-g277498a001
11:31:06 <advil> for CAO scoring, I'm considering ways of filtering out accounts that haven't been used in a while, anyone have opinions on the criteria?
11:31:36 <advil> I guess my favorite so far is: SELECT count(name) FROM players where last_game_end >= curdate() - interval 1 year or games_played > 50 or games_won > 0;
11:32:31 <advil> it's not a complete solution to CAO scoring by itself, but that one goes from 112915 total unique player names to 38307
11:32:55 <advil> some may find those criteria to steep though I guess
11:32:58 <advil> *too
11:34:10 <aidanh> is there a total playtime field?
11:35:06 <advil> not in the db right now
11:35:19 <advil> one could use score also in there somehow, that might be a proxy for what you
11:35:21 <advil> 're after
11:35:35 <aidanh> games_played > 50 does seem a little steep to me
11:36:08 <aidanh> I guess the idea i'm going for is 'total player investment'
11:36:19 <advil> it would also be possible to generate player pages for players outside the time threshold on a full db rebuild, so they're present, but still exclude those players from all-players (though it may make sense to get rid of all-players)
11:36:57 <advil> games_played > 20 brings that up to 45400
11:38:22 <advil> I guess the idea I had is that someone who hasn't won and hasn't played in a year probably isn't super invested right now, even if they did play many games
11:40:00 <aidanh> just to make sure I understand correctly, the only effect on these players is that their scores won't show up on CAO?
11:41:31 <advil> yeah, in http://crawl.akrasiac.org/scoring/best-players-total-score.html and http://crawl.akrasiac.org/scoring/all-players.html, and potentially they wouldn't have a personal scoring page, depending on how it's implemented
11:42:08 <aidanh> if they become active again, would they be automatically incorporated again?
11:42:43 <advil> yes
11:42:47 <advil> so an example player this would filter is: http://crawl.akrasiac.org/scoring/players/zzzzzzzzzzzzz.html
11:43:14 <Pinkbeast> This isn't an objection but I observe this would filter me
11:43:28 <advil> or another example: http://crawl.akrasiac.org/scoring/players/zzzaaaccc13.html
11:43:43 <advil> Pinkbeast what name do you play under?
11:43:50 <Pinkbeast> "damerell"
11:44:04 <advil> you have 14 wins
11:44:22 <Pinkbeast> Oh, I'm an idiot who can't read. Don't mind me.
11:45:52 <aidanh> right, if that's the case I'd say go for it
11:46:00 <aidanh> 50 is probably fine
11:46:03 <advil> Pinkbeast also 247 games ;-)
11:46:56 <Pinkbeast> I think I got and and or jumbled in my head and reached a silly conclusion
11:47:51 <advil> heh
11:49:19 <shalmezad> Though "and" instead of "or"? (happens a lot to my robotics students)
11:49:22 <shalmezad> *thought
11:52:37 <aidanh> what would be involved in dragging crawlcode kicking and screaming up to c++17?
11:53:11 <aidanh> the code compiles fine, so I imagine the major barrier would potentially be g++ versions on servers?
11:53:16 <geekosaur> backward compatibility; iirc there were reasons... that
11:53:33 <geekosaur> upgrading several of the servers even for c++11 was problematic
11:55:37 <aidanh> guess I'll trawl the logs
12:02:52 <advil> looks like there were a bunch of issues that aren't going to be relevant to c++17
12:02:58 <advil> e.g. it shouldn't be a problem on os x
12:05:09 <advil> cdo is using gcc 4.9.2 so that would need some work (doesn't even fully support c++14)
12:07:02 <advil> haha CAO seems to be 4.7.2 which doesn't even fully support *c++11*
12:10:20 <aidanh> heh that's pretty nuts
12:11:00 <aidanh> I also recently discovered that we only use Wall and not Wextra, which, if you enable, will unleash a veritable torrent of warnings
13:24:34 <Henzell> Unstable branch on crawl.akrasiac.org updated to: 0.25-a0-42-g6c54e25 (34)
13:47:45 <aidanh> gammafunk: how did you build to get those warnings? I can't reproduce them, even with -Wshadow
13:48:20 <gammafunk> aidanh: it was the installation of stable on CAO, I'd need to check the script to see the exact make args
13:48:34 <gammafunk> as advil pointed out earlier, CAO does run a pretty old gcc
13:48:41 <gammafunk> I'll try to find out those args and let you know
13:49:55 <aidanh> is the CAO build setup on github by any chance?
13:50:31 <gammafunk> yes
13:51:04 <gammafunk> https://github.com/neilmoore/dgamelaunch-config/blob/szorg/crawl-build/update-crawl-stable-build.sh#L67
13:52:01 <aidanh> ah, thanks
13:52:10 <aidanh> well nothing jumps out
13:52:41 <aidanh> the ui.cc warning is benign, but it'd be good to have warnings enabled in general
13:53:02 <aidanh> on that note, don't know if you saw, but there are a *lot* of warnings if you enable -Wextra
13:59:38 <Cheibriados> Level annotation "N runed translucent doors" isn't cleared when Xom opens these doors. 13https://crawl.develz.org/mantis/view.php?id=12105 by Yermak
14:02:21 <gammafunk> aidanh: hrm, so that warning just happens with older gcc? I didn't get it when compiling stable on my system
14:02:44 <gammafunk> maybe it's having somewhat different args
14:02:49 <ebering> !blame xom
14:02:50 <Sequell> I pronounce xom... Guilty!
14:03:12 <gammafunk> ah I did debug lite
14:04:31 <aidanh> seems that way; i tested a full build as well, no luck
14:09:03 <floraline> is it time to install the new version?
14:11:32 <advil> ah here is the explanation, behavior of -Wshadow changed in gcc 4.8 to ignore shadowing of function names by non-functions: https://stackoverflow.com/a/15554949/1393974
14:11:47 <advil> floraline yes you can install any time
14:13:06 <advil> realistically those two types are not at all confusable, so that specific shadowing seems not too worrisome
14:16:14 <aidanh> mystery solved
14:27:01 <gammafunk> Napkin: not sure if you'll have time to install 0.24 for tournament, but stone_soup-0.24 is available if so
14:29:27 <gammafunk> johnstein: also letting you know re: stone_soup-0.24 for CBRO
15:15:05 <Cheibriados> fedhas_altar_decomposition doesn't make sense anymore 13https://crawl.develz.org/mantis/view.php?id=12106 by kitchen_ace
15:43:43 <advil> does it need root to install a new version on CDO?
15:46:08 <gammafunk> advil: yes, or at least for the chroot setup with dgl scripts, it needs one with some kind of sudo access
15:46:14 <gammafunk> I think cdo is set up a bit differently though
15:46:30 <gammafunk> even then, I don't think we can edit e.g. webtiles config on cdo
15:46:46 <advil> well cdo doesn't have webtiles so that one would be hard :D
15:46:50 <gammafunk> er, right
15:47:08 <advil> but I know what you mean
15:47:19 <gammafunk> but yeah, there's a chroot which require write access, dgamelaunch menus that requires write access etc
15:48:10 <gammafunk> cdo is a little less important for tournament since it's console-only and low usage
15:48:30 <gammafunk> I think cbro and cwz are the only two major servers we're waiting on
15:55:52 <advil> well, that change got best-players-total-score down to 1.5 minutes, and all-players down to 2 minutes
15:57:26 <advil> maybe I will also make them update only a few times a day
15:59:29 <advil> it's v unsatisfying to make crawl-related commits that chei doesn't show
16:00:05 <advil> I guess anyone else who is using scoring may be interested in the fact that someone is actually making incremental improvements to it again
16:00:46 <advil> e.g. https://github.com/crawl/scoring/commit/99e7973e33b647b9d078ee6fb277d6624446045e
16:01:39 <advil> though that's much less important if you aren't dealing with 25G of logfiles
16:06:33 <Pinkbeast> We try and grab updates to the scoring when we notice them
16:06:47 <Pinkbeast> ... I think we're probably going to score bcrawl as well as us and Gooncrawl
16:06:52 <Pinkbeast> So, thanks.
16:06:54 <advil> it's been a long time since there have been real updates to it I think
16:09:28 <advil> lol at per-day going back to 2006, I had not ever actually scrolled down
16:10:46 <advil> this is cool and all but will anyone be upset if I truncate this page at 1 month?
16:10:53 <advil> http://crawl.akrasiac.org/scoring/per-day.html
16:11:29 <gammafunk> heh, yeah I don't think anyone will be bothered by that
16:11:33 <gammafunk> I'd forgotten about that page
16:12:29 <advil> it's sort of fun but doesn't need to be regenerated from scratch, maybe I'll make a static version of it to link to
16:12:48 <gammafunk> as neat as the first row is, where the only winners in december 2006 where green.snark and someone named jarmok
16:15:34 <amalloy> !gamesby jarmok
16:15:35 <Sequell> jarmok has played 258 games, between 2006-12-19 23:14:21 and 2011-08-28 23:06:48, won 24 (9.3%), high score 11056340, total score 53274451, total turns 6074451, play-time/day 0:25:36, total time 30d+11:33:16.
16:15:48 <advil> that's a pretty good win ratio for that time window
16:15:57 <amalloy> it is
17:13:36 <Cheibriados> Interlevel travel is broken (buggy now). 13https://crawl.develz.org/mantis/view.php?id=12107 by Yermak
18:09:53 <Yermak> Are there reasons against incorporating beem into Webtiles chat? No subscriptions, no spam on bot joining your game.
18:15:33 <alexjurkiewicz> Yermak: it's an interesting idea
18:16:45 <alexjurkiewicz> what version of gcc is required for c17?
18:17:46 <alexjurkiewicz> I could install 5.4 on 16.04 lts
18:24:24 <aidanh> mostly g++ 7
18:27:43 <aidanh> although it looks like there is a ppa for newer versions
18:28:56 <aidanh> clang++ 5 and later also support all of c++17, not sure about distro packages there though
18:33:54 <aidanh> Yermak: that would give webtiles players an unfair advantage
18:34:12 <Yermak> Don't they already have that?
18:35:15 <aidanh> arguably, yes, but I think there's still a meaningful distinction
18:37:47 <aidanh> i wouldn't be against beem or something like it being integrated with e.g. the built-in ?/ lookup help system though
18:38:25 <aidanh> although I am less design/philosophy-focused than other devs here, and I'm not too sure of the original design goals
18:48:43 <advil> I'm not sure I'd be too worried about anything like that, console players can just open irc
18:49:43 <alexjurkiewicz> yeah. might be better to say that sequell gives in-group players an advantage
18:49:45 <advil> beem is an interface to learndb
18:49:59 <alexjurkiewicz> and monsterdb!
18:50:12 <advil> right
18:50:32 <advil> for learndb in particular integrating it into the game might not be a good idea, because we don't really vet the contents
19:11:50 <advil> some upgraded graphs here: http://crawl.akrasiac.org/scoring/per-day.html
19:12:44 <advil> I drastically reduced how often that page updates, instead of removing it; also added http://crawl.akrasiac.org/scoring/per-day-monthly.html which is actually sort of interesting if you want to get a sense of who's been playing lately
19:16:00 <gammafunk> yeah, beem doesn't confer any advantage since anyone can get the information it relays from any of the public interfaces, nor is the information it provides particularly reliable (learndb is notoriously out of date/wrong)
19:16:58 <gammafunk> the monster information arguably gives some advantage, but it's not very significant, and we are generally against that kind of information overload being in the game itself
19:17:54 <gammafunk> better to work to reform things so that spoilers and the benefits they provide are as small as possible
19:33:31 <SenoraRaton> It would be rather nice if the spells had damage in them on examine.  Beem has this, and it and MR are the only reason I would use beem monster lookup.
19:34:04 <SenoraRaton> Everything else is in the interface.
19:41:57 <advil> did CJR morgues ever end up anywhere?
19:47:29 <advil> gammafunk: do you know how some of your games got quadrupled or quintupled? the doubling I understand
19:48:20 <advil> e.g. there's a DsFi win that appears in two logfiles but shows up 5 times in your score page
19:49:30 <gammafunk> advil: yes, the reason is that at two different points when installing stable versions, part of the cjr trunk logfile was at the beginning of the new stable version's logfile
19:49:41 <gammafunk> so it's just the case that there were duplicate entries
19:49:42 <advil> ah
19:49:48 <advil> and then the logfiles got cleaned or something?
19:50:14 <gammafunk> reading over the dgl instructions last night when setting up cao, I think I see which step in those instructions led zibudo to do that
19:50:17 <gammafunk> one step has you copy the trunk data dirs over to install a new version, which is unecessary
19:50:22 <gammafunk> probably we should remove that step
19:50:29 <advil> I'm just confused because I can see two logfile lines for the game, one in the -git logfile, and one in the 0.20 logfile, but that doesn't get 5 games
19:50:36 <advil> also, does anyone know if at this point there are any logfile/milestone files that are *only* present as the copy on CAO?
19:50:37 <gammafunk> the data dirs (data, saves) are where the logfiles/milestone files live
19:50:58 <gammafunk> advil: yeah, there were more copies in the past, some of the logfiles got cleaned up
19:51:27 <gammafunk> well, I can't comment on how cao scoring would react to that duplication though
19:51:48 <advil> I think it'll produce as many copies of the game in the db as you give it distinct logfile entries
19:51:54 <advil> I will add some sort of hashing thing for games
19:52:01 <gammafunk> yeah, that would be excellent
19:52:18 <advil> do you know why scoring reads milestones at all?
19:52:23 <gammafunk> and I can't give you an exact history of what all we fixed there without trying to delve through email and irc logs, but we didn't realize this until after the fact
19:52:30 <gammafunk> advil: possibly just for in-progress games
19:52:38 <gammafunk> which for games played on CAO, you can see this on your player page
19:52:49 <gammafunk> for games not played on CAO, I don't think it does that
19:52:54 <advil> ah right
19:52:59 <advil> it's gathering them for everyone
19:53:19 <gammafunk> regarding your earlier question about copies only on CAO, not quite sure what you mean by that
19:53:38 <gammafunk> as in are there data archived on cao that we need (from other servers)? if so, then no I don't believe so
19:53:47 <gammafunk> there's an archive on shalott.org (Sequell's server)
19:54:07 <gammafunk> we have some specific defunct servers also backed up to amazon S3 thanks to alex
19:54:23 <gammafunk> cjr and....well I guess the other was cxc, which is probably now unecessary
19:54:25 <advil> yeah...I guess for a while cxc logfile was mia (and so only to be found on cao), but it seems like the new owner must have acquired it somehow
19:54:39 <gammafunk> but afaik we aren't doing archives on cao
19:54:55 <advil> the way scoring works it has copies of every logfile and milestone file
19:54:59 <gammafunk> right, shalott.org has the old cxc stuff anyhow, but namanix got the backups for CXC from zkyp/medar directly
19:55:37 <advil> so I was wondering if there was anything that should be backed up from there
19:55:37 <gammafunk> by old cxc stuff I specifically mean the logfile/milestones file, not the saves etc (namanix got those from the cxc backup)
19:56:09 <gammafunk> I don't think so; I can give you the public backup on shalott.org for comparison
19:56:19 <gammafunk> but I haven't looked through what's on CAO
19:56:34 <advil> if we have accurate backups of cjr that's probably the only thing
19:56:59 <gammafunk> right, those we do have on shalott and we have the backups on S3
19:57:08 <advil> do servers ever close down old versions?
19:57:17 <advil> I guess they probably leave the logfile etc up
19:57:42 <advil> another obscure thing CAO has is http://rl.heh.fi/
19:57:57 <gammafunk> advil: https://github.com/crawl/sequell/blob/master/config/sources.yml#L215
19:58:15 <advil> ah it already has it
19:58:24 <gammafunk> shalott also has rhf stuff, yeah
19:58:26 <advil> oh that's sequell
19:58:32 <gammafunk> yeah, that's sequell
19:58:35 <advil> good, I needed the morgue link
19:58:40 <gammafunk> but that's our "official" backup of cjr
19:58:42 <gammafunk> hosted by alex
20:06:10 <Cheibriados> New branch created: pull/1192 (1 commit) 13https://github.com/crawl/crawl/pull/1192
20:06:10 <Cheibriados> 03theJollySin02 07https://github.com/crawl/crawl/pull/1192 * 0.25-a0-43-gee6f27c: Fixed typo in wands mutation (12103) 10(4 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/ee6f27c1301b
20:06:27 -!- amalloy is now known as amalloy_
20:14:46 <alexjurkiewicz> i think i also have a backup of cszo dont i
20:15:23 <alexjurkiewicz> yeah, i do
20:15:56 <alexjurkiewicz> are they not needed? I thought dobrazupa.org was going to close at some point
20:15:58 <gammafunk> right, the logfiles and milestones files?
20:16:06 <gammafunk> yes, dobrazupa.org is gone
20:16:28 <gammafunk> alexjurkiewicz: it's probably wise to keep those in place until we're confident that we have a more stable location
20:17:11 <alexjurkiewicz> yep, the logfiles/milestones only for cszo. I will leave them as-is
20:17:47 <alexjurkiewicz> there was mentiona bove of needing to use @??monster to look up monster MR values. I wonder how many monsters have non-pip-rounded values for MR, if they could all be removed there would be no value from Cheibriados to be gained
20:31:34 -!- amalloy_ is now known as amalloy
20:42:18 -!- cjm_ is now known as cjm
21:11:45 <kitchen_ace> There are quite a lot of creatures with 60 MR (1.5 pips)
21:12:03 <kitchen_ace> And also a lot with 40 > MR > 0 but those are much less important
21:14:34 <Yermak> FR: make monster hexes with low probability (say 4% and less) have zero chance to affect player. It's really annoying to play safely against such outcomes.
21:16:57 <advil> aidanh: do you know if you'll be able to get to the the loading screen click bug? I can try if you won't be able to, that seems like something we should fix before release
21:22:52 -!- jfcaron_ is now known as jfcaron
21:41:38 <advil> huh, scoring doesn't store server name right now
21:52:48 <ebering> /win 2
22:08:26 <gammafunk> Napkin: I'm getting fatal: unable to access 'https://github.com/crawl/dcss_tourney.git/': gnutls_handshake() failed: Public key signature verification has failed.
22:08:36 <gammafunk> when I try to fetch the dcss_tourney repo
22:10:42 -!- Amnesiac_ is now known as Amnesiac
22:12:36 <Bcadren> CKO just went down; what's going on?
22:18:31 <kitchen_ace> updating for 0.24/tournament probably
22:24:46 <gammafunk> are you talking to me? I'm the one updating cdo for tournament, and this error would have nothing to do with that
22:27:53 <gammafunk> Napkin: I'm getting around this by wgetting the repo stuff to CDO, and wget to github works just fine; it's just that git fetch fails, odly
22:28:05 <gammafunk> by repo stuff I mean just a zip of the branch
22:28:15 <gammafunk> s/odly/oddly/
22:36:46 <kitchen_ace> gammafunk: I was replying to Bcadren about CKO going down
22:37:37 <gammafunk> oh ok, couldn't the original messages, so it looked like a reply to me
22:44:22 <alexjurkiewicz> you could try cloning with git+ssh instead of https
22:47:22 <gammafunk> I was going to try that, but then I'd need to set up a private key on cdo itself
22:47:28 <gammafunk> which is kind of inconvenient
22:47:43 <gammafunk> or I could use ssh forwarding I guess
22:48:06 <gammafunk> might try that if it can't get fixed, since I do this already for getting builds to cdo
22:48:11 <gammafunk> *ssh auth forwarding
22:50:28 <Bcadren> Hey floraline, sup?
22:52:21 <floraline> hi
22:52:58 <floraline> 0.24 is installed on cko, and i doubled cpu and ram in case of higher player counts during the tournament
22:57:16 <gammafunk> got it, thanks
23:07:50 <gammafunk> 0.24 tournament page is up: https://crawl.develz.org/tournament/0.24/
23:08:09 <gammafunk> ??tournament
23:08:09 <Sequell> tournament[1/3]: The 0.24 tournament will run from 20:00 UTC Friday 25 October through 20:00 UTC Sunday 10 November. Tournament page TBA soon.
23:08:52 <gammafunk> !learn edit tournament[1] s|page.*|page: https://crawl.develz.org/tournament/0.24/|
23:08:53 <Sequell> tournament[1/3]: The 0.24 tournament will run from 20:00 UTC Friday 25 October through 20:00 UTC Sunday 10 November. Tournament page: https://crawl.develz.org/tournament/0.24/