02:09:09 <alexjurkiewicz> non-monospace font, and new colour theme https://imgur.com/a/3Mqh1md
02:10:45 <alexjurkiewicz> annoying that there's chroma subsampling in macos screenshots. so blurry
02:20:08 <aidanh> is there a repo for this I can hack on a bit?
02:20:08 <aidanh> looks good
02:31:47 <alexjurkiewicz> dcss_tourney new-scoring
02:31:58 <alexjurkiewicz> hacking on the design?would love help
02:32:21 <alexjurkiewicz> i haven't pushed this yet, gimme a couple of hours unless you are keen to start now I can push wip work to a branch
03:11:47 <Rotatell> Unstable branch on crawl.beRotato.org updated to: 0.26-a0-21-gf97e55d (34)
03:18:04 <alexjurkiewicz> advil: re 1c4fc3e855e459dee8c3e4c48a431692465506b8 Don't require typing -- typing is included in requirements.txt so the import should be fine, right? What case would it fail?
03:27:03 <alexjurkiewicz> aidanh: pushed
03:30:48 <Kelbell> Fork (bcrawl) on crawl.kelbi.org updated to: 0.23-a0-3122-gcd7237df04
03:37:30 <Kelbell> Fork (bcadrencrawl) on crawl.kelbi.org updated to: 0.22.1-2735-ge551f443f3
05:05:07 <cebolla> just ran into my own @_adjective_@ randart on trunk  ^_^
05:05:20 <cebolla> the ring of the Lost Tusks {rPois rN+ Str+6 Stlth+}
05:38:38 -!- amalloy is now known as amalloy_
06:01:05 -!- amalloy_ is now known as amalloy
07:14:52 <alexjurkiewicz> I've updated the Docker instructions in README for dcss_tourney. Hopefully they make more sense. Also, loaddb depends on the mysql container so it should take care of automatically starting it for you
08:06:34 <alexjurkiewicz> more updates. made challenges smaller for player pages and they can be >1 per row now (not sure if this is good or not): https://imgur.com/a/Y1UR5Bq
08:07:04 <alexjurkiewicz> the heights are too big now, need to fix the layout so they don't take up unnecessary vertical space
09:59:28 <elliotd123> Hi guys, brand new at this open source stuff - but I'd like to start contributing a little bit. Any good advice?
10:00:15 <elliotd123> I have a typo to fix, so should I create an entire local branch for that and then create a PR, or should I just update my forked master and PR that?
10:00:45 <ebering> For a typo fix an PR from your forked master is fine, except that other commits might appear in the interim
10:00:57 <ebering> Good practice is to branch (branches are cheap in git) and submit the PR based on the branch
10:01:12 <ebering> If you're looking to get more involved, it's best to start with good habits now!
10:01:16 <elliotd123> ok should I rebase from master before submitting the PR?
10:01:20 <elliotd123> Definitely!
10:02:10 <ebering> It's appreciated but we can do that on our end when landing the PR
10:05:18 <elliotd123> For a typo like this, I probably don't need to do this, but is there a set of tests to run on code changes before I submit a PR?
10:11:11 <advil> alexjurkiewicz I'm running it locally on a python 2 install without the typing module, ime it's often not there and the py2 version is just a backport on pip
10:11:31 <Cheibriados> New branch created: pull/1424 (1 commit) 13https://github.com/crawl/crawl/pull/1424
10:11:31 <Cheibriados> 03Elliot Dronebarger02 07https://github.com/crawl/crawl/pull/1424 * 0.26-a0-22-g795f99b: Fixed a typo in god-passive.cc 10(5 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/795f99b74d3e
10:12:07 <ebering> It's appreciated but we can do that on our end when landing the PR
10:12:14 <Cheibriados> 03Elliot Dronebarger02 {ebering} 07* 0.26-a0-22-g2a348b2: Fixed a typo in god-passive.cc 10(6 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/2a348b289f77
10:12:36 <ebering> oops
10:12:42 <ebering> elliotd123: we have github actions (and for a little while longer travis-ci) set up to run continuous integration compiling and tests
10:12:51 <ebering> how would you like to appear in the CREDITS?
10:15:03 <elliotd123> Yeah I saw those github actions. I guess I was wondering if I can test to make sure code changes pass tests before I create a PR and break everything lol
10:15:41 <elliotd123> me in the credits? Maybe I should make a few more contributions first
10:17:11 <Kelbell> Unstable branch on crawl.kelbi.org updated to: 0.26-a0-22-g2a348b289f (34)
10:18:05 <ebering> Every contribution goes in the credits, unless you'd rather not
10:18:11 <ebering> It's fine to open a PR to get the actions run
10:18:25 <ebering> If its early in your development process just marke the PR as a draft!
10:23:37 <elliotd123> Sounds good, thanks for the help ebering
10:27:29 <ebering> Also, if you're planning a major gameplay change you should check-in here before you get too deep into the weeds
10:27:59 <ebering> And if you're planning on getting more involved, take the time to check out the contribution guide at some point (for small fixes and tweaks its not so urgent)
10:28:02 <ebering> ??patch
10:28:05 <Sequell> patch[1/2]: For details about how to discuss and submit a patch or pull request, see: https://github.com/crawl/crawl/blob/master/crawl-ref/docs/develop/contribution-process.md
10:29:04 <elliotd123> Just getting my feet wet at the moment, but I'll keep that in mind - I've read some of the contrib guide already, I've just never contributed to an Open Source project before, so wanted to make sure I'm doing this right
10:29:10 <elliotd123> Most of my coding work has been in stuff that I have commit access to
10:41:49 <ebering> alexjurkiewicz: re mono vs. sans, how do you feel about using google fonts to load a more legible mono as the default
10:44:23 <ebering> or at least using mono for some display fonts
10:44:44 <aidanh> maybe for table figures but not heading or body text?
10:46:31 <advil> yeah, I do not like mono for body text :-)
10:51:13 <ebering> a long shelved type design project of mine: crawl.otf a display font based on ontoclaasm's title lettering
10:58:40 <aidanh> unfortunately I think that lettering would be overkill for anything more than just a main title graphic
10:59:34 <aidanh> I rummaged through a bunch of serif fonts a while back and the nice ones were all eye-poppingly expensive
11:05:44 <Cheibriados> 03advil02 07* 0.26-a0-23-g3ae7099: Fix accidental no-daemonize by default (johnstein) 10(64 seconds ago, 1 file, 4+ 3-) 13https://github.com/crawl/crawl/commit/3ae7099c4749
11:08:13 <ebering> yeah crawl.otf would definitely be a display font for light usage
11:14:52 <aidanh> truth be told I've been itching to get non-mono font support into local tiles for a while, but it'd require a big rewrite of the text rendering code
11:17:04 <Kelbell> Unstable branch on crawl.kelbi.org updated to: 0.26-a0-23-g3ae7099c47 (34)
11:25:10 <elliotd123> Any reason there's a spider named "Tarantella" instead of "Tarantula"?
11:31:01 <|amethyst> https://en.wikipedia.org/wiki/Tarantism https://en.wikipedia.org/wiki/Tarantella
11:37:31 <elliotd123> Yeah I mean I kind of thought it was on purpose to do it that way
12:05:01 <advil> does anyone know the current status of shallott?
12:05:19 <advil> johnstein is unable to back up ttyrecs there right now so cbro is going to hit a disk space issue soon
12:38:47 <Cheibriados> 03advil02 07* 0.26-a0-24-g700258a: Enhance an error message 10(70 seconds ago, 1 file, 3+ 1-) 13https://github.com/crawl/crawl/commit/700258a8a8b0
12:47:06 <Kelbell> Unstable branch on crawl.kelbi.org updated to: 0.26-a0-24-g700258a8a8 (34)
12:48:58 <cebolla> what about dejavu sans mono?  that seems to exist for free on all linux installs, and imo looks good in  console games
12:49:09 <Cheibriados> 03advil02 07* 0.26-a0-25-g5c64649: Fix a bytes/str issue in error message 10(61 seconds ago, 1 file, 4+ 3-) 13https://github.com/crawl/crawl/commit/5c64649191f3
12:50:06 <advil> crawl already uses dejavu sans mono
12:50:22 <cebolla> >_>
12:50:25 <cebolla> i see!
12:50:40 <aidanh> the discussion was about non-monospace options
12:51:06 <advil> and the tourney scripts, I'm not sure if there is a good way to use it on the web
12:51:23 <aidanh> although dejavu sans does indeed have a non-mono variant, web fonts have their own distribution issues
12:51:43 <cebolla> ah, my mistake
12:52:27 <aidanh> no worries
12:55:39 <cebolla> does crawl have any probability rubberbanding?  like, if food, ammo, curing, fails to drop for several floors, does it become more likely to appear on the next floor? (or whatever items are basic-needs-ish)
12:56:21 <|amethyst> not in general, but there are a few concessions that have a similar effect
12:58:41 <|amethyst> e.g. there is a guaranteed ration somewhere between D2 and D5; and which overflow altars go on which levels is determined before the dungeon is even generated
13:02:09 <Kelbell> Unstable branch on crawl.kelbi.org updated to: 0.26-a0-25-g5c64649191 (34)
13:06:35 <Henzell> Unstable branch on crawl.akrasiac.org updated to: 0.26-a0-25-g5c64649 (34)
13:36:43 -!- amalloy is now known as amalloy_
14:36:15 <Cheibriados> 03advil02 07* 0.26-a0-26-ge22fd893: Don't send ui_state_sync when not in game 10(83 seconds ago, 2 files, 5+ 1-) 13https://github.com/crawl/crawl/commit/e22fd8933b49
14:37:02 <advil> will be excited to see where that row of dominos leads
14:37:38 <aidanh> ah, nice catch
14:44:51 <advil> heh well we'll see if I broke anything
14:46:13 <advil> that in_game check was not widely used so it might not be very reliable
14:46:31 <advil> although I wonder now if that tweak I just made will fix other elusive bugs with chat
14:47:35 <Kelbell> Unstable branch on crawl.kelbi.org updated to: 0.26-a0-26-ge22fd8933b (34)
14:47:35 <advil> I'm pretty sure that before that commit if you got a close message (e.g. via a crash or spectating a game while it closed), in_game() would never return true until reload
15:19:59 <Cheibriados> Electric Eel stumbled onto land in its confusion 13https://crawl.develz.org/mantis/view.php?id=12273 by Midn8
15:33:35 <gammafunk> WARNING: WebTiles: Told to go to lobby while watching user suddenbreakfast.
15:34:10 <gammafunk> beem gets these go to lobby messages in a funny order sometimes
15:34:59 <gammafunk> just after a game has ended, I need to verify the exact message sequence
15:36:53 <gammafunk> yeah, looks like we got game_ended but watching state not updated internally before getting go_lobby
17:28:32 -!- aidanh_ is now known as aidanh
17:51:30 <alexjurkiewicz> advil: is that a fix for an issue reported with trunk?
17:51:34 <alexjurkiewicz> i've got reports like "Anyone else got UI issues with trunk? Dialogs don't seem to close properly"
17:51:38 <alexjurkiewicz> "I'm having the same issue. Neither incognito nor clearing cache helps."
17:51:54 <gammafunk> uh oh
17:52:25 <gammafunk> it was a fix for log warnings
17:52:27 <alexjurkiewicz> ebering: i'm happy to try a better monospace font. But I doubt it will be as readable. Perhaps a compromise could be monospace for certain areas
17:52:46 <gammafunk> hopefully didn't induce a new bug (something advil mentioned might happen)
17:53:28 <alexjurkiewicz> +ebering | It's fine to open a PR to get the actions run <-- actually, if you fork the repo you fork the CI config (in .github/workflows). So you should see tests against commits in your fork cc elliotd123
17:56:12 <alexjurkiewicz> advil: it's impossible to close dialogs on CPO trunk. If you open inventory and press escape, the following console error log appears https://imgur.com/a/B2bZY5Y
18:07:32 <advil> alexjurkiewicz that looks like something that would happen if client.js did not get updated for whatever reason
18:09:20 <advil> yes, when I look at the version being loaded it is not up-to-date
18:09:32 <advil> is it possible you aren't automatically updating it?
18:10:52 <advil> cko (left) vs cpo (right): https://www.dropbox.com/s/lkgtrrtzdye6vju/Screenshot%202020-05-26%2018.10.44.png?dl=0
18:11:48 <advil> that `exports...` line being the crucial one
18:12:31 <alexjurkiewicz> i'm not updating webserver automatically
18:12:54 <alexjurkiewicz> so you expect the same sort of behaviour as on dgamelaunch-config, where I update the files but don't necessarily restart the server every release?
18:12:55 <advil> ah including static/scripts I guess?
18:12:58 <alexjurkiewicz> *every commit
18:13:46 <alexjurkiewicz> there is no automatic update for any part of webserver code
18:13:48 <advil> yes, I have generally been assuming that static/ generally gets updated on a rebuild
18:14:07 <alexjurkiewicz> ok, gotcha
18:14:09 <advil> too many generallys
18:14:29 <advil> style.css in there also gets minor tweaks regularly
18:16:21 <advil> you must already be updating game_data right?
18:17:09 <alexjurkiewicz> no, only when i update the server
18:17:16 <advil> huh
18:17:22 <alexjurkiewicz> which is not that often
18:17:32 <alexjurkiewicz> it's about 1.5mo old right now
18:17:47 <advil> but game_data *must* have been updated for this bug to happen
18:17:58 <advil> because you got a change to js in game_data without the corresponding change in static/scripts
18:18:18 <alexjurkiewicz> would that be because game_data is actually served by the crawl install's web folder?
18:19:15 <alexjurkiewicz> # g log game_data | head -3
18:19:17 <alexjurkiewicz> commit 20d90ebb82add7df39850d2e74ccd3c138c4d5ba
18:19:19 <alexjurkiewicz> Author: Kate <601195+semitonal@users.noreply.github.com>
18:19:21 <alexjurkiewicz> Date:   Tue Mar 17 19:57:54 2020 +0000
18:19:46 <advil> ah, that would be it
18:20:00 <advil> also true of dgl setups
18:21:22 <alexjurkiewicz> ugh, I don't really want to copy the dgl script hackiness. I think the best approach is to update the full webserver checkout and restart it periodically
18:21:22 <advil> well, static/scripts doesn't change that often, https://github.com/crawl/crawl/commits/master/crawl-ref/source/webserver/static/scripts, so it's mostly safe
18:21:34 <advil> only 7 changes in the last year
18:22:00 <advil> yes, now that I know about this I can probably plan ahead for commits like this and warn you
18:22:11 <alexjurkiewicz> by the way, so this change broke newer trunk versions, but it doesn't break older builds like 0.24?
18:22:37 <advil> the break is that code served from game_data/static/scripts calls client.in_game(), which was not previously exported as a function
18:23:06 <alexjurkiewicz> ah, funny! I have a local branch which exported that too!
18:23:08 <Lantell> Unstable branch on underhound.eu updated to: 0.26-a0-26-ge22fd8933b (34)
18:23:08 <advil> %git
18:23:08 <Cheibriados> 07advil02 * 0.26-a0-26-ge22fd893: Don't send ui_state_sync when not in game 10(4 hours ago, 2 files, 5+ 1-) 13https://github.com/crawl/crawl/commit/e22fd8933b49
18:23:14 <advil> heh
18:23:57 <alexjurkiewicz> ok, so I can update the webserver checkout every trunk update. But restart it less frequently, say daily or weekly. That should minimise breakage due to back/forwards incompatibility
19:15:08 -!- jfcaron_ is now known as jfcaron
19:23:00 <advil> it's actually sort of odd that static/scripts/*.js, static/style.css, and templates/client.html are not versioned
19:23:18 <advil> I'm surprised there hasn't been any major backwards compat issues that I know of
19:32:04 <advil> alexjurkiewicz did you actually update client.js earlier, or just set it up to update?
19:32:12 <Cheibriados> Abilities menu does not cancel. 13https://crawl.develz.org/mantis/view.php?id=12274 by Tenaya
19:32:41 <advil> still seems broken
19:32:41 <advil> I can commit an interim fix if needed
19:32:53 <advil> (^ is still this issue)
19:40:56 <alexjurkiewicz> i haven't had a chance to fix yet. in the next hour
19:46:12 <advil> hm, I'm not even sure after all this trouble if I've gotten rid of that log msg
19:49:09 <alexjurkiewicz> oh crap. the old template paths were re-added to the repo?
19:50:34 <advil> %git e9962a393ca
19:50:34 <Cheibriados> 07advil02 * 0.25-a0-1091-ge9962a3: Add back in all non-default templates for compatibility 10(11 days ago, 3 files, 8+ 0-) 13https://github.com/crawl/crawl/commit/e9962a393cac
19:50:36 <advil> yes, sorry
19:50:42 <advil> it was breaking everything to not have the
19:50:43 <advil> m
19:50:50 <alexjurkiewicz> damn you dgl scripts
19:51:08 <alexjurkiewicz> lucky i have some backups, even if it's a full disk image
19:51:14 <advil> ugh, sorry
19:51:52 <alexjurkiewicz> all good, not your fault. I'm happy to blame the code which never really thought about "how do people update this"
19:58:03 <alexjurkiewicz> great, updated & confirmed fix
19:58:06 <alexjurkiewicz> thanks for the help advil
20:01:11 <gammafunk> poor dgl, always gettin made fun of
20:02:09 <advil> no problem...confirming that it is fixed but I had to force flush my browser cache
20:04:32 <alexjurkiewicz> dgl is like... every time I try and improve webtiles, there's some wart in dgl making the improvement non-functional :(
20:07:27 <advil> well, in fairness that one is very specifically a consequence of what happens when you delete shudown.html from under a running webtiles server started before the commit that handled the .default files
20:07:34 <advil> *shutdown.html
20:08:22 <advil> if all the webtiles servers get restarted for 0.25, it should be possible to revert that commit afterwards
20:09:26 <advil> so it's more of a webtiles server robustness thing as much as a dgl thing, they just didn't work well together
20:11:27 <advil> the timing of these ui_state_sync messages is hard to pin down
20:11:57 <advil> I think what is happening is that the message can get queued before the client is gone, but not sent until afterwards
20:24:47 <Cheibriados> 03advil02 07* 0.26-a0-27-g1f02269: Add an ugly exception for ui_state_sync 10(31 seconds ago, 1 file, 8+ 1-) 13https://github.com/crawl/crawl/commit/1f0226971c1c
20:32:05 <Kelbell> Unstable branch on crawl.kelbi.org updated to: 0.26-a0-27-g1f0226971c (34)
20:52:23 <Cheibriados> 03Implojin02 07https://github.com/crawl/crawl/pull/1409 * 0.25-a0-1060-gd852dcc: Remove the SPARM_LUNGING evoke toggle. 10(7 minutes ago, 14 files, 8+ 79-) 13https://github.com/crawl/crawl/commit/d852dcce0016
21:06:34 <Implojin> hi hi
21:08:28 <Implojin> I have some stuff to work on crawlstuff tonight and I'd like to finish up SPARM_LUNGING to the point where it could maybe go into trunk; if anyone has some time to review the current state of that PR that would be nice
21:09:00 <Implojin> some time to work on crawlstuff*
21:09:08 <Implojin> all that's really jumping out at me, at the moment, is rebasing it and finding a replacement name
21:10:16 <gammafunk> Implojin: one thing to keep in mind is that we avoid merging new features like that in trunk until after tournament
21:10:27 <Implojin> oh, sure
21:10:38 <gammafunk> for rebase, we can certainly do that at time of merge
21:10:39 <Implojin> i thought you guys had already tagged that and had some kind of 0.26 branch going, though?
21:10:59 <gammafunk> well we tage 0.25 and make a branch for that, but trunk is already 0.26-a
21:11:16 <gammafunk> we just avoid putting new features in that trunk (so 0.26-a) until after tourney
21:11:23 <gammafunk> s/tage/tagged/
21:11:28 <Implojin> okay
21:11:35 <Implojin> well, in that case, i'll hold off on rebasing
21:11:35 <gammafunk> and I should say we tagged the 0.25 beta, not final version
21:11:55 <gammafunk> yeah, rebase is not a big deal; definitely people can still give you feedback in PR (and you can still make commits as you like of course)
23:04:12 <gammafunk> ebering: I was planning on removing the part of 340c9e81949e71a15bc that considered the player to be a ranged ally, since it's pretty frustrating for the most common use cases of allies and a number of people have complained about it
23:04:28 <gammafunk> that is, stop making allies move out of the lof between the player and a hostile
23:10:24 <Implojin> if i have a particularly large clua script, would it be generally okay to run that on the online crawl servers? should i be rate-limiting it somehow, or is rate-limiting of user clua adequately handled serverside? i'm vaguely aware that some servers have bumped against hardware limitations in the past; i don't want to cause problems for anyone
23:11:23 <Implojin> ballpark size of this script is somewhat smaller than qw
23:11:30 <gammafunk> Implojin: we asked that people have a 100ms (I think) delay for that
23:11:37 <gammafunk> between actions
23:11:42 <alexjurkiewicz> theoretically it's handled by the server, but I don't think the server applied rate limiting is very good
23:11:59 <gammafunk> so if you implement your clua in a similar way to qw with a similar delay, should be ok
23:12:04 <Implojin> okay, thanks :)
23:12:53 <gammafunk> and yeah I'm pretty ignorant of what can actually cause lag beyond that; perhaps not all lua is created equal and some code needs a longer delay
23:13:16 <Implojin> yes, i haven't profiled this at all, it's just thousands of lines of clumsy hacks at present
23:13:31 <Implojin> i'll try to implement something similar to qw's rate limiting
23:17:23 <advil> I would test it out locally with rate limiting on using whatever cpu monitor is appropriate for your os
23:18:29 <advil> would also check with the admin for the specific server, some server admins may have more specific preferences about running bots
23:19:21 <Implojin> it's really only half-bot, it returns control to the user whenever it thinks something interesting is happening, but it does regularly go for multiple floors worth of turns on its own
23:19:27 <Implojin> but okay
23:21:02 <gammafunk> is this iw
23:21:16 <gammafunk> or Iw I guess
23:21:24 <Implojin> its not *w related except that i've browsed qw code in the past
23:21:41 <Implojin> https://github.com/Implojin/ff-rc
23:21:57 <Implojin> it began life as an "o tab (e)at G>" script
23:22:19 <Implojin> and then, i guess, feature creep happened
23:22:50 <gammafunk> cool
23:22:53 <gammafunk> "ancient, cursed crawl Lua"
23:23:04 <gammafunk> so that's what clua stands for, cursed lua
23:23:22 <Implojin> if i can apply that retronym to clua i think i could die happy
23:24:29 <advil> are those tabs
23:25:13 <Implojin> it's possible that whatever text editor came default with this distribution might have been set to tabs and i might not have noticed until much later
23:25:20 <Implojin> theoretically
23:25:39 <gammafunk> advil: pretty appropriate that this rc uses TAB, don't you think?
23:26:10 <advil> fair
23:28:01 <advil> also I literally did not know that userbase.lua exists, what is this stuff
23:28:18 <Implojin> crawl lua is vaguely horrifying
23:29:35 <advil> I'm not entirely sure what some of this stuff will do on a server
23:29:52 <advil> chk_lua_save is in dlua though so it may simply not be accessible from ready()
23:30:31 <Implojin> at present it's not using chk_lua_save, i think
23:30:35 <advil> I'm a little surprised that c_persist works
23:30:43 <advil> yeah I'm responding to your note that you couldn't get it to work
23:31:05 <Implojin> hmm
23:31:52 <advil> I guess qw uses c_persist so it must work at least
23:41:20 <alexjurkiewicz> awesome
23:41:27 <gammafunk> advil: c_persist definitely works and I have some custom code using it as well
23:41:32 <alexjurkiewicz> if you want to test on a server with an admin around, you can try on cpo now
23:41:47 <gammafunk> I think userbase.lua documents some other c_foo callback lua functions, right?
23:41:59 <gammafunk> I recall putting the documentation for the aquirement interface there
23:42:41 <gammafunk> elliptic implemented c_persist specifically to support qw
23:42:44 <gammafunk> !source userbase.lua
23:42:45 <Sequell> https://github.com/crawl/crawl/blob/master/crawl-ref/source/dat/dlua/userbase.lua
23:48:07 <alexjurkiewicz> wow, preferred treatment for dev-run bots
23:48:20 <alexjurkiewicz> and you won't even implement Implojin's boots of lunging
23:54:06 <Implojin> alexjurkiewicz: that's generous, thank you, much easier than setting up dgamelaunch or whatever to test locally
23:54:20 <Implojin> let me put a delay into the action loop real quick