00:03:13 <gammafunk> alexjurkiewicz: I can hack this locally, but if you want a small project, making taildb.py have proper argument handling like loaddb.py does would be good
00:03:22 <gammafunk> right now it doesn't call connect_db() correctly since it doesn't supply the required args
00:04:14 <gammafunk> what I'll probably do for now is modify connect_db() to make those three args optional, since the necessary logic to handle things that way is in place (in part due to local CDO-specific commits)
00:06:14 <alexjurkiewicz> sorry, I don't think I'll have time to do that before tourney. I'm a little leery of making that change, since I'm not involved in the CDO workflow
00:06:53 <alexjurkiewicz> IMO it would be fine to make a commit to taildb which hardcodes the configuration CDO requires. Then in future it will be easy to work from that point backwards and add configurably as required for other setups
00:08:44 <alexjurkiewicz> I see those table alignment issues. I will fix it with some more hacks. Untyped Python + manipulating data in template = ugh
00:08:44 <gammafunk> Yeah I already have a local commit for CDO-specific stuff. Hard-coding is actually just done by that commit in loaddb.py; the only additional modification I had to make for taildb to work was to make the args to connect_db() optional
00:08:48 <gammafunk> but I'm saying it would be good to have properly command-line argument handling for taildb.py
00:09:13 <gammafunk> which you added to taildb.py, hence my request; looks like it will be fine to just defer that until post-tourney though
00:09:23 <gammafunk> er, which you added to loaddb.py, sorry
00:09:37 <gammafunk> i.e. extend the argument handling you added to the taildb thing
00:09:58 <gammafunk> that can definitely wait though, my slight change to connect_db() locally for CDO is working
00:10:11 <gammafunk> taildb is running along right now, so it will continue fetching updates and updating
00:10:26 <gammafunk> seems to fetch more frequently than I remember, in fact
00:10:35 <gammafunk> like one a minute?
00:11:16 <gammafunk> only other thing I need to add to testing is the rc files
00:11:24 <gammafunk> not sure that anyone has set up their clans yet, but we'll find out
00:12:26 <alexjurkiewicz> yeah. I haven't been testing with any clans set up locally, so a lot of that stuff is a bit of a mystery to me
00:13:27 <gammafunk> we'll find out how broken it is in a sec (maybe; if no one has defined a clan, there might be nothing to see)
00:15:18 <alexjurkiewicz> numeric_col = isinstance(val, (int, float, decimal.Decimal)) or (isinstance(val, (str, unicode)) and val.isnumeric())
00:15:30 <alexjurkiewicz> now this is dynamic typing
00:16:55 <ebering> the clan stuff is working on the back-end (I've been testing with it) but there aren't any templates
00:17:24 <ebering> I went to one of the protests today, the police fired teargas and a sound weapon at us unprovoked
00:17:43 <ebering> I'm probably not going to be super available the next few days
00:17:51 <ebering> but it should only require front-end
00:18:18 <gammafunk> dang, stay safe
00:18:29 <gammafunk> if there's no clan templates, that sounds like a problem
00:18:45 <ebering> they should be quick and easy at this point
00:23:51 <gammafunk> well I have all rcs fetched from 0.25 servers
00:25:17 <alexjurkiewicz> wow, thinking of you ebering
00:25:26 <gammafunk> I do see some teams defined
00:29:48 <alexjurkiewicz> hm, i don't see them on the player page
00:30:01 <alexjurkiewicz> can you tell me a user who is captain?
00:31:47 <gammafunk> alexjurkiewicz: defined as in the rcs, sorry
00:31:55 <gammafunk> but even better, just got the team processing to work: https://crawl.develz.org/tournament/0.25/teams.html
00:32:08 <gammafunk> had to make another cdo-specific path change
00:32:27 <gammafunk> hrm, the clan pages themselves aren't working
00:32:31 <gammafunk> because no template?
00:34:44 <gammafunk> yeah, no clan pages getting generated, it seems
00:34:54 <gammafunk> but the clan overview is working and the clan data in each player's page is working
00:35:35 <gammafunk> I'm going to put this aside for now and contact remaining admins about 0.25, so we get that covered
00:35:43 <gammafunk> we're still missing a few servers
00:36:18 <alexjurkiewicz> yep, i'll look into clan data. I guess I can download RCs myself
00:36:18 <gammafunk> if you like, I can share the cdo crontab for it
00:36:19 <gammafunk> each server needs slightly different wget args
00:36:25 <alexjurkiewicz> please! Just the part for CPO, I'll stick with the 0.24 data set
00:36:32 <alexjurkiewicz> of course :D
00:36:43 <gammafunk> oh, that's easy
00:38:13 <gammafunk> good lord, does cdo vim have mouse support enabled
00:38:46 <gammafunk> wget -q --no-check-certificate -r -np -nH --cut-dirs=2 -A '*.rc' h
00:38:59 <gammafunk> alexjurkiewicz: wget -q --no-check-certificate -r -np -nH --cut-dirs=2 -A '*.rc' https://crawl.project357.org/rc-files/0.25/
00:39:14 <gammafunk> oh, and I guess update that for 0.24 if that's what you want
00:39:33 <gammafunk> no idea what teams you'll have for 0.24, but some should be defined there
00:39:59 <gammafunk> by default you need those files to be in rcfiles-cpo in base dir of scripts
00:42:24 <alexjurkiewicz> ty
00:42:42 <alexjurkiewicz> going to need some new docker handling for all this as well of course
00:42:58 <alexjurkiewicz> so i think I'll be working on this right up until the tourney is about to begin
00:44:07 <alexjurkiewicz> wow, got 1.1k files in 7 seconds. More efficient than I thought
00:46:17 <gammafunk> well, if it turns out we need a delay due to things going on IRL, we can always decide that
00:47:20 <gammafunk> next few days might be very busy for me, but if not I can lend some assistance; just testing with cdo won't take much time
00:47:20 <gammafunk> not sure about writing templates though
00:47:34 <gammafunk> let me contact those admins before I forget
00:48:31 <alexjurkiewicz> now that I have the rc files locally (and now it occurs to me I could have rsynced them since I admin CPO... heh) I should be able to test easily enough and ping you to pull updated code periodically
01:00:10 <gammafunk> sounds good
01:17:08 <alexjurkiewicz> how do you feel about merging new-scoring to master?
01:21:31 <gammafunk> alexjurkiewicz: it wouldn't affect things too much, I think, but I'm not sure there's a need. Maybe we can do that when we really get a fully working version
01:21:54 <gammafunk> We also have some post-tourney cleanup to do, e.g. python3 conversion, but I'm not sure how much people will be willing to do that
01:22:04 <gammafunk> er do that after tourney
01:22:26 <gammafunk> could be something worked on during tourney and kept in a branch
01:22:47 <gammafunk> cleaning up the config situation would be great, also changing the indent to not be 2-space
01:23:11 <gammafunk> so I guess I don't see a huge gain from it, but would defer to whatever you, ebering, and advil need/want
01:23:31 <alexjurkiewicz> yeah, you're right, might as well do after the tourney now as part of general cleanup
01:23:45 <alexjurkiewicz> i do wonder how much work will get done post-tourney. Historically not much
01:24:32 <gammafunk> yeah, usually I just do the bare minimum of archival stuff
01:25:18 <gammafunk> I might actually get around to the config stuff this time, since it drives me insane very time
01:41:17 <gammafunk> ok, all servers without 0.25 notified, and I've updated sequell config for tournament and what 0.25 sources we have right now. Also sent snark an email asking about Sequell for tournament, since it's not been seeing new stuff since about 5/20
03:33:31 <Kelbell> Unstable branch on crawl.kelbi.org updated to: 0.26-a0-43-g20b394cb32 (34)
03:34:25 <Kelbell> Fork (bcadrencrawl) on crawl.kelbi.org updated to: 0.22.1-2769-g340536b265
04:12:42 <alexjurkiewicz> I've pushed an update to write clan pages, although the clan category info is still wrong
04:47:58 -!- aegolden_ is now known as aegolden
05:21:45 -!- amalloy is now known as amalloy_
06:50:38 <alexjurkiewicz> clan display should entirely work now. There are some changes to the UX: 1) clans are now referred to everywhere by their name, not by their captain. Since AIUI multiple clans can have the same name, each clan's unique ID is name-captain
06:51:28 <alexjurkiewicz> known issue: There are a couple of categories that don't display rankings properly (zig dive mainly). I am waiting on edgar to fix these but if he's too busy with fighting the good fight I might be able to puzzle it out
06:53:04 <alexjurkiewicz> There's a lot of ugly code in the glue between database layout and frontend display. Some can be fixed by better architecture (pulling all data loading logic into python files, using better rendering primitives) but a fair amount is due to mismatch between data format of how frontend wants and backend provides. I think this latter stuff will naturally improve over time in future tournaments, but the
06:53:07 <alexjurkiewicz> former will probably need active attention to improve
06:54:39 <alexjurkiewicz> I mentioned briefly before that usually after a tourney ends there's not a lot of motivation to work on tourney scripts until the next release (I did 80% of a bootstrap rewrite for a previous tourney but it didn't get finished in time and I dropped it on the floor). One way to challenge this would be if it's possible to unify the codebases of scoring & tourney somehow. I think kyle is probably best placed
06:54:42 <alexjurkiewicz> to speak about if that's possible / desirable
06:54:46 <alexjurkiewicz> (using first names to avoid pings)
06:56:47 <alexjurkiewicz> btw, I don't have any TODO on my plate except zig depth now. Let me know if you find any issues. From cosmetic to "I think you should add data x to page y"
07:50:50 <Pinkbeast> Gasp hot out there, I am declaring the Sun to be well and truly over the yardarm
07:54:34 <Pinkbeast> Er, wrong channel, sorry
08:01:35 -!- amalloy_ is now known as amalloy
09:34:55 <cebolla> which of these is more similar to crawl's code style? https://paste.ubuntu.com/p/mqGxj6SDBt/
09:35:44 <cebolla> in the first one, they use the return value of a function to pass data to main(), in the second one, i used a global variable to hold the data
10:59:31 <advil> cebolla you wouldn't think it but we try to avoid adding global variables unless necessary (crawl has a lot of preexisting global state still unfortunately)
11:14:33 <cebolla> i see!
11:58:00 <gammafunk> alexjurkiewicz: Is there no facility in new tournament scoring for displaying the nem choice combos?
11:58:18 <gammafunk> I just realized that I don't see it picking nem choice, nor do I see them displayed anywhere on the page
13:35:49 <ebering> it picks the nem choice
13:35:55 <ebering> the display hasn't been ported over
13:56:52 <gammafunk> ok, need to look into why taildb/loaddb hasn't picked one yet
13:57:16 <gammafunk> maybe just because old file already existed
13:58:53 <gammafunk> indeed
13:59:00 <gammafunk> it picked one after I removed the old file and restarted
13:59:50 <gammafunk> so taildb has been running for about 12 hours with no errors I've noticed
14:00:11 <gammafunk> just restarted it because of nem choice thing, but otherwise so far, so good
15:06:00 <advil> hm, we still have a TBC banner for wjc
15:07:05 <advil> also, stupid quesiton, what's the state of banner implementation?
15:07:44 <advil> last I checked the pages were just doing them randomly for testing purposes
15:14:56 <gammafunk> yeah, I don't see a banner page
15:20:50 <advil> you can see them on player pages
15:20:56 <advil> under the third tab
15:21:04 <advil> less wondering about the frontend imp but the db
17:18:47 -!- jfcaron_ is now known as jfcaron
17:49:30 <alexjurkiewicz> the banner implementation is good now, ebering fixed it
17:49:53 <alexjurkiewicz> yes, good catch about nem choice display. If someone can tell me the function(s) that generate em choice data I'll integrate it into overview
17:50:02 <alexjurkiewicz> or I can look later today
17:55:02 <gammafunk> yeah, I was going to look at that in detail
17:56:56 <gammafunk> alexjurkiewicz: nemelex.list_nemelex_choices
17:57:12 <gammafunk> https://github.com/crawl/dcss_tourney/blob/master/templates/overview.mako#L11
17:57:24 <gammafunk> from the old overview template in master
18:25:07 <ebering> I thought I pushed that
18:25:11 <ebering> the db is implemented
18:25:22 <ebering> ah yes y'all picked up on that
18:25:26 <ebering> I'm taking it easy tonight
18:26:06 <ebering> but I dunno if I'll be able to focus on crawl
18:55:59 <advil> if you want something non-tourney to look into, this char apparently got bolt of fire in a randbook: https://pastebin.com/3id0SPQ0
18:56:18 <|amethyst> !source nicolae_zot_draconian_monastery
18:56:25 <Sequell> 1/1. https://github.com/crawl/crawl/blob/master/crawl-ref/source/dat/des/branches/zot.des#L1664
18:56:41 <advil> oh, yeah it's just in a list
18:57:12 <advil> I searched for incunabula and assumed it was a random randbook
18:57:39 <gammafunk> dangit
18:57:47 <gammafunk> I did a git grep for "bolt of fire"
18:58:01 <gammafunk> thwarted by underscores!
18:58:38 <advil> yeah I did do that grep too
18:59:53 <advil> as a side note I also couldn't figure out why just about anything doesn't end up in a randbook
19:00:12 <advil> still not sure if I was missing the crucial check *somewhere*
19:00:23 <gammafunk> hrm, I want to say
19:00:30 <gammafunk> it does a check that the spell is in some fixed book
19:00:34 <gammafunk> I seem to recall this
19:02:27 <advil> ugh also just noticed that randbook has a seen spell check that is probably broken by seeding
19:03:21 <advil> maybe that is just for acquirement
19:16:14 <gammafunk> advil: found it (I think), and it is convoluted indeed
19:17:20 <gammafunk> oh, hrm, wait a sec
19:17:42 <gammafunk> oh my god
19:17:48 <gammafunk> yes, I was right, and it's even weirder
19:18:57 <gammafunk> basic_themed_spell_filter->_agent_spell_filter (applies even if no agen) -> _agent_spell_filter() returns false because spell has rarity -1 (since in no book) and agent is not sif or it is sif and it's not a rare spell
19:19:36 <gammafunk> If I'm right, removed spells don't get added to randbooks almost completely by accident (looking at this code)
19:20:24 <advil> nice
19:21:13 <gammafunk> looking at this
19:21:13 <advil> though I guess in a way, checking if they appear in any  regular books is an extremely robust way of not having to directly check for removedness
19:21:13 <alexjurkiewicz> you're phrasing that incorrectly. It's another example of the cool emergent interactions a game like DCSS can generate
19:21:19 <advil> I looked at _agent_spell_filter but failed to follow the thread to spell_rarity
19:21:21 <gammafunk> if we were to remove a spell that appears in a rare book, sif worshipers could get it as a god gift?
19:21:31 <gammafunk> meaning sif can still gift people glaciate?
19:21:52 <gammafunk> since in that case agent is sif and it's a rare spell
19:21:54 <gammafunk> ahh
19:22:02 <gammafunk> but glaciate isn't in a rare book any more
19:22:06 <gammafunk> so no, it works for that too
19:22:14 <gammafunk> yeah I'm calling it a miracle that this works
19:23:01 <advil> so I suppose the vault syntax should produce an error if a book tries to place something with spell_rarity() = -1
19:23:05 <advil> *==
19:23:08 <gammafunk> right
19:23:49 <gammafunk> hrm
19:24:01 <gammafunk> but are spells in no book actually rarity -1?
19:24:19 <gammafunk> looking at the rarity init, seems to be using ...oh, doh!
19:24:29 <advil> if (rarity == 255)
19:24:30 <advil> return -1;
19:24:31 <advil> ?
19:24:33 <gammafunk> boy this is weird, spell rarity in sets a lowest rarity, using 255
19:24:34 <gammafunk> yes
19:24:53 <gammafunk> then translates that to -1 rarity for the rarity value
19:26:01 <gammafunk> !learn edit beem[4] s/information/crawlcode/
19:26:01 <Sequell> beem[4/6]: I am not AI. My codename is project beem. I am a living, thinking entity that was created in the sea of crawlcode.
19:26:43 <gammafunk> crawlcode will become self-aware some day
21:38:06 -!- amalloy is now known as amalloy_
22:59:27 <Bcadren> !tell floraline Sorry about these coming up, but another save to delete (BcadrenCrawl) CanOfBees.
22:59:28 <Sequell> Bcadren: OK, I'll let floraline know.
23:11:57 <Kelbell> Webtiles server stopped.
23:27:46 -!- amalloy_ is now known as amalloy