01:07:11 <alexjurkiewicz> ebering: are you around to explain query.update_rank / query.update_all_player_ranks ?
01:38:56 <Gretell> Unstable branch on crawl.develz.org updated to: 0.26-a0-58-g79b31b8662 (34)
02:09:59 <alexjurkiewicz> ebering: ok, I got it, look out for a big refactor to how those functions work and how the relevant metadata is stored in scoring_data.py
03:15:47 <alexjurkiewicz> ebering: advil: please see this branch with my new work. It's a relatively large change so didn't want to push without giving a chance to review: https://github.com/crawl/dcss_tourney/pull/new/new-scoring-typed-tables
03:18:45 <alexjurkiewicz> better link: https://github.com/crawl/dcss_tourney/compare/new-scoring...new-scoring-typed-tables
03:30:40 <Kelbell> Fork (bcrawl) on crawl.kelbi.org updated to: 0.23-a0-3205-gb897bc9fbb
04:02:24 <alexjurkiewicz> this branch has a couple of bugs. I ripped out the hacks I did to rename clan pages from {team_captain} to {clan_name}-{team_captain}
04:03:06 <alexjurkiewicz> several places need updates to revert their clan page links (and to use the crawl_utils.clan_link in future which I discovered)
04:04:13 <alexjurkiewicz> beyond that, I need to replace all places using the old table logic with new table logic. Only a couple of places remain. Places showing games list (overview, player & clan pages), and top players/clans on overview
04:04:29 <alexjurkiewicz> will get on it in the next day or so
06:43:51 -!- amalloy is now known as amalloy_
06:59:04 -!- amalloy_ is now known as amalloy
07:00:20 -!- amalloy is now known as amalloy_
08:10:22 <Cheibriados> 03Elliot Dronebarger02 07https://github.com/crawl/crawl/pull/1427 * 0.26-a0-45-gdcfaf3d: Some cleanups 10(3 minutes ago, 5 files, 26+ 26-) 13https://github.com/crawl/crawl/commit/dcfaf3ddd825
08:10:22 <Cheibriados> 03Elliot Dronebarger02 07https://github.com/crawl/crawl/pull/1427 * 0.26-a0-46-g4065d8b: Forgot to check linter 10(2 minutes ago, 2 files, 2+ 2-) 13https://github.com/crawl/crawl/commit/4065d8b54ea3
09:44:09 <advil> alexjurkiewicz: probably don't want to blanket enable datatables on e.g. the overview page
09:44:53 <advil> I also am not sure about how well its paging mode works for us, though maybe you just hadn't gotten to adding any config
09:49:06 <advil> code-wise I'm fine with you going ahead and merging this
10:09:28 <Cheibriados> New branch created: remove-habitat-real-amphibiousness (1 commit) 13https://github.com/crawl/crawl/tree/remove-habitat-real-amphibiousness
10:09:28 <Cheibriados> 03Aidan Holm02 07[remove-habitat-real-amphibiousness] * 0.26-a0-59-g2a36f74: Move real_amphibious mons_habitat hack to separate method 10(26 minutes ago, 5 files, 80+ 23-) 13https://github.com/crawl/crawl/commit/2a36f74b56e0
10:20:08 <ebering> alexjurkiewicz: I'll review and merge that today
10:20:15 <ebering> I think I have energy to work normally
11:17:34 <advil> well, if you mean the branch he just posted, give him some more time to work out some ui stuff maybe
11:17:54 <advil> doesn't look like the table formatting is done to me
11:43:18 <ebering> ok
11:43:33 <ebering> it's backend refactor makes adding the transform functions to display additional data in the rankings easy to implement
11:43:40 <ebering> vv nice factorization
11:44:01 <ebering> I'll try to add as many of those as I can
11:45:17 <ebering> alexjurkiewicz: one nit to pick: you re-added displaying the Winning category in the overview! per advil and I's discussion we think it's best to just leave that off in the overview and have it appear in rules and individual/clan pages
11:45:29 <ebering> since it isn't a "ranked category" in the usual sense
11:58:18 <advil> maybe it's still more expedient to merge even if some of the rendering is wip
11:58:58 <advil> I guess sequell is currently dead dead :-/
12:16:04 <Cheibriados> New branch created: pull/1447 (1 commit) 13https://github.com/crawl/crawl/pull/1447
12:16:04 <Cheibriados> 03Elliot Dronebarger02 07https://github.com/crawl/crawl/pull/1447 * 0.26-a0-59-g8a062cd: Added SDL_image to dependency requirements in INSTALL.md 10(5 minutes ago, 1 file, 1+ 0-) 13https://github.com/crawl/crawl/commit/8a062cd4b47c
13:06:00 <Henzell> Unstable branch on crawl.akrasiac.org updated to: 0.26-a0-58-g79b31b8 (34)
13:19:14 <Pinkbeast> AFAICT the use of fake_noisy for orb pickup and apportation and Xom doesn't work (and never has); noisy() lets it go off at a silenced square but it still can't propagate into (through?) silenced squares. Just mentioning it here in case I'm missing something because slapping in a Mantis issue
13:19:38 <Pinkbeast> s/because/before/;
13:40:05 <advil> Pinkbeast yeah, doesn't seem like it propagates, there's noise only at the orb location
13:43:20 <ebering> advil: running it with local test there are still some things to be done with the ui
13:44:16 <ebering> but I think I'm going to merge it to main so I can start building on the "additional display info" using the column tables
13:44:16 <advil> that seems reasonable (do you mean new_scoring?)
13:44:16 <ebering> yes
13:44:23 <advil> I guess we could merge all this into master some day
13:48:33 <Cheibriados> 03advil02 07* 0.26-a0-59-gd69c119: Tweak how seed games are displayed in ?v and chardump 10(2 minutes ago, 3 files, 13+ 8-) 13https://github.com/crawl/crawl/commit/d69c119aaa7d
13:49:12 <Cheibriados> 03advil02 07[stone_soup-0.25] * 0.25-b1-42-gfa7c8ce: Tweak how seed games are displayed in ?v and chardump 10(3 minutes ago, 3 files, 13+ 8-) 13https://github.com/crawl/crawl/commit/fa7c8ce0536d
13:58:03 <Cheibriados> 03advil02 07* 0.26-a0-60-g00fb324: Show seed for seeded play with welcome message 10(63 seconds ago, 2 files, 5+ 2-) 13https://github.com/crawl/crawl/commit/00fb324cd09f
13:58:41 <Cheibriados> 03advil02 07[stone_soup-0.25] * 0.25-b1-43-g9cfca56: Show seed for seeded play with welcome message 10(2 minutes ago, 2 files, 5+ 2-) 13https://github.com/crawl/crawl/commit/9cfca5693d92
14:03:35 <Kelbell> Unstable branch on crawl.kelbi.org updated to: 0.26-a0-60-g00fb324cd0 (34)
14:21:46 <dcss-ci> Build failed for stone_soup-0.25 @ fa7c8ce0 https://github.com/crawl/crawl/actions/runs/128885483 (failed: Catch2/codecov)
14:36:52 <dcss-ci> Build failed for stone_soup-0.25 @ 9cfca569 https://github.com/crawl/crawl/actions/runs/128892089 (failed: Catch2/codecov)
15:01:41 <Cheibriados> 03advil02 07[stone_soup-0.25] * 0.25-b1-44-gc3591d3: Fix catch2 tests for map panning / clamping 10(26 hours ago, 3 files, 34+ 1-) 13https://github.com/crawl/crawl/commit/c3591d366c2b
15:01:41 <Cheibriados> 03advil02 07[stone_soup-0.25] * 0.25-b1-45-g73d902b: Fix viewmap catch2 even more 10(26 hours ago, 1 file, 1+ 0-) 13https://github.com/crawl/crawl/commit/73d902bd6c5a
15:37:23 <ebering> alexjurkiewicz: check out  4ae2c62e269 in new-scoring
15:56:53 <Cheibriados> New branch created: pull/1448 (1 commit) 13https://github.com/crawl/crawl/pull/1448
15:56:53 <Cheibriados> 03shummie02 {GitHub} 07https://github.com/crawl/crawl/pull/1448 * 0.26-a0-61-g36a7bbd: Change = to == in checking crawl_state.type 10(3 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/36a7bbd71978
15:58:31 <Cheibriados> 03shummie02 {advil} 07* 0.26-a0-61-g6c07f0e: Change = to == in checking crawl_state.type 10(5 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/6c07f0ee8d46
15:58:54 <advil> uh oh
16:00:00 <Cheibriados> 03shummie02 {advil} 07[stone_soup-0.25] * 0.25-b1-46-gb8a1ec0: Change = to == in checking crawl_state.type 10(6 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/b8a1ec05dff4
16:01:32 <Kelbell> Unstable branch on crawl.kelbi.org updated to: 0.26-a0-61-g6c07f0ee8d (34)
16:13:11 <travis-ci> The build passed. (stone_soup-0.25 - 73d902b #13587 : advil): https://travis-ci.org/crawl/crawl/builds/696168034
16:16:48 <elliotd123> fastest merge in the west
16:33:19 <shummie> Anyone know whether or not the bug that set the crawl_state.type would cause it to not show up on the scoreboard or if it's unaffected by it?
16:33:19 <shummie> The game still shows up as a "Custom game" when I do ?V and when i log back in
17:00:50 <alexjurkiewicz> actually, I ran into one issue with how my refactoring works
17:00:50 <alexjurkiewicz> you can only format a cell based on its own data
17:01:00 <alexjurkiewicz> which is a problem in the case you want to add a game link to the score column, because that requires the name, src and end date
17:01:35 <alexjurkiewicz> i have a few options for fixing this, I thought about it a lot last night but couldn't pick the better approach
17:01:40 <alexjurkiewicz> will chat more soon
17:03:09 <ebering> alexjurkiewicz: mysql's json functions to the rescue here
17:03:37 <ebering> I've been adding columns to the relevant views that package all the needed info
17:04:05 <ebering> The only tricky one is streaks
17:20:26 <advil> shummie I'm afraid it will mess up which scoreboard it shows up in
17:20:30 <advil> is it on cko?
17:20:37 <shummie> Yes
17:20:57 <advil> trunk or 0.25?
17:21:03 <shummie> I hadn't decided if I should quit the game or not. Playing Op's aren't exactly my favorite thing but i don't want to go through w/ it if it's not going to show up
17:21:06 <shummie> trunk
17:21:11 <advil> let me check something re quitting
17:21:32 <shummie> ok thanks!
17:26:28 <advil> shummie: I'm pretty certain that no milestones ever showed up in the regular milestones file so there are no repercussions in terms of score or sequell for quitting
17:26:45 <advil> if there had been I could have tried to fix the game, but maybe that's not needed
17:27:29 <advil> well, eventually there may be a scoreboard for seeded games, so you would have a quit on that in that hypothetical future
17:28:31 <shummie> so that means there's no reason to continue the game since it won't show up in the scoreboard right?
17:28:36 <advil> yeah
17:28:38 -!- jfcaron_ is now known as jfcaron
17:28:39 <advil> or sequell
17:28:57 <advil> milestones from now on definitely won't but I was worried there would have been a game start milestone
17:29:14 <shummie> well, i was playing on shummie007
17:29:20 <advil> yeah, I found
17:29:22 <shummie> and the game got upgraded to that one
17:29:22 <shummie> ok
17:29:29 <advil> hm
17:29:36 <advil> ok let me double check then...
17:29:50 <advil> actually can you just link to your morgue so I make sure I am checking the right game?
17:30:00 <shummie> https://crawl.kelbi.org/crawl/morgue/shummie007/shummie007.txt
17:30:19 <shummie> This is Dungeon Crawl Stone Soup 0.26-a0-61-g6c07f0ee8d
17:30:23 <shummie> * Milestones
17:30:24 <shummie> That's the ?V shows as well
17:31:17 <advil> ok yeah I should try to fix that game
17:32:04 <shummie> should i quit the game first so that there's no milestone for that either?
17:34:53 <advil> oh I'm looking at the wrong milestones file, no wonder I didn't see it
17:34:58 <advil> don't quit
17:35:04 <advil> do you know how to do a save backup?
17:35:18 <shummie> i saw the instructions somewhere... let me see if ic an find it again
17:35:21 <advil> I can probably try to get it from floraline if not
17:36:07 <advil> https://crawl.develz.org/wiki/doku.php?id=mantis:savebackup
17:37:31 <advil> well, to be clear, the consequences for quitting are that you will have some milestones for the game whenever sequell comes back up, and it will show (by some ways of accounting) as an incomplete game forever. Quitting will not show up in any of the regular scoreboards, won't break streaks. Though the existence of milestones could cause problems with a streak in progress if you have one (looks like not?)
17:38:21 <advil> what I will do with a save backup is edit it so that its game type is not seeded, and request that floraline replace your save with the fixed one, and then you can continue the game as-is
17:39:22 <shummie> http://crawl.kelbi.org/saves/shummie007-crawl-git-6c07f0ee8d-200608-2139.tar.bz2
17:39:22 <advil> I'm happy to do that
17:39:22 <shummie> hopefully i did it right
17:39:22 <shummie> Thank you very much :)
17:39:34 <shummie> of course, watch me kill the guy 2 minutes after you fix it, but that's dcss for ya =D
17:40:52 <advil> yes, that worked
17:43:05 <shummie> thanks again, i know that's the risk of playing on trunk, just usually not completely game ending =D
17:54:24 <advil> lol I kept editing this and being like why isn't it working???? but I'd actually forgotten to pull and recompile after your bugfix
17:58:57 <advil> floraline: when you have a sec could you replace a trunk save for shummie007 on which I've fixed a bad trunk bug?
17:59:40 <advil> ??bad bugs
17:59:40 <shummie> thanks again!
17:59:49 <advil> oh right, no sequell
18:00:03 <advil> I guess I can escape having my first(?) entry in bad/epic bugs :D
18:00:14 <advil> we do already have the xom one which was similar
18:00:17 <shummie> lol
18:00:43 <advil> shummie may take a bit for floraline to be around
18:00:51 <shummie> No worries, I won't be playing for a few hrs anyway
18:01:20 <shummie> I only noticed it because i was like, why am i able to see my game seed when i logged in
18:01:20 <shummie> and saw the commit on github
18:01:20 <advil> yeah...thanks again for spotting it so quickly
18:01:41 <advil> luckily it only rolled out to cko, that would have been ...unfortunate if it had been wider
18:01:44 <shummie> my only code contribution to dcss =)
18:01:54 <advil> heh
18:01:59 <advil> you probably will get a credits entry for that
18:02:11 <shummie> lol
18:19:45 <ebering> ok alexjurkiewicz advil I've ordered the remaining todo cards in a "rough priority to make sure we can launch friday" order
18:20:14 <ebering> imo everything down to the "whereis table" entry is "should do", and everything down to updating the blacklist is "must do"
18:23:35 <Lantell> Unstable branch on underhound.eu updated to: 0.26-a0-61-g6c07f0ee8d (34)
18:39:57 <floraline> advil: sure, where's the file? it is a trunk game right?
18:56:21 <alexjurkiewicz> ebering: great, thanks
18:57:19 <alexjurkiewicz> ebering: one question from reddit: are banner scores cumulative? (eg do you get 4+2+1 for tier 3)? The rules are ambiguous
18:57:55 <alexjurkiewicz> or perhaps the rules standard allows for undefined behaviour by the tourney implementation :)
19:19:31 <alexjurkiewicz> ebering: the clan name conduct rules could be construed as forbidding BLACK LIVES MATTER clan names, which is arguably an advertisement for unrelated content
19:20:05 <alexjurkiewicz> just saw the second half of that commit :) :)
19:22:25 <ebering> alexjurkiewicz: re banners, they are not currently cumulative, the wording neeeds disambiguation
19:23:02 <ebering> alexjurkiewicz: re conduct I guess a deftly applied "commercial" or similar is needed
19:24:55 <alexjurkiewicz> ok, I'll add. I expect there to be some attempts to game this rule regardless of its wording
19:25:16 <ebering> I'm pretty sick of people who bad faith rules lawyer to advance icky causes
19:25:35 <ebering> I have no problem telling them "you're awful get stuffed"
19:25:49 <alexjurkiewicz> about clan names. After the refactor, I think it's best to stick with the legacy approach of using team captain name in most places. Using team name consistently is a little too big a project for this tourney, I think
19:26:26 <ebering> it's just in the columns in the ranking table right?
19:26:37 <ebering> that's ez.pz. within your current framework
19:26:52 <ebering> by just making the query return a json object or array specifyingn name/owner pairs
19:27:00 <ebering> so the link can work w/a transform function
19:27:20 <alexjurkiewicz> hm! I saw you mention that before but I didn't think much about it. It sounds plausible. I'll see what I can do
19:27:45 <ebering> look at what I did with adding morgue links and displaying combo high scores /nemelex combos on the ranking pages
19:29:34 <ebering> I'm going to use some similar trickery with streaks
19:29:58 <ebering> which I thought was a headache but just caching the full streak data as a json object in the streak cache table is easy
20:12:52 <floraline> shummie: save file was copied
20:15:24 <shummie> Thank you!
20:22:12 <Lantell> Stable (0.25) branch on underhound.eu updated to: 0.25-b1-46-gb8a1ec05df
20:24:26 <gammafunk> Err: 1994, type: <type 'str'>
20:24:29 <gammafunk> getting a lot of these from taildb
20:24:43 <gammafunk> er, ah!
20:24:47 <gammafunk> that's my debugging code on that error
20:25:01 <gammafunk> seems I am still getting that string conversion error
20:25:06 <gammafunk> perhaps latest code will fix this?
20:31:34 <alexjurkiewicz> the one you posted about a couple of days ago? it should be fixed with the merge of my big rewrite I posted overnight (it's merged to new-scoring)
20:44:57 <gammafunk> I got an interesting new exception just now about AttributeError: 'module' object has no attribute 'category_table'
20:45:05 <gammafunk> but let me just fetch the latest code and rerun
20:46:27 <alexjurkiewicz> nothing like big refactors three days out
20:57:45 <alexjurkiewicz> ebering: idea for wu jian banner: collect a rune within N turns of entering its run branch
20:58:16 <alexjurkiewicz> Tier 1: 66% of recent games would do it, Tier 2: 20% of recent games, Tier 3: 5% of recent games
20:59:29 <gammafunk> unfortunately that's a bit too lock-step with just low turncount
20:59:59 <gammafunk> you might be able to get at something slightly different if you did
21:00:09 <gammafunk> well
21:00:29 <gammafunk> it's still quite similar but if you really want to amp up the "get this thing fast" thing in a way that's slightly more distinctive
21:00:40 <gammafunk> get a rune within N turns of entering the rune branch level
21:00:47 <alexjurkiewicz> it's also a little cheesable, with both entering an S branch at XL 27 and going for the abyssal rune
21:00:48 <gammafunk> some ways to differentiate it a bit: be specific
21:01:16 <gammafunk> N turns of entering any rune branch level, N turns of entering vaults:5, N turns of entering zot:5 (for the orb)
21:01:28 <gammafunk> last one was a bit of a cheevo some people tracked
21:02:15 <alexjurkiewicz> that sort of matches Uskayaw's banners about getting to certain places within N turns
21:02:57 <gammafunk> er
21:02:57 <gammafunk> to be clear
21:02:57 <gammafunk> that should be "get Foo within N turns of arrive at the branch end level containing Foo"
21:03:03 <gammafunk> so you don't care about the trip down to the branch end
21:03:06 <gammafunk> just what you do at branch end
21:03:14 <gammafunk> so how fast can you get the orb upon entering zot:5
21:03:20 <alexjurkiewicz> oh, interesting
21:03:24 <gammafunk> or how fast get silver rune upon entering V:5
21:03:45 <gammafunk> could just use specific places, like Tier 1: wherever, Tier 2: Vaults:%, Tier 3: Zot:5
21:03:48 <gammafunk> using same cutoff if you like
21:03:59 <gammafunk> a few ways you could do it
21:04:05 <gammafunk> s/%/5/
21:04:17 -!- amalloy_ is now known as amalloy
21:04:52 <gammafunk> it's going to end up being fairly redundant no matter what you pick, but some specificity for the more interesting places could help
21:05:03 <gammafunk> basically asking people to orb ninja
21:05:04 <alexjurkiewicz> that's a nice idea. You could even use the same number for each tier for aesthetic reasons
21:05:15 <gammafunk> yeah, same cutoff
21:05:56 <alexjurkiewicz> so to make sure I understand you correctly, the tiers are: Collect the {rune,orb} within 1000 turns of entering {any rune branch,Vaults:5,Zot:5}.
21:07:01 <gammafunk> well I think 1k is just too generous by far
21:07:33 <gammafunk> and for tier 1, I was more envisioning s/any rune branch/any rune branch end level/, but this does overlap with V:5
21:07:35 <alexjurkiewicz> this mostly overlaps with Okawaru (get score X), and Uskayaw (reach X in Y turns)
21:07:51 <gammafunk> yeah always going to be a lot of overlap
21:08:27 <alexjurkiewicz> oh yes, Tier 1 should only start from the branch end since it's meant to be the easiest
21:09:49 <gammafunk> a more reasonable turn cutoff for this might be 270 if you want the cutoff to actually be difficult; the scaling with this is a bit weird though, since tier 1 and 2 and 3 are actually close in terms of difficulty
21:09:58 <gammafunk> perhaps different cutoffs for different tiers would be good
21:10:18 <gammafunk> so you increase depth of the challenge and decrease available turns simultaneously
21:10:33 <alexjurkiewicz> Another approach, based on your suggestions: Get 2 runes within 1000 turns of entering either rune branch's final level.
21:11:35 <gammafunk> that doesn't seem to offer any new gameplay above just asking for a fast dive
21:12:25 <alexjurkiewicz> I was thinking it lets you scale the tiers better. Change 2 -> 3 -> 4 for each tier
21:12:57 <alexjurkiewicz> travel between branches sort of ruins it though...
21:13:23 <gammafunk> sort of, but as you scale it you're just getting closer to "play for good turncount"
21:13:58 <alexjurkiewicz> yeah. That's true. OK, I'm done brainstorming this. Detritus for ebering to sift through!
21:14:10 <gammafunk> we tried, it's his fault
21:17:58 <alexjurkiewicz> "Wu Jian Council is still bikeshedding their banner, and thus it is not ready in time for this tournament."
21:18:30 <gammafunk> just have a blank and let people write in their own points they feel they deserve
21:23:15 <alexjurkiewicz> you get one point per friend referred to DCSS
22:23:32 <alexjurkiewicz> ebering: what do you think about this view for combo high score details: https://imgur.com/a/upbo4yy
22:23:43 <alexjurkiewicz> (one per line)
22:24:17 <alexjurkiewicz> (I also wonder about removing the numeric columns entirely)
22:40:59 <alexjurkiewicz> I'll commit it. You can back out if you prefer
22:48:47 <gammafunk> alexjurkiewicz: let me know when you do as I'm about to do another restart of scripts after recent changes
22:49:00 <gammafunk> unless this is in a branch
22:49:08 <gammafunk> *another branch
22:50:24 <alexjurkiewicz> hmm, i'll do a big push later today, in my evening
22:51:15 <gammafunk> no problem
22:52:16 <gammafunk> running now so we can see latest changes and if those errors go away
22:53:03 <gammafunk> of course I have to keep moving the test tournament dates forward so it doesn't have so much data that it takes hours to process the first time
22:53:03 <gammafunk> which can make catching these bugs a bit of a problem in the first place
22:55:09 <alexjurkiewicz> fair enough, pushed now
22:56:07 <alexjurkiewicz> I think either combo scoring is wrong or I don't understand it. In my test data, someone has a single eligible game: `HOFi (HO high score)` (it's not a win) and they have 4 points total
22:56:35 <alexjurkiewicz> my counting is: 1 point for combo high score, 5 points for HO high score = 6
22:57:02 <gammafunk> https://crawl.develz.org/tournament/0.25/overview.html is updated
22:57:22 <alexjurkiewicz> ah. Possibly it's "1 point for getting a combo high score, +1 point if it's a win, +3 points if it's a species high score"
22:58:39 <gammafunk> alexjurkiewicz: where's the original image for https://crawl.develz.org/tournament/0.25/images/clan/winning.png ?
22:58:42 <gammafunk> haven't seen that one
22:59:47 <alexjurkiewicz> that's a newish one from dcss-kimchi fork
22:59:54 <alexjurkiewicz> https://github.com/kimjoy2002/crawl/pull/109
23:00:06 <alexjurkiewicz> COUNT(*) + SUM(c.killertype='winning')
23:00:08 <alexjurkiewicz> + 3 * COUNT(sp.raceabbr) + 3 * COUNT(cl.class) AS total,
23:00:08 <gammafunk> alexjurkiewicz: not sure if you've fixed this, but the numeric column alignment for Recent Games is still off; looks like 0 is correctly right-aligned, but non-zero values are left-aligned?
23:00:36 <gammafunk> The Score column is what I'm talking about
23:01:11 <gammafunk> alexjurkiewicz: oh, heh, you should probably use the scarf-free version of that image!
23:01:22 <gammafunk> I can make that change
23:04:04 <alexjurkiewicz> ha ok thanks
23:04:04 <alexjurkiewicz> before you commit it, please crush the image
23:04:04 <alexjurkiewicz> pngcrush -ow -brute in.png && advpng -z4 in.png
23:04:24 <alexjurkiewicz> if it's large (>100kb) I also used `pngquant 64 in.png` before these steps
23:04:46 <alexjurkiewicz> (pngquant will create in-fs8.png, which I would compare with the original to make sure the degradation was not too bad)
23:15:04 <alexjurkiewicz> gammafunk: i see the issue about score alignment
23:15:17 <alexjurkiewicz> i will make sure it's fixed as part of the transition to datatables
23:15:28 <alexjurkiewicz> gotta fix that fugly all-players page first
23:57:52 <advil> thanks floraline!