01:22:09 <Lantell> Experimental (bcrawl) branch on underhound.eu updated to: 0.23-a0-3177-gbaa60938a7
01:38:33 <Gretell> Unstable branch on crawl.develz.org updated to: 0.26-a0-43-g20b394cb32 (34)
02:10:54 <Gretell> Windows builds of master branch on crawl.develz.org updated to: 0.26-a0-43-g20b394cb32
02:41:18 <gammafunk> alexjurkiewicz: does the current clan template have an error?
02:44:23 <gammafunk> alexjurkiewicz: http://dpaste.com/0RQ2G3B
02:45:31 <gammafunk> alexjurkiewicz: I've added a debug statement printing the final query, which you can see at the top; looks like the 'player IN %s' clause has an unsubstituted value
02:46:35 <gammafunk> alexjurkiewicz: looks like line 58 of clan.mako
02:55:13 <alexjurkiewicz> hmm
02:56:54 <alexjurkiewicz> I did add support for the IN operator there
02:57:34 <alexjurkiewicz> i can't replicate with my local data set though. Could you please modify query.py to print(repr(value)) on line 536, just after `elif isinstance(value, (list, tuple)):`?
02:57:54 <gammafunk> sure
02:58:52 <alexjurkiewicz> if this doesn't pinpoint the issue, I'll ask for your loaddb START_TIME etc settings so I can try and replicate locally (plus all the rc wgets)
02:59:13 <alexjurkiewicz> and you can temporarily disable the issue by commenting out the games table in clan.mako with `##` comments
02:59:37 <gammafunk> alexjurkiewicz: line printed is ['cron3']
03:00:30 <alexjurkiewicz> damn. My assumption was that it was an empty list
03:01:37 <alexjurkiewicz> ok, I recommend ocmmenting out that table in clan.mako and I'll get back to you. If you could send me all the tourney config (or push the current state of CDO to the repo) I'll be able to use that too
03:01:44 <gammafunk> alexjurkiewicz: clan that does get through right before the one that doesn't: http://dpaste.com/2NN64QP
03:02:15 <gammafunk> or at least I assume the one before did render properly
03:02:42 <gammafunk> so it got through the first clan and died on the second
03:02:50 <gammafunk> first as in the first clan processed
03:03:08 <alexjurkiewicz> OH. I think it's doing some unexpected tuple unpacking
03:03:20 <gammafunk> right
03:03:27 <gammafunk> was going to ask if clan of one was the cause
03:03:46 <alexjurkiewicz> I've tested clan of one locally, so I'm still not really sure what's going on, but...
03:03:47 <gammafunk> I recall weirdness like that with python
03:04:23 <alexjurkiewicz> please replace query.py line 537 (I guess 538 for you now) with a listified value:
03:04:23 <alexjurkiewicz> append_where(where, key + " IN %s", [value])
03:04:23 <gammafunk> ok
03:05:37 <gammafunk> alexjurkiewicz: now it fails on first clan
03:05:45 <alexjurkiewicz> MySQLdb._exceptions.OperationalError: (1241, 'Operand should contain 1 column(s)') ?
03:06:03 <gammafunk> no, still that same error
03:06:14 <gammafunk> ProgrammingError: (1064, "You have an error in your SQL syntax; check the manual that corresponds to your MySQL server version for the right syntax to use near ') ORDER BY end_time DESC LIMIT 10' at line 1")
03:06:29 <gammafunk> SELECT source_file,src,version,lv,player,uid,race,raceabbr,class,charabbrev,xl,skill,sk_lev,title,place,branch,lvl,ltyp,hp,maxhp,maxmaxhp,strength,int
03:06:32 <gammafunk> elligence,dexterity,ac,ev,god,start_time,duration,turn,score,killertype,killer,kgroup,damage,piety,penitence,end_time,terse_msg,verb_msg,kaux,kills,nr
03:06:35 <gammafunk> une,runes,gold,gold_found,gold_spent FROM games WHERE player IN %s ORDER BY end_time DESC LIMIT 10
03:06:38 <gammafunk> TEMPLATE ERROR
03:06:46 <gammafunk> which isn't helpful I guess since you're not seeing the substitution
03:07:02 <alexjurkiewicz> that's weird. Can you back out that change please?
03:07:03 <gammafunk> or the actual final query
03:07:11 <alexjurkiewicz> yeah. And getting that in python is real hard
03:07:27 <gammafunk> backed out
03:07:33 <alexjurkiewicz> ok, last remote debugging attempt. On line 540 (after the for loop ends), please add these lines:
03:07:35 <alexjurkiewicz> print(repr(where))
03:07:37 <alexjurkiewicz> print(repr(values))
03:08:01 <alexjurkiewicz> for clan pages, we should see output like this
03:08:04 <alexjurkiewicz> loaddb_1  | ['player IN %s']
03:08:06 <alexjurkiewicz> loaddb_1  | [['chequers']]
03:08:08 <alexjurkiewicz> loaddb_1  | ['player IN %s']
03:08:10 <alexjurkiewicz> loaddb_1  | [['Ihryk', 'Illumine']]
03:08:15 <alexjurkiewicz> (that's two clans of one and two people respectively)
03:09:53 <gammafunk> again fails on second clan as before and I get (for both clans):
03:09:53 <gammafunk> ['player IN %s']
03:09:53 <gammafunk> [['backslashecho', 'shittywizard']]
03:09:59 <gammafunk> ['player IN %s']
03:09:59 <gammafunk> [['cron3']]
03:12:50 <alexjurkiewicz> ok, i'm out of ideas. Well, out of easily testable ones
03:13:13 <alexjurkiewicz> I wonder if the mysql connector doesn't support converting a single item list into the appropriate query form
03:13:44 <alexjurkiewicz> eg, it's written the query into `WHERE player IN 'cron3'` instead of `WHERE player IN ('cron3')`
03:14:38 <alexjurkiewicz> https://stackoverflow.com/questions/7071166/print-the-actual-query-mysqldb-runs
03:15:23 <alexjurkiewicz> if you want to try and jimmy this debugging into the query_class.py code...
03:16:01 <gammafunk> I can play more with it later, will look into that
03:16:15 <gammafunk> alexjurkiewicz: do you think you'll be able to make an implementation of the nemelex choice listing?
03:16:45 <gammafunk> I started to look into that but not really sure of the best bootstrap thingy to use
03:21:43 <alexjurkiewicz> i can't this evening but can look tomorrow
03:22:01 <alexjurkiewicz> I've got a bunch of other miscellaneous improvements queued up too
03:22:42 <alexjurkiewicz> animated altar tiles, server url fixes, wording tweaks, player banner display improvements
03:43:11 <gammafunk> well, thanks for the help; we're cutting it close, but hopefully can get this actually working in time
03:43:11 <Kelbell> Fork (bcadrencrawl) on crawl.kelbi.org updated to: 0.22.1-2774-g2bb65ec93b
04:05:55 <alexjurkiewicz> gammafunk: pushed nemelex display for overview. Thanks for pointing to the old overview.mako, I could copy that wholesale and reformat minimally
04:06:08 <alexjurkiewicz> do you want me to disable clan recent games display for now until I can fix it?
04:07:07 <gammafunk> alexjurkiewicz: it's not too important until we need the test to run continuously, but it had been running fine through taildb
04:07:29 <gammafunk> so you can just push the clan fix when you have time
04:07:45 <alexjurkiewicz> alright, sounds good
04:07:51 <gammafunk> ping me and I'll fetch + restart
04:08:55 <gammafunk> er s/until we need the test to run/unless we need the test to keep running/
04:59:36 -!- amalloy is now known as amalloy_
06:03:41 -!- amalloy_ is now known as amalloy
06:06:54 -!- amalloy is now known as amalloy_
09:02:47 <Cheibriados> 03Elliot Dronebarger02 07https://github.com/crawl/crawl/pull/1435 * 0.26-a0-37-g69f780e: Added missing comment 10(2 minutes ago, 1 file, 3+ 1-) 13https://github.com/crawl/crawl/commit/69f780ed706f
09:24:19 <Lantell> Stable (0.25) branch on underhound.eu updated to: 0.25-b1-29-g4cfa5c7bda
09:25:56 <Lantell> peterlawrence (L3 MiFi)  (D:2)
09:25:56 <Lantell> Webtiles server stopped.
09:26:05 <Lantell> Webtiles server stopped.
09:26:06 <Lantell> Webtiles server started.
12:24:59 <Henzell> jtang582 (L13 HuCj) ASSERT(ray.pos() == oldpos) in 'beam.cc' at line 4400 failed. (Gauntlet)
14:16:59 <Cheibriados> New branch created: pull/1443 (1 commit) 13https://github.com/crawl/crawl/pull/1443
14:16:59 <Cheibriados> 03Byrel Mitchell02 07https://github.com/crawl/crawl/pull/1443 * 0.26-a0-44-gb842c5a: Rename Spiny Frogs to Cane Toads 10(6 minutes ago, 25 files, 51+ 43-) 13https://github.com/crawl/crawl/commit/b842c5a3384b
15:28:00 <Cheibriados> 03advil02 07* 0.26-a0-44-gc699729: Add a global scaling option for local tiles 10(4 days ago, 6 files, 46+ 17-) 13https://github.com/crawl/crawl/commit/c699729d39af
15:28:00 <Cheibriados> 03advil02 07* 0.26-a0-45-ga07a990: Fix font glitches at very high device density 10(4 days ago, 1 file, 12+ 10-) 13https://github.com/crawl/crawl/commit/a07a990f6579
15:28:00 <Cheibriados> 03advil02 07* 0.26-a0-46-gedd8c26: Let tile_filter_scaling=false work on mipmapped textures 10(4 days ago, 1 file, 3+ 1-) 13https://github.com/crawl/crawl/commit/edd8c26f9183
15:28:00 <Cheibriados> 03advil02 07* 0.26-a0-47-g55d99e3: Consolidate code for applying game_scale 10(3 days ago, 5 files, 112+ 89-) 13https://github.com/crawl/crawl/commit/55d99e38fcda
15:28:00 <Cheibriados> 03advil02 07* 0.26-a0-48-g98f9fe6: Remove a stray bolt of fire 10(2 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/98f9fe671ce0
15:30:00 <Cheibriados> 03advil02 07[stone_soup-0.25] * 0.25-b1-30-gfd4b7da: Don't print "levelgen mode: deterministic" 10(3 days ago, 2 files, 4+ 4-) 13https://github.com/crawl/crawl/commit/fd4b7daab93b
15:30:00 <Cheibriados> 03advil02 07[stone_soup-0.25] * 0.25-b1-31-g5b5d564: Restore note about pasting to local tiles seed selection 10(3 days ago, 1 file, 5+ 0-) 13https://github.com/crawl/crawl/commit/5b5d564e08f1
15:30:00 <Cheibriados> 03advil02 07[stone_soup-0.25] * 0.25-b1-32-g5c6e1b8: Add a global scaling option for local tiles 10(4 days ago, 6 files, 46+ 17-) 13https://github.com/crawl/crawl/commit/5c6e1b839fe0
15:30:00 <Cheibriados> 03advil02 07[stone_soup-0.25] * 0.25-b1-33-g943f020: Fix font glitches at very high device density 10(4 days ago, 1 file, 12+ 10-) 13https://github.com/crawl/crawl/commit/943f02021d53
15:30:00 <Cheibriados> 03advil02 07[stone_soup-0.25] * 0.25-b1-34-g082b4b2: Let tile_filter_scaling=false work on mipmapped textures 10(4 days ago, 1 file, 3+ 1-) 13https://github.com/crawl/crawl/commit/082b4b2af4fd
15:30:00 <Cheibriados> 03advil02 07[stone_soup-0.25] * 0.25-b1-35-gd6afb6c: Consolidate code for applying game_scale 10(3 days ago, 5 files, 112+ 89-) 13https://github.com/crawl/crawl/commit/d6afb6c36c88
15:30:00 <Cheibriados> 03advil02 07[stone_soup-0.25] * 0.25-b1-36-g55d9e7b: Remove a stray bolt of fire 10(4 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/55d9e7bdc558
15:32:21 <Cheibriados> 03Nikolai Lavsky02 {advil} 07* 0.26-a0-49-g0e6732e: Don't pick up duplicate tins of tremorstones 10(6 days ago, 1 file, 1+ 0-) 13https://github.com/crawl/crawl/commit/0e6732e0a6af
15:33:52 <Cheibriados> 03Nikolai Lavsky02 {advil} 07[stone_soup-0.25] * 0.25-b1-37-g2669288: Don't pick up duplicate tins of tremorstones 10(6 days ago, 1 file, 1+ 0-) 13https://github.com/crawl/crawl/commit/2669288c555a
15:46:46 <Mandevil> Hullo devs.
15:47:12 <Mandevil> I am working on clan stats for my clan in the tournament and I have found log entries that have this field:
15:47:20 <Mandevil> "start=20190930001745S"
15:47:33 <Mandevil> Er.
15:47:36 <Mandevil> Wrong one.
15:48:03 <Mandevil> "end=20190931000038S"
15:48:05 <Mandevil> This one.
15:48:22 <Mandevil> Can you see the problem?
15:48:46 <Mandevil> There are many like this one.
15:48:59 <Mandevil> borek@voyager:~/dev/teamstats$ grep ....0931......S logs/log.cue.24.games |wc -l
15:49:03 <Mandevil> 590
15:49:10 <Mandevil> 590 in CUE/0.24 log
15:49:59 <gammafunk> as you can see, the month field is 0-based
15:50:08 <Mandevil> Oh god.
15:50:09 <gammafunk> so it's not actually a problem
15:50:43 <Mandevil> WHY?
15:50:43 <Mandevil> Month is zero based, but day is 1-based?
15:51:07 <Mandevil> But it superficially looks like almost ISO 8016 date.
15:51:15 <Mandevil> Well done, guys.
15:51:28 <gammafunk> are you trying to troll or what?
15:51:37 <Mandevil> I'm trying to make sense of the log.
15:52:31 <Mandevil> BTW, that 'S' in the end means something?
15:52:38 <Mandevil> And is the date/time in UTC?
15:54:14 <gammafunk> alexjurkiewicz: we had a discussion about the art usage on the tournament pages, and sort of agreed that we shouldn't use any images that we don't have explicit rights to use
15:54:43 <gammafunk> alexjurkiewicz: a lot of that art I sent you is unattributed and without any grants from the original artists, just stuff taken from e.g. korean roguelikes forum
15:55:24 <gammafunk> alexjurkiewicz: so one thing we might do is just use splash images, images from froggy (one person who's given us explicit permission), and only use images that we've otherwise gotten permission to use
15:55:38 <gammafunk> alexjurkiewicz: not high priority, but it's a change I'll probably make tomorrow when I have more time
16:00:55 <gammafunk> not that I think any of those artists would actually be mad at us, but it's the right thing to do
16:01:44 <gammafunk> will at least go through and try to see if we have permission from any of those artists
16:06:11 <advil> Mandevil this inherited from old POSIX localtime (also C89+)
16:06:26 <advil> see e.g. https://linux.die.net/man/3/localtime
16:07:15 <advil> I'm not sure why that was originally chosen (maybe convenience) but it was a while ago and once chosen it's not something that's very easy to change ever
16:07:27 <advil> it's useful to be aware of that date format because it's also used by javascript
16:23:32 <Cheibriados> New branch created: pull/1444 (1 commit) 13https://github.com/crawl/crawl/pull/1444
16:23:32 <Cheibriados> 03Nikolai Lavsky02 07https://github.com/crawl/crawl/pull/1444 * 0.26-a0-50-gfe383db: Update `restart_after_save` default in the options guide 10(9 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/fe383db47f58
16:30:04 <cebolla> have folks ever used the dcss "engine" to make a game?  eg, strip out monsters, races, roles, etc and put in other stuff
16:31:26 <gammafunk> not to my knowledge; the game's internal design is not set up to make that easy
16:37:34 <bh> that would be like slamming your head in a door until you lost consciousness
16:57:23 <Mandevil> advil: Yeah, I knew I saw this peculiar format somewhere :)
17:06:06 <alexjurkiewicz> Mandevil: the format of timestamps is based on struct_t
17:06:36 <alexjurkiewicz> gammafunk: that's fair enough. We have permission for the mummy zigg image too
17:06:54 <gammafunk> oh that's good; one of the ones I was going to ask bout
17:07:13 <gammafunk> going to try printing cursor._last_executed now
17:07:47 <alexjurkiewicz> all the rest are either title images, from the fanart tumblr, or the korean art dump imgur you sent
17:07:47 <alexjurkiewicz> alright. I might be around again later today
17:07:47 <gammafunk> the froggy ones are ok, since I have permission relayed to me from hong
17:07:50 <gammafunk> but that's a secondary concern to just getting stuff working
17:10:57 <alexjurkiewicz> we should probably add some image credits to the credits area anyhow
17:11:22 <alexjurkiewicz> even for title-sourced ones
17:16:44 <nlavsky> It looks like [skip ci] has no effect on crawl's Github Actions. Is there something similar to use for trivial, documentation only, PRs?
17:18:05 <alexjurkiewicz> nope
17:25:36 <nlavsky> I see. Googling shows that there's no standard way and other projects use something like "jobs: ... if: "! contains(github.event.head_commit.message, '[skip ci]')"" for that.
17:26:31 <gammafunk> advil: ebering: alexjurkiewicz: good time to make a final assessment. Do you all think we should delay tournament a week? Current state of affairs: tournament scripts will probably get completed in the nick of time if we can nail down a few issues, no word on Sequell working again soon from snark (sent an email yesterday), I will have enough time to monitor tournament and apply fixes, but might be
17:26:37 <gammafunk> gone for ~10 hours a day 4 days a week starting this week or next, so we need a backup admin around
17:27:02 <gammafunk> 0.25 installed on all servers by cbro, cdo, with both admins for those servers saying they'll get them installed by tournament start
17:27:48 <gammafunk> and I say delay it a week, but if there was a delay, could be e.g. 2 weeks; just wanted to give that as an option to people so we can get that ball rolling if it turns out to be something we need
17:28:18 <gammafunk> and by "that as an option" I mean having a tournament delay at all; does look like we'll have everything necessary to start the tournament ok
17:28:31 <gammafunk> I guess we also have to make the release tomorrow
17:55:40 <ebering> I won't be able for any finishing tasks (actions continue here) before next week. A week delay seems prudent to me
18:08:46 <gammafunk> ok, if anyone else feels that's a problem, please let me know
18:22:45 <Lantell> Unstable branch on underhound.eu updated to: 0.26-a0-49-g0e6732e0a6 (34)
19:01:20 <alexjurkiewicz> suits me. I feel giving people a week to soak in the new rules is good
19:02:46 <alexjurkiewicz> given how many other things have been delayed due to the george floyd protests I think it's inevitable that will be seen as part of the reason why. I'm not sure how the dev team feels about political statements
19:04:24 <gammafunk> alexjurkiewicz: oh I don't think that perception is a problem. I will mention that events in the US are a factor in people not being available
19:06:41 <gammafunk> what I'll do is patch up whatever small issues remain for making the scripts present a working example of the pages with updated tournament dates, then make an announcement on wordpress, then contact all the admins with an update and link to said announcement
19:07:47 -!- Amnesiac_ is now known as Amnesiac
19:30:56 <alexjurkiewicz> gr8. If you can't figure out that clan issue, I'm happy to look into it. Please send me the wgets for all server rc files, and details about any uncommitted changes on the CDO checkout
19:30:56 -!- Tux[Qyou] is now known as goat
19:31:00 -!- goat is now known as Tux[Qyou]
19:55:31 <gammafunk> alexjurkiewicz: heh, this is pretty good; you can debug the final sql statement with something as simple as:
19:56:12 <gammafunk> try:
19:56:12 <gammafunk> cursor.execute(self.query, self.values)
19:56:12 <gammafunk> except Exception as e:
19:56:12 <gammafunk> error("Unable to execute statement:\n"
19:56:12 <gammafunk> "%s\n", cursor._last_executed)
19:56:14 <gammafunk> raise
19:56:18 <gammafunk> and this is what it found:
19:56:41 <gammafunk> 2020-06-04 01:52:54,660 [ERROR] Unable to execute statement:
19:56:41 <gammafunk> SELECT source_file,src,version,lv,player,uid,race,raceabbr,class,charabbrev,xl,skill,sk_lev,title,place,branch,lvl,ltyp,hp,maxhp,maxmaxhp,strength,int
19:56:44 <gammafunk> elligence,dexterity,ac,ev,god,start_time,duration,turn,score,killertype,killer,kgroup,damage,piety,penitence,end_time,terse_msg,verb_msg,kaux,kills,nr
19:56:47 <gammafunk> une,runes,gold,gold_found,gold_spent FROM games WHERE player IN ("'cron3'",) ORDER BY end_time DESC LIMIT 10;
20:08:12 <alexjurkiewicz> wow nice
20:09:30 <alexjurkiewicz> so i guess the mysql query converter doesn't convert tuple args properly
20:09:33 <alexjurkiewicz> it's just done a repr() on the tuple we pass it
20:10:12 <ebering> the mysql wrapper we are using is Not Smart™
20:10:23 <gammafunk> it's pretty old I guess
20:10:45 <gammafunk> so is just that trailing comma the problem?
20:10:47 <ebering> also, politics being seen as part of the reason why: that's the truth, a volunteer spent his time marching instead of writing crawlcode and we're not quite ready
20:10:49 <gammafunk> the double quoting is bizarre
20:10:56 <alexjurkiewicz> append_where(where, key + " IN %s", "(%s)" % ",".join(a))
20:11:09 <alexjurkiewicz> do you want to try this at query.py line 537 before I push it?
20:11:15 <gammafunk> sure, can try
20:11:26 <alexjurkiewicz> ebering: hope you are still safe btw, seen some crazy videos on this side of the world
20:11:40 <ebering> ya all quiet here today, no tussles
20:11:47 <ebering> a tornado just blew through tho
20:12:18 <alexjurkiewicz> unsure if you mean python Tornado or a weather system
20:12:22 <gammafunk> did we just hit 2 million turns on the 2020 game timer or something
20:12:32 <ebering> a weather system
20:15:10 <alexjurkiewicz> sheesh
20:15:27 <ebering> looks like the funnel cloud didn't touch down?
20:16:41 <ebering> power's still on so things seem ok now
20:19:07 <PleasingFungus> devs, i'm here to report a crash
20:19:23 <gammafunk> alexjurkiewicz: oh, I think you means join(value) and not join(a)?
20:19:25 <PleasingFungus> however i have zero idea if sequell will pick up the milestone...
20:19:28 <PleasingFungus> !lm . crash
20:19:38 <gammafunk> it won't, sorry
20:19:40 <Sequell> 6. [2016-09-10 16:00:37] pleasingfungus the Devastator (L24 HEAE of Hepliaklqana) ASSERT(col < MAX TERM COLOUR) in 'options.h' at line 614 failed on turn 56749. (Vaults:3)
20:19:44 <PleasingFungus> rip
20:19:49 <gammafunk> nice 2016 crash tho
20:19:54 <gammafunk> you can find it in your morgue dir though
20:19:55 <PleasingFungus> the crash message was "paperback manual"
20:19:58 <alexjurkiewicz> gammafunk: doh
20:20:31 <PleasingFungus> ok, looking
20:20:36 <gammafunk> alexjurkiewicz: looks like you need quoting in that
20:20:37 <PleasingFungus> http://crawl.berotato.org/crawl/morgue//PleasingFungus/crash-PleasingFungus-20200604-001604.txt
20:20:53 <gammafunk> SELECT source_file,src,version,lv,player,uid,race,raceabbr,class,charabbrev,xl,skill,sk_lev,title,place,branch,lvl,ltyp,hp,maxhp,maxmaxhp,strength,int
20:20:56 <gammafunk> elligence,dexterity,ac,ev,god,start_time,duration,turn,score,killertype,killer,kgroup,damage,piety,penitence,end_time,terse_msg,verb_msg,kaux,kills,nr
20:20:59 <gammafunk> une,runes,gold,gold_found,gold_spent FROM games WHERE player IN '(backslashecho,shittywizard)' ORDER BY end_time DESC LIMIT 10;
20:21:20 <alexjurkiewicz> oh right
20:21:22 <alexjurkiewicz> evil solution: append_where(where, key + " IN %s", "(%s)" % ",".join(repr(value)))
20:21:42 <alexjurkiewicz> probably better: append_where(where, key + " IN %s", "(%s)" % ",".join('"%s"' % v for v in value ))
20:22:07 <gammafunk> that does seem better
20:22:42 <Lantell> Stable (0.25) branch on underhound.eu updated to: 0.25-b1-37-g2669288c55
20:24:39 <PleasingFungus> so it's generating a new level, placing a vault (I think a Vaults vault?), trying to save a manual, and crashing
20:36:11 <alexjurkiewicz> gammafunk: any luck?
20:36:25 <gammafunk> alexjurkiewicz: it's not quite the problem that you think, but I'm working on a fix
20:37:03 <gammafunk> spoiler, this is the problem:
20:37:03 <gammafunk> SELECT source_file,src,version,lv,player,uid,race,raceabbr,class,charabbrev,xl,skill,sk_lev,title,place,branch,lvl,ltyp,hp,maxhp,maxmaxhp,strength,int
20:37:03 <gammafunk> elligence,dexterity,ac,ev,god,start_time,duration,turn,score,killertype,killer,kgroup,damage,piety,penitence,end_time,terse_msg,verb_msg,kaux,kills,nr
20:37:03 <gammafunk> une,runes,gold,gold_found,gold_spent FROM games WHERE player IN '(\"backslashecho\",\"shittywizard\")' ORDER BY end_time DESC LIMIT 10;
20:37:06 <gammafunk> the fix is pretty simple though
20:37:29 <gammafunk> just a case of not quite using the sql cursor's concept of "values" correctly
20:39:49 <gammafunk> success!
20:40:44 <gammafunk> alexjurkiewicz: what I went with, which works for all test data:
20:40:45 <gammafunk> +       append_where(where, key + " IN (" +
20:40:45 <gammafunk> +               ",".join(['%s'] * len(value)) + ")", *value)
20:41:23 <gammafunk> I'm not sure how to feel about building the statement template up dynamically based on input data like that
20:41:42 <gammafunk> but what was happening is that any attempt to quite before passing it as a cursor value runs into escaping problems
20:42:34 <gammafunk> since I guess it just can't deal with list-like objects properly?
20:42:34 <gammafunk> also a bit awkward given how that append_where() sub-function works
20:42:38 <gammafunk> we can check the pages now though
20:42:53 <alexjurkiewicz> well, excellent
20:42:57 <alexjurkiewicz> nice fixup
20:43:04 <gammafunk> hrm, getting 404
20:43:11 <gammafunk> https://crawl.develz.org/tournament/0.25/teams.html
20:43:15 <gammafunk> this still loads, but not the clan page?
20:43:52 <alexjurkiewicz> can you ls html.tourney0.25/clans
20:44:09 <alexjurkiewicz> see what page names it's generating
20:46:19 <ebering> are y'all still pushing changes to the github repo
20:46:31 <ebering> I haven't pulled since last night but git pull says all up to date
20:46:43 <alexjurkiewicz> I just pushed now
20:46:53 <gammafunk> ebering: yeah, alex has been pushing stuff
20:46:57 <alexjurkiewicz> one related fix to the above I just pushed: "Fix clan links on all player rank page"
20:47:12 <gammafunk> I have local cdo stuff, only some of which can be pushed (since it has config info)
20:48:04 <gammafunk> alexjurkiewicz: hrm, pages there, e.g.: League_of_Redundancy-cron3.html
20:48:26 <ebering> hm now pull pulled a bunch of stuff
20:48:30 <ebering> unclear what was happening there
20:48:36 <gammafunk> alexjurkiewicz: oh, case issue
20:48:36 <ebering> clan_name = c.fetchone()[0]
20:48:37 <ebering> TypeError: 'NoneType' object has no attribute '__getitem__'
20:48:51 <ebering> when I run on 0.24 test data locally
20:48:57 <alexjurkiewicz> did you rebuild database?
20:49:24 <ebering> I did not!
20:49:25 <alexjurkiewicz> gammafunk: in update_page.py, try:
20:49:27 <alexjurkiewicz> -    dest = '%s/%s-%s' % (crawl_utils.CLAN_BASE, clan_info[0], captain.lower()),
20:49:29 <alexjurkiewicz> +    dest = '%s/%s' % (crawl_utils.CLAN_BASE, clan_info[2]),
20:50:27 <gammafunk> alexjurkiewicz: ok, but I think we do need that captain, since duplicat clan names
20:50:45 <PleasingFungus> cool, my crash happens every time
20:50:47 <alexjurkiewicz> ebering: that is a Shouldn't Happen error, if you still get it please print repr(captain) immediately before that line
20:50:49 <gammafunk> (you may expect a few clans named BLACK LIVES MATTER, for example)
20:51:28 <alexjurkiewicz> gammafunk: clan_info[2] is the "clan page name", with case and special characters all fixed up
20:51:33 <gammafunk> gotcha, great
20:52:41 <alexjurkiewicz> yeah. All the code I've written was built to allow multiple clans to have the same name. But I didn't test that, and there might be some issues
20:52:46 <gammafunk> PleasingFungus: yeah, sorry we're in the middle of tourney patch-up; was going to suggest downloading the save, but you'd have to debug from that
20:53:04 <gammafunk> I feel this specific bug has been fixed by an advil commit, but can't offer more
20:53:35 <gammafunk> I do recall some kind of manual crash in trunk at one point
20:53:35 <gammafunk> specific to 0.25-a, I mean
20:53:47 <gammafunk> alexjurkiewicz: well, I guess positive side effect of delay is we can test that!
20:54:26 <PleasingFungus> i did download the save, yeah
20:54:29 <PleasingFungus> sorry, didn't mean to nag
20:54:31 <PleasingFungus> just thinking out loud
20:55:10 <advil> that is a wild stack trace
20:55:35 <advil> I would generally have hoped that level excursions are impossible in levelgen
20:55:56 <advil> also that levelgen wouldn't update your shopping list
20:56:43 <gammafunk> PF what did you do!
20:57:43 <PleasingFungus> i imported a save from 2017
20:58:01 <advil> it's maybe an upgrade issue but I'm not sure
20:58:03 <gammafunk> alexjurkiewicz: ok, that fixed the links: https://crawl.develz.org/tournament/0.25/clans/futureplayerghosts-eyebeam.html
20:58:32 <gammafunk> er, well, I guess that's not an example of a link, but the link to that page is right
20:59:05 <advil> re tournament, if we do go forward I can act as sysadmin
20:59:26 <alexjurkiewicz> kewl
20:59:54 <advil> re that crash, basically if ShoppingList.item_type_identified can ever be triggered while placing a manual, I would expect it to always be if the manual is on your shopping list
21:00:00 <advil> so I suspect it isn't an upgrade issue
21:01:27 <advil> also did you do a save backup
21:02:32 <alexjurkiewicz> gammafunk: when you announce the new date, could you please put out the call for anyone with frontend / ux experience to help improve the site, and for feedback to be posted in the reddit thread or on the blog post thread?
21:06:07 <advil> ah are we postponing?
21:08:11 <gammafunk> advil: well, ebering said it would be prudent when I asked the channel earlier
21:08:29 <gammafunk> because he doesn't have time this week under present circumstances
21:08:41 <gammafunk> and just because circumstances this week are awful for a lot of people in general
21:08:56 <advil> yeah, that's fine
21:09:06 <advil> I just hadn't read all the backlog
21:09:09 <PleasingFungus> advil: yeah i have a save backup
21:09:19 <PleasingFungus> and repro'd it locally
21:09:22 <PleasingFungus> it's consistent, which is nice
21:10:07 <PleasingFungus> http://crawl.berotato.org/crawl/saves/PleasingFungus-crawl-git-20b394cb32-200604-0028.tar.bz2
21:39:45 <gammafunk> alexjurkiewicz: looks like conqueror banner is bugged: https://crawl.develz.org/tournament/0.25/players/p0werm0de.html
21:40:03 <gammafunk> powermode has only a single 3 rune win worth ~2.6M
21:40:13 <gammafunk> yet he's given all 3 tiers of that banner
21:46:31 <ebering> hm I'm not sure I updated all the banner adding code, one of my remaining todos was to double check all of them
22:11:35 <alexjurkiewicz> I don't do maths, i just print what the boss tells me
22:11:54 <ebering> :P