00:15:55 <alexjurkiewicz> hellmonk: d:15 currently has orc + vaults shortcut + depths shortcut. I suggest moving depths to end of S branch
00:15:55 <Sequell> alexjurkiewicz: You have 1 message. Use !messages to read it.
00:18:09 <alexjurkiewicz> !messages
00:18:10 <Sequell> (1/1) hellmonk said (2h 14s ago): sounds like a good idea, removing it also sounds like a good idea and is prob easier but I can try to rewrite the spectral weapon code so it works as permabuff
00:24:15 <alexjurkiewicz> !source monster_info
00:24:16 <Sequell> 1/5. https://github.com/crawl/crawl/blob/master/crawl-ref/source/mon-info.cc#L423
00:24:20 <alexjurkiewicz> !source monster_info 2
00:24:21 <Sequell> 2/5. https://github.com/crawl/crawl/blob/master/crawl-ref/source/mon-info.cc#L303
00:24:23 <alexjurkiewicz> !source monster_info 3
00:24:23 <Sequell> 3/5. https://github.com/crawl/crawl/blob/master/crawl-ref/source/mon-info.h#L237
00:36:22 <hellmonk> the ideal way is probably making it act like hep ancestor and then raising the spell level a lot so it's not overpowered
00:37:04 <alexjurkiewicz> the damage numbers are pretty tweakable, you could keep the spell level the same quite easily
00:38:25 <alexjurkiewicz> hm, can monster_infos update dynamically?
00:38:34 <alexjurkiewicz> eg if I want to add some data that updates during the game
01:01:38 <gammafunk> !seen elliptic
01:01:38 <Sequell> I last saw elliptic at Mon Aug 14 03:45:03 2017 UTC (1h 16m 34s ago) saying 'maybe we just need to have a couple month-long tourneys again, to kill off the 10 hours a day people' on ##crawl.
01:01:54 <gammafunk> harsh
01:02:49 <gammafunk> !tell elliptic I was interested in implementing that xp tracking for spawns. You mentioned you had a branch for this but ran into some technical issues, can you share that code and what the issues are?
01:02:49 <Sequell> gammafunk: OK, I'll let elliptic know.
01:19:10 <elliptic> gammafunk: no, the only branch I have is an old one that did some of the work required for aut_scoring
01:19:10 <Sequell> elliptic: You have 2 messages. Use !messages to read them.
01:19:31 <gammafunk> oh, maybe I misremembered then
01:19:39 <elliptic> and I stopped because I wasn't sure how to handle the spawn timer
01:19:51 <gammafunk> ah, that was the connection to the spawning thing I was remembering
01:20:01 <elliptic> and the xp tracking for spawns would be useful for deciding how to handle the timer, yeah
01:20:32 <elliptic> so please go ahead and implement the tracking if you are interested, but I don't have any code for it :)
01:20:48 <gammafunk> sure, hopefully it won't be difficult
01:31:17 -!- Menche_ is now known as Menche
01:38:33 <gammafunk> hrm, how would we want to report that
01:38:48 <gammafunk> I suppose the per-branch table like we do for some other morgue things?
01:39:08 <gammafunk> ??options
01:39:08 <Sequell> rcfile[1/4]: https://github.com/crawl/crawl/blob/master/crawl-ref/docs/options_guide.txt
01:57:20 <Gretell> Windows builds of master branch on crawl.develz.org updated to: 0.21-a0-193-g085b6a0
02:52:12 <Gretell> Monster database of master branch on crawl.develz.org updated to: 0.21-a0-193-g085b6a0
03:11:30 <Rotatell> Unstable branch on crawl.beRotato.org updated to: 0.21-a0-193-g085b6a0 (34)
04:26:58 -!- amalloy is now known as amalloy_
09:20:52 -!- pantaril_ is now known as pantaril
11:32:51 <advil> "It [making an options menu with gui] is not even hard to do (I did something like that in 2 programming languages but I don't know C++ good enough). It can take literally 2-3 days I believe."
11:32:55 <advil> hahahahaha
11:45:24 <|amethyst> that would be true if we only had to do it for webtiles
11:45:32 <advil> yeah, I suspect that's what he has in mind
11:45:41 <|amethyst> but
11:45:47 <|amethyst> that also ignores things like list options
11:45:56 <|amethyst> unless we want to make people type in comma-separated lists
11:47:17 <advil> I probably wouldn't estimate less than months, at least at my spare-time speeds of programming
11:47:25 <advil> including testing
11:51:23 <|amethyst> probably it's only a week of sitting-down-and-working-on-nothing-else to get a prototype?
11:51:32 <|amethyst> but yeah, ain't nobody got time for that
11:51:35 <advil> seems plausible
11:51:49 <shmup> trying to get crawl built in msys2 and was building the dependencies in contrib and, any thoughts about this? https://gist.githubusercontent.com/shmup/34fe32f06c1080171054065d2e777bae/raw/1a9196263a9b8d0e52283b41fdc22ba2a74a6c17/gistfile1.txt
11:52:28 <advil> hm, haven't seen that one before
11:53:20 <advil> |amethyst: can I mention that estimate in the tavern thread?
11:53:36 <|amethyst> advil:  sure, but remember I'm not a software engineer
11:53:42 <|amethyst> advil:  that is not a professional estimate :)
11:54:36 <advil> well, tavern didn't pay you either, so they're getting their money's worth
11:54:53 <|amethyst> advil:  hmm,. that's weird
11:54:57 <|amethyst> err
11:54:58 <|amethyst> oop
11:55:02 <shmup> link thread please
11:55:03 <|amethyst> shmup:  hmm, that's weird
11:55:47 <|amethyst> shmup:  because that file does #include <stdlib.h>, and it has a mechanism in place to define setenv if it's not present
11:57:05 <|amethyst> oh, hm
11:57:25 <|amethyst> I see, our Makefile doesn't run configure on the submodules
11:58:17 <|amethyst> not sure of a good way to fix that in the distribution, but
11:59:21 <|amethyst> you can try editing  contrib/sdl2-mixer/Makefile   and removing  -DHAVE_SETENV  from the CLFAGS
11:59:50 <|amethyst> oh
11:59:55 <|amethyst> actually another thing you might try
12:00:07 <advil> shmup https://crawl.develz.org/tavern/viewtopic.php?f=17&t=23989&p=319217#p319217
12:00:11 <shmup> that worked
12:00:19 <shmup> but i'd like to learn your other idea anyways, |amethyst
12:00:20 <advil> it's the 0.21 improvements thread
12:01:06 <shmup> oh no it did not work, |amethyst. i jumped the gun.
12:01:12 <|amethyst> hm
12:01:17 <|amethyst> my other idea probably won't work
12:01:41 <shmup> external/libmikmod-3.1.12/drivers/drv_nos.c:72:19: error: variable 'drv_nos' definition is marked dllimport
12:02:21 <|amethyst> hm
12:02:31 <|amethyst> now we get into Windows stuff that I have no clue about
12:03:52 <|amethyst> hm
12:05:15 <|amethyst> https://stackoverflow.com/questions/20644151/error-function-definition-is-marked-dllimport   suggests adding  -DAVT_VMBAPI_CPP_EXPORTS  to the cflags while building libmikmod
12:05:33 <|amethyst> but I'm not sure why this would only have started happening now
12:06:00 <|amethyst> since clearly people have built with msys2 before now, and we haven't changed those contribs
12:06:38 <shmup> hmmm
12:07:36 <advil> pretty sure johnstein is building with msys2
12:09:01 <johnstein> I am
12:09:07 <shmup> johnstein, what version of msys2 do you use and do you install your tools with pacman?
12:09:12 <gammafunk> I saw that sandman "options UI estimate"
12:09:17 <gammafunk> pretty good stuff
12:09:31 <advil> I posted a response just in case someone believes him
12:09:36 <johnstein> shmup: I used whatever the latest version was a few weeks ago
12:09:44 <johnstein> I can check later when I'm home
12:09:49 <shmup> all right thanks
12:09:52 <johnstein> I used pacman for some things
12:09:59 <johnstein> I had to fix sdl2
12:10:07 <shmup> well, here we have it lol
12:10:12 <johnstein> that was the only blip
12:10:14 <advil> that was for a bug that presented differently, though
12:10:17 <shmup> oh
12:10:40 <shmup> to verify: you built all the stuff in source/contrib ?
12:11:52 <johnstein> seems like it
12:12:20 <johnstein> I didn't have to screw around with getting contribs elsewhere
12:12:29 <johnstein> I did have to install a few pacman things but I don't think I kept a list
12:12:53 <shmup> it was probably the standard stuff yeah
12:13:07 <shmup> base-devel and git and mingw toolchain
12:13:35 <johnstein> yea. it felt like standard stuff
12:34:45 <gammafunk> hrm, dump_kill_breakdowns is mentioned in the options.txt table at the top, but not described
12:40:10 <gammafunk> looks like there's support for this in dat/clua/kills.lua
12:40:10 <gammafunk> but I don't think it's possible to use currently
12:56:56 -!- amalloy_ is now known as amalloy
12:58:15 <Cheibriados> 03johnstein02 {amalloy} 07* 0.21-a0-194-g8a2c2cd: Make Dith hate ALL fire spells (bepbepimjep, #10969) 10(5 days ago, 4 files, 11+ 1-) 13https://github.com/crawl/crawl/commit/8a2c2cdf8838
13:04:34 <johnstein> can't wait to see what unintended bug that one introduced
13:05:00 <johnstein> I'm sure it's ok though
13:05:24 <Cheibriados> 03johnstein02 {amalloy} 07* 0.21-a0-195-gc079aee: Fix some spell highlighting 10(34 hours ago, 3 files, 14+ 7-) 13https://github.com/crawl/crawl/commit/c079aeec1028
13:05:37 <amalloy> i'll ruin the experiment by merging two at once
13:05:37 <Sequell> amalloy: You have 2 messages. Use !messages to read them.
13:05:48 <amalloy> now you'll never know which introduced the bug
13:05:53 <johnstein> whew
13:09:02 <Jorgrell> Unstable branch on crawl.jorgrun.rocks updated to: 0.21-a0-194-g8a2c2cd (34)
13:10:01 <amalloy> johnstein: i was thinking about how we could improve the codebase so that a change to spell_highlight_by_utility to make it better in one place doesn't have ripple effects making it worse in another place. i think basically the problem is that "utility" means a lot of different things in different contexts, and we're trying to pack all the information into too small a return type (a colour)
13:10:46 <johnstein> hmmm
13:10:56 <amalloy> we compensate for this by passing it flags to say what context we're in, but then it's basically several different functions glued together, depending on what flags it gets. better would be to have it return a larger result: a struct containing several utility fields
13:11:41 <amalloy> and then callers can evaluate that struct based on their context
13:11:45 <johnstein> yea. then the calling function gets to
13:11:47 <johnstein> yea
13:11:56 <johnstein> that sounds somewhat sane
13:14:41 <amalloy> are you interested in trying to put together a design/patch for that, or are you never touching highlights again?
13:16:02 <johnstein> I don't believe I've done enough coding to swear off any particular section of it yet :)
13:16:11 <johnstein> I can make an attempt
13:16:54 <amalloy> it's up to you. nothing wrong with making me eat my own dogfood if it turns out to be a silly idea
13:18:19 <johnstein> I can wrap my head around it so it sounds like a pretty good task
13:18:26 <amalloy> great
13:21:24 <johnstein> if you get a few minutes can you take a look at that prism/spire trap interaction PR? no rush but I was surprised to see doesnty's fix.
13:22:58 <Henzell> Unstable branch on crawl.akrasiac.org updated to: 0.21-a0-194-g8a2c2cd (34)
13:24:55 <amalloy> that one's a little less in my domain; i mostly stay away from game-design commits (what is the right thing to do according to the Crawl Philosophy) and stick to code health, which i'm better at
13:26:24 <amalloy> the one thing i noticed about yours is that it seems weird to say "shimmering outline" to refer to a shaft; the text makes a lot more sense for invisible monsters
13:38:18 <johnstein> yea. shimmering sounded weird for shafts
13:39:13 <johnstein> ghostly outline works well for invis monsters. could just allow it to ID the trap like |amethyst suggested in mantis
13:39:40 <amalloy> finally ultraviolent will get a 100% trap-safe Su game
13:40:13 <johnstein> not sure how interesting if a situation it would actually be to know there's either a shaft or teleport trap there. not sure if it would affect decision making much at all
13:41:22 <johnstein> interesting OF
14:09:04 <Jorgrell> Unstable branch on crawl.jorgrun.rocks updated to: 0.21-a0-195-gc079aee (34)
14:18:57 -!- amalloy is now known as amalloy_
14:27:05 -!- amalloy_ is now known as amalloy
14:58:14 <gammafunk> .gfgk
14:58:15 <Sequell> 678. iamoutofideas the Chopper (L8 MiFi of Trog), blasted by a gargoyle (stone arrow) on D:6 (gammafunk_temple_overflow_arena) on 2017-08-14 17:00:53, with 1397 points after 4938 turns and 0:12:13.
14:58:29 <gammafunk> I love it when players send messages like that through their usernames
14:59:00 <shmup> what name seems best for a config setting that enables/disables 'easy_modifier_keys'. / and * that is. a thing |amethyst and I were chatting about a few days ago
15:01:24 <gammafunk> is that referring to how * as a key prefix is the same as CTRL?
15:02:01 <shmup> yeah, you got it. the idea is, I might like to bind / to search, and at the moment I have to tap / twice to activate it
15:02:24 <shmup> with my patch, setting easy_modifier_keys to false, I only have to hit / once
15:02:42 <gammafunk> hrm, well you just referred to it as enabling "easy_modifier_keys"
15:02:53 <gammafunk> I take it that this is a proposed name? is that a bad one?
15:02:55 <shmup> yeah i am doing that on purpose :P but also open to better names
15:03:00 <gammafunk> seems like an ok name to me
15:03:22 <shmup> all right, cool
15:03:22 <gammafunk> maybe "easy" is bad
15:03:33 <gammafunk> "alternate_modifier_keys" ?
15:04:03 <gammafunk> but yeah, I don't think the name has to be too perfect or anything; maybe others have stronger opinions
15:04:14 <shmup> i just chose the "easy_eat_chunks" option
15:04:19 <shmup> but really I am open to whatever
15:04:28 <gammafunk> yeah, and what's good is that people will only start to object
15:04:34 <gammafunk> *after* you've coded it
15:04:40 <|amethyst> modifier_prefix_keys = true   perhaps
15:04:40 <shmup> haha, i'll open the question in the PR
15:04:54 <shmup> that's not bad, and I like it more than alternate
15:05:13 <gammafunk> take_it_easy_modifier_keys
15:05:20 <|amethyst> even nicer would be   modifier_prefix_keys = */   but that is a bit outside the scope of what you were doing :)
15:05:38 <|amethyst> (and has the potential to break things if set to something bad)
15:07:41 <amalloy> i would totally use a config setting that turns off / and * so i can bind them to something useful. even just mapping / to ?/ would be handy
15:08:11 <shmup> that sounds like a nice thing to have bound for sure
15:09:00 <shmup> i'm gonna adopt the use_* naming convention
15:09:13 <shmup> the one |amethyst suggests is nice if I really am doing the = */ thing
15:09:31 <shmup> aka "use_fake_player_cursor"
15:52:18 <gammafunk> elliptic: btw, regarding tracking that spawn XP, I'm wondering how much detail that needs. If it's by branch like some of the other dump tables, will that be sufficient?
15:52:26 <gammafunk> Will probably also need a grand total
15:52:56 <gammafunk> maybe it needs to summarize totals xp values as well as percent of XP within the branch and percent of XP overall?
15:53:41 <gammafunk> I suppose just the two values, total XP by branch and total, and percentage of XP for the branch and for overall
15:54:33 <elliptic> not sure
15:55:18 <elliptic> if you just count up xp by branch then you'll get stuff like D:1-3 not mattering at all
15:55:27 <gammafunk> yeah
15:55:32 <elliptic> since it will be insignificant next to late D
15:55:37 <gammafunk> could do by level, but then that's a pretty big table in a morgue
15:55:52 <gammafunk> Which I think is the only way we can present this data
15:56:11 <gammafunk> but it's more of a debug table that we could enable in e.g. qw games or our own games
15:56:14 <gammafunk> so maybe that's fine
15:57:14 <Cheibriados> New branch created: pull/597 (1 commit) 13https://github.com/crawl/crawl/pull/597
15:57:14 <Cheibriados> 03Jared Miller02 07https://github.com/crawl/crawl/pull/597 * 0.21-a0-165-gbecc837: Toggle setting for modifier prefix keys / and * 10(9 minutes ago, 4 files, 26+ 15-) 13https://github.com/crawl/crawl/commit/becc837c3f19
15:57:17 <elliptic> yeah, and I don't think it would be much bigger than stuff like the list of generated vaults
15:57:20 <gammafunk> alright, so I'll do by level
15:58:05 <gammafunk> but for the actual values, it shouldn't be necessary to do more than the total spawn XP of that level and spawn XP as a percentage of total gained on the level
15:58:41 <gammafunk> I guess it could also just include non-spawn XP
15:59:18 <elliptic> the percentage of total is the only really important number
15:59:43 <elliptic> but having the total XP value might be okay for comparing different games
15:59:53 <gammafunk> and a monster count?
16:00:04 <gammafunk> probably not too meaningful I guess
16:00:56 <elliptic> probably I'd just do:   spawn xp, non-spawn xp, spawn percentage of total (spawn / (spawn+non-spawn))
16:01:19 <elliptic> but if you want monster count or anything else then might as well add it
16:01:31 <gammafunk> alright, I'll start with that; monster count might also be interesting to see
16:01:38 <gammafunk> I honestly have no idea how many I'm fighting
16:01:46 <elliptic> yeah
16:02:11 <elliptic> btw, this table should probably only appear in finished games
16:02:46 <gammafunk> right
16:02:53 <elliptic> since otherwise it could be used to check whether a monster that you just killed was a spawn or not, which is very minorly relevant in places like zot:5
16:03:32 <gammafunk> just the sort of tactics qw would employ
16:30:44 -!- Tuxedo[Qyou] is now known as told_you
16:30:55 -!- told_you is now known as Tux[Qyou]
16:59:42 <Cheibriados> 03Jared Miller02 07https://github.com/crawl/crawl/pull/597 * 0.21-a0-165-g9ebbd0a: Toggle setting for modifier prefix keys / and * 10(71 minutes ago, 6 files, 27+ 18-) 13https://github.com/crawl/crawl/commit/9ebbd0a54f34
17:00:41 <amalloy> |amethyst: ^ looks good to me, but i wanted to give you a chance to review before i merge it
17:04:39 <|amethyst> looks good to me as well, though the documentation could go into a little more detail
17:04:58 <|amethyst> might also be good to double-check that none of our built-in Lua stuff relies on those
17:07:30 <|amethyst> I don't see anything relevant
17:08:00 <|amethyst> though I notice that automagic has not been updated to use do_commands the way autofight has
17:35:01 <shmup> what's the message you'd like docs to get across, |amethyst?
17:35:05 <shmup> why * and / are useful to begin with?
17:35:52 <shmup> (cases of not having shift/ctrl right)
17:37:21 <shmup> oh, I see your example in the comment, nvm
17:44:17 <geekosaur> shift and control arrows are not guaranteed to be meaningful in terminals; browsers on non-mac often do special local things with control
17:44:43 <|amethyst> yeah, giving an example of why you might want the option would be good too
17:45:37 <geekosaur> (you kinda don't want control-w to close the browser window when you intended to set a waypoint)
18:14:09 <Lantell> Unstable branch on underhound.eu updated to: 0.21-a0-195-gc079aee (34)
18:41:58 <shmup> it does in chrome though, geekosaur, right? *w does not, however. i mean ^w will warn you about closing, though
18:58:08 <Cheibriados> 03Jared Miller02 {amalloy} 07* 0.21-a0-196-g6a663c0: Toggle setting for modifier prefix keys / and * 10(3 hours ago, 6 files, 27+ 18-) 13https://github.com/crawl/crawl/commit/6a663c05849a
18:59:13 <amalloy> shmup: geekosaur's point was that *w is useful because ^w closes the window in some browsers
18:59:54 <shmup> yeah, realized that since heh. was saying in another channel, would a webtiles setting that said to respect in game modifier keys First be ok? it'd be opt in, so people by default had their normal browser behavior
19:00:08 <shmup> if the setting did exist, it'd preventDefault on ^w etc
19:00:48 <shmup> shift-tab as well
19:02:14 <amalloy> i'm not a good person to answer that question. i don't play webtiles, and if i did i close the window a lot more often than i set waypoints, so i wouldn't turn it on
19:09:09 <Jorgrell> Unstable branch on crawl.jorgrun.rocks updated to: 0.21-a0-196-g6a663c0 (34)
20:04:10 <alexjurkiewicz> gammafunk: not sure exactly what the purpose of the xp table is, but it would be cool to see a 27 line table showing the breakdown by source level for the XP required to reach each XL. Eg XL2: 100% D:1 \n XL3: 12% D:2 88% D:3