00:13:43 <amalloy> |amethyst: is there a good way to join a vector<string> with a delimiter in crawl? i was surprised to see there doesn't seem to be anything good for it in the standard library
00:14:45 <amalloy> alternatively, even something more special-cased that can turn ["x", "y", "z"] into "x, y, and z" would be even better if we have it
00:15:54 <amalloy> i was going to suggest johnstein use a vector<string> in his halo PR rather than checking whether a string is empty before appending " and " to it, but if there's no nice way to turn that into a joined string the benefit is not obvious
00:18:51 <gammafunk> oh, there's some stringutil function we have like that
00:18:56 <gammafunk> I used it
00:19:55 <gammafunk> amalloy: comma_separated_line() ?
00:20:19 <gammafunk> allows you to specify a "comma" and an optional final delimiter
00:20:26 <gammafunk> both as trings
00:21:07 <amalloy> wow, that is quite a template
00:21:22 <gammafunk> s/trings/strings/
00:21:30 <amalloy> gammafunk: looks useful, thanks
00:23:26 <johnstein> amalloy, yea, I figured there had to be a less clunky way to handle those strings. thanks for looking into it
00:36:43 <alexjurkiewicz> hellmonk: animate dead feels really useless at low spellpower
00:37:29 <alexjurkiewicz> two ideas: a) spellpower changes duration of summons b) remove it from Ne starting book and bump the level up to like 6
00:37:54 <hellmonk> yeah watertreatmentrl mentioned the same thing, effect is a little too weak rn
00:38:15 <hellmonk> can probably just bump the chance at low power
00:52:53 <nikheizen> is that a cdda joke
00:59:18 <espais> !messages
00:59:18 <Sequell> (1/1) Napkin said (9h 2m 58s ago): all hail espais! reCaptcha is live!
01:00:26 <espais> !tell napkin yay! glad it works :)
01:00:27 <Sequell> espais: OK, I'll let napkin know.
01:00:40 <espais> seriously though, nobody let gamma in
01:13:06 <gammafunk> too late
01:13:32 <gammafunk> espais: https://cdn.discordapp.com/attachments/205316046230388737/352181745069588491/unknown.png
01:13:43 <gammafunk> that was hours ago, I already have complete control of the mainframe
01:13:55 <gammafunk> and I have nearly hacked the Core's quantum encryption
01:13:58 <gammafunk> it is only a matter of time
01:15:50 <espais> and this is me
01:15:54 <espais> handing you the keys to the mainframe
01:15:54 <espais> https://i.makeagif.com/media/6-24-2015/202wbR.gif
02:52:34 <Gretell> Monster database of master branch on crawl.develz.org updated to: 0.21-a0-227-g58a7c62
03:11:33 <Rotatell> Unstable branch on crawl.beRotato.org updated to: 0.21-a0-227-g58a7c62 (34)
03:18:37 -!- amalloy is now known as amalloy_
09:28:54 <shmup> glad you chose to update captcha instead of mantis, although i suggested updating mantis. i like this old ass look
09:50:38 <espais> i must be blind, are the player character sprites in dat/tiles/player.png?
09:54:44 <|amethyst> espais:  they start way down, after the orb guardian fetus and the net tile
09:54:45 <Sequell> |amethyst: You have 1 message. Use !messages to read it.
09:55:56 <|amethyst> espais: also, they're all in their underwear or less, since they have paper doll clothing
09:56:11 <espais> ah, naturally
09:56:30 <espais> actually i forgot about the paper doll thing
09:56:48 <espais> one *possible* issue i noticed is that draconians don't get a hat sprite when they put on a hat
09:58:07 <espais> octopodes get a little rag on their head...i kind of thought dracs should get something too
09:58:28 <|amethyst> is there even room?
10:01:22 <|amethyst> ah
10:01:35 <|amethyst> TILEP_PART_DRCHEAD goes on top of TILEP_PART_HELM
10:02:05 <|amethyst> for local tiles you could try changing the order in pack_doll_buf in tiledoll.cc
10:02:12 <|amethyst> and see how it looks
10:03:21 <|amethyst> in webtiles there's presumably similar logic somewhere in the game_data javascript
10:03:25 <espais> ok i'll play with that a bit
10:03:48 <espais> silly question, but is there a debug way to spawn a hat nearby when i start off a game for testing?
10:03:59 <|amethyst> &%hat
10:04:23 <|amethyst> (&Y%hat   if you're not in wizmode yet)
10:04:34 <|amethyst> spawns it at your feet
10:05:24 <|amethyst> you can also use &rgreen draconian   etc to test the different draconian colours
10:06:35 <espais> ok th
10:06:37 <espais> *thx
11:13:00 <gammafunk> We use a check like !pointer over something like pointer != nullptr, right?
11:13:43 <gammafunk> I recall that we want to avoid explicitely using 0 for a pointer over nullptr, but I think for boolean tests like that it's ok to do the more direct truth test?
11:14:10 <espais> is nullptr defined to be NULL?
11:14:46 <espais> i haven't dug into the codebase yet, i'm just curious as an outsider looking in
11:23:59 <gammafunk> espais: http://en.cppreference.com/w/cpp/language/nullptr
11:24:16 <gammafunk> part of C++11 it seems
11:24:29 <espais> eww
11:24:39 <espais> if (!ANSI-C) {return -1;}
11:24:54 <wheals_> yeah
11:25:04 <wheals_> except it would be just if (pointer) not if (!pointer)
11:25:12 <gammafunk> yeah, wheals_
11:25:27 <gammafunk> espais: don't look at the crawl codebase then, since it's over 400k lines of !ANSI-C
11:26:26 <espais> heh
11:26:26 <espais> i was just kidding
11:29:56 <|amethyst> gammafunk:  I think either is fine, but slightly prefer the more concise version
11:30:09 <|amethyst> it's not an issue the way 0 or NULL is
11:32:20 <|amethyst> (NULL is allowed to be either 0 or nullptr as of C++11: http://en.cppreference.com/w/cpp/types/NULL)
11:34:05 <|amethyst> gammafunk:  (but oddly, in the case of integers, I definitely prefer foo == 0 to !foo)
11:34:16 <gammafunk> yeah, I understand that
11:34:32 <|amethyst> I guess it comes from thinking of pointers as "optional references"
11:34:54 <gammafunk> |amethyst: one thign I wondered about the model we seem to have wrt "pointer vs reference"
11:35:33 <gammafunk> is that it really seems to me, speaking naively as a non-expert, that we shouldn't be using pointers just because "the thing might be 'empty'"
11:35:42 <gammafunk> it seems to me that having pointer syntax just to allow that is bad
11:35:57 <gammafunk> and ideally there would be things like a ref to "not any monster"
11:36:08 <gammafunk> special globals/defines that encapsulate that idea
11:36:25 <gammafunk> like why switch to pointer syntax if all we care about is the fact that the thing might not 'exist'?
11:36:51 <gammafunk> if you're actually making an output parameter, that's one thing
11:37:14 <gammafunk> but using monster *mons just because mons might be nullptr strikes me as a bad design
11:37:35 <gammafunk> but maybe it's not feasible for refs to work that way
11:39:13 <gammafunk> wheals_: see? I can be pretty deep on the programming stuff sometimes
11:40:03 <|amethyst> gammafunk:  IMO the problem with that is that you have to have a special not-a-value instance for each type
11:40:32 <gammafunk> yeah, I guess you would need quite a few of those wouldn't you
11:41:21 <gammafunk> we do have a number of things like that already
11:41:31 <gammafunk> like the coord_def define that's (-1, -1)
11:41:50 <gammafunk> which, sadly, is not used consistantly for that kind of purpose
11:42:15 <gammafunk> I see various uses of coord_def to denote "not a valid place"
11:42:25 <|amethyst> yeah
11:42:30 <gammafunk> I think (0, 0) gets used a lot as a global location, for instance
11:42:43 <|amethyst> possibly a lot of things should be optional<coord_def> once we can use C++17
11:43:14 <|amethyst> (or implement optional<> ourselves)
11:44:35 <gammafunk> I see, and that's not for use with references
11:45:02 <|amethyst> yeah, I think at least in part because it's idiomatic in C++ to use pointers for that purpose
11:45:58 <|amethyst> ah
11:46:21 <|amethyst> the real problem was a lack of agreement on what to do if you assign one optional reference to another
11:47:21 <|amethyst> you can do optional<reference_wrapper<T>>
11:47:26 <|amethyst> but that's ugly of course
11:49:18 <|amethyst> http://tristanbrindle.com/posts/optional-references  discusses the issue somewhat (and has links to the original proposal, which did in fact include optional references)
13:04:47 <Henzell> Unstable branch on crawl.akrasiac.org updated to: 0.21-a0-227-g58a7c62 (34)
13:29:33 -!- amalloy_ is now known as amalloy
15:16:20 <ai_4_life> hey, is anyone here that could help me understand how the webserver returns map-info via json?
15:17:17 <ai_4_life> when a client connects to the webserver, it sends json data with a 'cells' tag
15:17:37 <ai_4_life> some cells have x, y, g, ... etc and other fields
15:18:00 <ai_4_life> but I don't see how to reconstruct the map based on this info
15:18:37 <ai_4_life> any tips? Or could someone point to the code that reconstructs the map in the web browser after getting this data?
15:25:51 <gammafunk> ai_4_life: well, how do you mean "reconstruct the map", exactly?
15:26:07 <gammafunk> I think the json is structured so that it sends packed tile data for each cell
15:26:37 <gammafunk> so it's just a case of the renderer figuring out which tiles to display on a given cell
15:26:49 <gammafunk> if by map you mean "game data", that's now how the client works
15:27:08 <gammafunk> client is basically a renderer and an input coordinator
15:27:14 <gammafunk> it does not process game logic for the most part
15:27:21 <ai_4_life> right
15:27:37 <ai_4_life> i'd like to be able to create a 2d grid after getting the packed tile data
15:27:39 <gammafunk> it renders data and processes input, but the crawl binary handles the logic of interpreting that input and processing the game world
15:28:00 <ai_4_life> right, i'm looking for where the interpretation of the packed tile data happens
15:28:18 <ai_4_life> it didn't seem all cells have an x or y value
15:28:42 <ai_4_life> and i wasn't sure if that's relative to the player's current position, or from like the top left corner of the screen
15:29:30 <gammafunk> ai_4_life: have you been looking at the game_data/static code?
15:29:41 <gammafunk> most specifically cell_renderer.js
15:29:46 <gammafunk> do_render_cell: function(cx, cy, x, y, map_cell, cell)
15:29:50 <gammafunk> in that file, for instance
15:30:37 <gammafunk> and dungeon_renderer.js
15:30:58 <ai_4_life> okay
15:31:04 <gammafunk> it would only have to render a view window
15:31:06 <ai_4_life> i think i poked around there briefly
15:31:20 <gammafunk> so it probably has no notions of global map coordinates
15:31:20 <ai_4_life> I will look into that in more detail
15:31:21 <gammafunk> like the crawl binary does
15:31:22 <ai_4_life> okay
15:32:05 <gammafunk> ai_4_life: see the functions render_loc() and shift() in dungeon_renderer.js for how cells are iterated
15:32:14 <ai_4_life> okay great
15:32:57 <ai_4_life> I think i was trying to avoid going through that code :D
15:33:10 <ai_4_life> sorry to bug you, thanks for the pointers
15:33:17 <amalloy> render_loc is where i wound up too after poking around a bit
15:33:22 <amalloy> probably an important function
15:34:35 <amalloy> i am surprised to see that chrome's network tab doesn't obviously show the ongoing connection between the client and server: just a bunch of static js loaded at the beginning of the game. but obviously there must be some back-and-forth on every player action and i'm just not seeing it
15:35:12 <gammafunk> amalloy: it's over the websocket
15:35:15 <ai_4_life> go to network, ws
15:35:26 <ai_4_life> you can see all the json that you send via key commands, etc
15:35:44 <ai_4_life> but what the server returns is in binary, so its not directly readable on the chrome developer tools panel
15:35:46 <amalloy> ah. i'd looked there, but i'd done it wrong somehow and that tab was blank
15:35:52 <ai_4_life> click the frames
15:35:55 <ai_4_life> tab
15:36:13 <ai_4_life> might have to refresh the page
15:36:19 <amalloy> right
15:37:13 <ai_4_life> amalloy: what are you working on if you don't mind me asking
15:37:38 <amalloy> you mean why am i looking in chrome's network tab?
15:37:53 <amalloy> i'm not working on anything; i was just looking around since you'd asked a question
15:37:59 <ai_4_life> oh gotcha haha
15:38:01 <ai_4_life> thank you
15:40:58 <gammafunk> ai_4_life: ...he works for....them....
15:41:18 <gammafunk> I cannot say more...they are watching me right now....
15:45:03 -!- mol is now known as Guest60268
15:45:09 <amalloy> dang. i have tried to come up with a clever response to gammafunk's insinuations, and everything is either too sinister or super lame
15:45:13 <amalloy> you win again, gammafunk
15:45:51 <amalloy> i'll think of something funny half an hour from now. chei unreasons
16:07:54 <gammafunk> nothing more terrifying than the looming prospect of a chaos dunk
16:13:01 <Cheibriados> Can still use Okawaru abilities under penance 13https://crawl.develz.org/mantis/view.php?id=11203 by UsaSatsui
16:23:00 <amalloy> ^ tiles reasons
16:28:38 <amalloy> i guess the question is, when you are under penance should we remove invocation icons from the menu, or leave them in but have them print an error message?
16:38:06 <geekosaur> imo allow them with a warning... but they put you deeper into penance
16:38:27 <geekosaur> (and don't work as well)
16:40:14 <geekosaur> and piety costs doubled
16:40:33 <gammafunk> and you get an immediate wrath action
16:42:06 <geekosaur> that ought to be implied by deeper into penance...
16:48:20 <gammafunk> why would it? I don't think getting more pennance guarantees a wrath action, does it?
16:49:28 <geekosaur> a god going "I'm all pissed off but will take a whack at you some other time" seems weird to me...
16:51:00 <amalloy> that's pretty much what dcss gods do though
16:51:33 <gammafunk> sif never gets mad at me <3 <3 <3
17:07:08 <amalloy> i don't really understand why tiles is so careful to include items in the ability tray that are supposed to be unusable
17:07:46 <amalloy> so players can remember that heroism is in the second slot from the left maybe, instead of remembering the image?
17:20:07 <gammafunk> I could see how shifting around ability locations based on game state could be a problem
17:20:19 <gammafunk> but I also don't use the mouse for games like this
17:27:32 -!- gressup_ is now known as gressup
17:28:16 <geekosaur> amalloy, maybe you missed the versionsof microsoft word that "optimized" its menus based on usage
17:28:34 <amalloy> i remember everyone being real mad about the ribbon
17:29:22 <geekosaur> I don't mean the ribbon. I mean menus moving often used stuff up to the top
17:29:52 <geekosaur> which meant you always had to stop and read the damn menu to see which thing you want to use, instead of just letting muscle memory save its location
17:30:42 <geekosaur> I should not have to play a menu minigame to use a word processor or spreadsheet...
17:39:26 <amalloy> interestingly there are exactly two active god powers that are usable even while under penance. anyone want to guess what they are?
17:49:04 <geekosaur> not counting renunciation?
17:51:21 <gammafunk> hrm
17:52:03 <gammafunk> ??gods
17:52:03 <Sequell> gods[1/3]: '?' '/' 'G' for a complete list (in game). Type "??<full god name> reasons" (in IRC) to get a list of reasons which might motivate you to worship that particular god.
17:52:05 <gammafunk> ??gods[2
17:52:05 <Sequell> gods[2/3]: Ashenzari, Beogh, Cheibriados, Dithmenos, Elyvilon, Fedhas, Gozag, Hepliaklqana, Jiyva, Kikubaaqudgha, Lugonu, Makhleb, Nemelex Xobeh, Okawaru, Pakellas, Qazlal, Ru, Sif Muna, Trog, Uskayaw, Vehumet, Wu Jian Council, Xom, Yredelemnul, Zin, the Shining One.
17:53:07 <gammafunk> zin's donate gold?
17:53:26 <gammafunk> ??elyvilon
17:53:26 <Sequell> elyvilon[1/4]: The divinity of healing. Most reliable source of protection from being killed (see {divine protection}), but will impose penance (i.e. not answer your requests for help) if you kill living creatures while invoking said protection. Provides healing invocations, but dislikes cannibalism, the use of necromancy, and allowing allies to die.
17:53:41 <gammafunk> maybe one of ely's abilities
17:54:23 <amalloy> not counting renounce, indeed
17:55:03 <amalloy> gammafunk got one of them, donate gold. i thought maybe ely's divine protection would be one too, but it's not
17:55:09 <gammafunk> hrm
17:55:23 <gammafunk> I was vaguely thinking it might be pacify
17:55:29 <gammafunk> but it didn't seem too likely a candidate either
17:55:53 <gammafunk> there's beogh gifting stuff to orcs
17:56:09 <gammafunk> which could concievably not be tied to pennance, but I'd be surprised if one of those were it
17:56:55 <amalloy> ??burn books
17:56:55 <Sequell> burn books[1/3]: Trog invocation that gives piety (2) and creates a cloud of flame. Each book in LOS extends the duration of the clouds. Won't burn manuals.
17:57:17 <amalloy> for whatever reason, this is allowed under penance
17:57:27 <amalloy> maybe because it generates piety instead of costing it, like donating gold?
17:57:40 <amalloy> if so we should add fedhas's fungal bloom thing, now that it's an ability and not a (p) command
17:58:27 <gammafunk> ahh
17:58:31 <gammafunk> I didn't even consider that one
17:58:38 <gammafunk> not sure it would have been a guess even then
17:58:50 <gammafunk> since it does have a tactical effect
17:59:03 <gammafunk> but yeah zin I settled on because it's a way to gain piety
17:59:20 <gammafunk> pacify for the same reason, although I think that one spends more piety than it gains
18:02:01 <Lantell> Unstable branch on underhound.eu updated to: 0.21-a0-227-g58a7c62 (34)
18:03:25 <amalloy> i don't think pacify gives you any piety anymore
18:03:50 <amalloy> it used to be a piety source but now it is a piety sink
18:04:04 <amalloy> see _try_to_pacify and cast_healing
18:06:29 <gammafunk> oh
18:31:54 <|amethyst> !tell johnstein re #594 / #536, I probably won't have time to look into it in more detail for a while, since I've been busy with classes
18:31:55 <Sequell> |amethyst: OK, I'll let johnstein know.
18:32:22 <|amethyst> !tell johnstein I prefer the more complicated solution myself, so that it works more like invisible monsters
18:32:23 <Sequell> |amethyst: OK, I'll let johnstein know.
18:32:57 <|amethyst> !tell johnstein but I do see PF's point about simplicity (and preventing using spells to search for traps, even if it does cost turns)
18:32:58 <Sequell> |amethyst: OK, I'll let johnstein know.
18:36:26 <amalloy> |amethyst: while you're here, care to look over https://github.com/crawl/crawl/pull/605 for me?
18:36:42 <amalloy> oh, i see you're checking in to report that you don't have time :P
18:36:48 <amalloy> well, if not no big deal of course
18:36:53 <|amethyst> well, this one is just one commit, so let's see
18:37:21 <|amethyst> johnstein's is several, and while I did look over many of the commits earlier, I also forgot most of it
18:37:49 <|amethyst> how about you look over #594 and #536 and weigh in? :)
18:38:33 <|amethyst> amalloy:  hm, re 605, are the buttons not even greyed out when you are under penance?
18:38:47 <Cheibriados> New branch created: pull/605 (1 commit) 13https://github.com/crawl/crawl/pull/605
18:38:47 <Cheibriados> 03amalloy02 07https://github.com/crawl/crawl/pull/605 * 0.21-a0-228-g5a351e1: Don't let local-tiles players use god abilities under penance (UsaSatsui) 10(11 minutes ago, 5 files, 52+ 13-) 13https://github.com/crawl/crawl/commit/5a351e1e72be
18:38:50 <amalloy> they weren't, but are now
18:39:10 <amalloy> the code for greying them out is the same code that decides what happens when you click them
18:39:14 <|amethyst> amalloy:  it's kind of weird to have to add a new message when we presumably didn't need that message in console
18:39:40 <amalloy> well. console removes the entry from the (a) menu
18:40:04 <amalloy> but removing from the tiles menu is a bit more disruptive, since it changes the placement of other buttons
18:40:07 <|amethyst> hm
18:40:29 <|amethyst> I think it's perfectly acceptible for clicking on a greyed-out thing to do nothing with no feedback, but I guess having a message in that case isn't terrible
18:40:53 <amalloy> compare to what happens when you're confused
18:41:07 <amalloy> in both tiles and console the ability is selectable but prints a message
18:41:11 <|amethyst> huh
18:41:41 <amalloy> er
18:41:52 <|amethyst> oh, remove those outer () on the return in god_power-usable
18:41:54 <amalloy> no, the (a) button is disabled entirely when you're confused
18:41:58 <|amethyst> s/-us/_us/
18:42:21 <|amethyst> if unbrace doesn't warn about that, it's just because it's not smart enough to parse the multi-line thing
18:42:25 <amalloy> and the buttons are *not* greyed out in the tiles menu when you're confused
18:42:28 <|amethyst> hm
18:42:43 <|amethyst> I'd say make those (also silence) consistent with the penance thing
18:42:54 <|amethyst> in one way or another
18:44:23 <|amethyst> oh, also
18:44:32 <|amethyst> in god_power_from_ability
18:44:42 <|amethyst> I would kind of prefer just to return a const god_power *
18:45:01 <|amethyst> you can use nullptr for false and an actual pointer for true
18:45:02 <amalloy> ah, unbrace does fix that. i left it in because i was copying from that other file which had it inside an if()
18:45:15 <amalloy> |amethyst: agree, will do
18:45:25 <amalloy> i forget why i thought it was a good idea to use a bool and an out-param
18:45:41 <|amethyst> if it hadn't been a pointer that would make sense
18:45:47 <|amethyst> at least until we can use std::optional
18:46:57 <|amethyst> at the very least, if you do use a pointer for an out-parameter, you should assert that the (top level) pointer isn't NULL rather than just assigning through it blindly
18:47:17 <|amethyst> (though I'm sure there are places in the code where we don't do that)
18:48:42 <amalloy> |amethyst: i agree it'd be good to be consistent with silence and confuse, but that's more work; do you mind if i merge this functionality fix and leave the UI consistency fix for some other time?
18:49:12 <Cheibriados> 03amalloy02 07https://github.com/crawl/crawl/pull/605 * 0.21-a0-229-g483def1: Style fixes (|amethyst) 10(2 minutes ago, 3 files, 16+ 20-) 13https://github.com/crawl/crawl/commit/483def11de62
18:50:00 <|amethyst> amalloy:  sure, as long as you either add a FIXME or TODO comment about that, or file a bug :)
18:50:32 <|amethyst> amalloy:  I mean, I wouldn't really *complain* if you pushed without doing that...
18:50:57 <|amethyst> but better to have that recorded somewhere other than IRC logs :)
18:51:08 <amalloy> oh, and re #594/#536 i told johnstein that that's more of a game-design thing and i mostly stick to code-cleanliness because i don't have strong feelings about most game-design stuff
18:51:32 <|amethyst> hm
18:51:35 <amalloy> both behaviours seem reasonable enough to me
18:51:55 <|amethyst> !tell johnstein  I wonder if it would simplify code more to also move the invisible creature check into _check_fixed_summon_loc...
18:51:55 <Sequell> |amethyst: OK, I'll let johnstein know.
18:53:39 <|amethyst> !tell johnstein if you can get it down to net positive millimarvins, not counting comments, it would be hard for anyone to make the simplicity argument :)
18:53:40 <Sequell> |amethyst: OK, I'll let johnstein know.
18:54:47 <|amethyst> amalloy:  I'm even less of a game-design person
18:55:13 <|amethyst> otherwise I would have already either merged food-unification, or removed food :)
18:57:32 <amalloy> oh, i was wrong. it is the same as confuse/silence already
18:57:57 <amalloy> i got mixed up because i was accidentally comparing penance+confuse to just penance, and they looked the same
18:57:58 <|amethyst> before or after your commit?
18:58:01 <amalloy> after
18:58:07 <|amethyst> ah, cool
18:58:48 <|amethyst> btw, we probably still have the bug where if you click on a spell fast enough you either get it for 0 MP, or crash Crawl :)
18:58:52 <|amethyst> in local tiles
18:59:08 <|amethyst> we really shouldn't be doing anything about those clicks until input()
18:59:14 <amalloy> except...augh. when confused, it greys out but prints no message. when silenced or in penance, it greys out and prints a message
18:59:58 <amalloy> i see why it happens but i'm not entirely sure what the right fix is
19:00:02 <|amethyst> amalloy:  heh, it sounds like there is some potential for more simplification, since different behaviour implies that different code is handling it :)
19:00:12 <amalloy> !source get_talent
19:00:12 <Sequell> 1/1. https://github.com/crawl/crawl/blob/master/crawl-ref/source/ability.cc#L1024
19:00:19 <amalloy> !source AbilityRegion::handle_mouse
19:00:19 <Sequell> 1/1. https://github.com/crawl/crawl/blob/master/crawl-ref/source/tilereg-abl.cc#L53
19:00:41 <|amethyst> hm
19:00:43 <amalloy> get_talent returns NON_ABIL if you're too confused to use the thing
19:00:56 <|amethyst> but not in the several other cases that prevent using abilities
19:01:01 <amalloy> right
19:01:13 <amalloy> !source _tile_for_ability
19:01:13 <Sequell> 1/1. https://github.com/crawl/crawl/blob/master/crawl-ref/source/tilereg-abl.cc#L185
19:01:25 <|amethyst> I wonder if we need an abflag::silence_ok and/or abflag::penance_ok ?
19:02:16 <|amethyst> get_talent has always made me kind of unhappy
19:02:39 <|amethyst> but not unhappy enough to actually do something about it, because the work to define a better interface would make me even more unhappy
19:05:01 <amalloy> |amethyst: incidentally, what do you think `// TODO(amalloy): we should do xyz` means? do i plan to fix it myself? was it my idea? am i a good person to ask for more details?
19:05:45 <|amethyst> amalloy:  hm, I would wonder myself if I encountered that
19:06:11 <amalloy> i ask because i assumed it meant one thing but i found out google's internal coding guidelines define it another way
19:06:59 <|amethyst> we don't seem to have a lot of those, and most of them are from enne
19:07:25 <|amethyst> (written as  "TODO enne - pool mcache types[...]"
19:07:26 <|amethyst> )
19:08:23 <|amethyst> which still doesn't answer whether enne intended to fix those himself, or was just saying "enne thinks this is a good idea"
19:09:48 <amalloy> anyway, i thought it meant i plan to fix it myself, but thinking about it more i prefer google's definition: ask amalloy for an explanation if you'd like to complete this TODO
19:10:04 <amalloy> encouraging participation rather than discouraging it
19:10:20 <amalloy> sadly it doesn't actually *work* unless people agree on what it means
19:15:13 <Cheibriados> 03amalloy02 07https://github.com/crawl/crawl/pull/605 * 0.21-a0-230-gfbb9852: Add a TODO (|amethyst) 10(4 minutes ago, 1 file, 3+ 0-) 13https://github.com/crawl/crawl/commit/fbb9852c99f5
19:15:39 <gammafunk> yeah, we should implement |amethyst some day, you're right
19:16:14 <Cheibriados> 03amalloy02 {GitHub} 07* 0.21-a0-228-g14ccbf3: Don't let tiles players use god abilities under penance (UsaSatsui) 10(7 seconds ago, 6 files, 51+ 13-) 13https://github.com/crawl/crawl/commit/14ccbf34992f
19:28:54 <amalloy> i always forget to join ##crawl before pushing a commit with an exciting title like "don't let tiles players cheat" to see how people react
19:32:24 -!- amalloy is now known as amalloy_
19:42:43 -!- Webmant9 is now known as Webmant
19:55:55 <gammafunk> amalloy_: there were no amused reactions this time
19:58:30 <johnstein> how exactly do I compute millimarvins?
20:00:13 <gammafunk> ??Millimarvin
20:00:14 <Sequell> millimarvin[1/1]: 22.751 lines of code removed (as of 2017-04-01)
20:00:28 <gammafunk> divide your removed lines by that value
20:01:13 <gammafunk> unless you want the current value of a millimarvin, in which case you have to look up how many lines MarvinPA has removed, and divide that number by 1000
20:05:45 <johnstein> just trying to see how I can get net positive millimarvins for my PR
20:06:26 <johnstein> in need to divide net removed lines right?
20:06:36 <johnstein> s/in/I
20:07:35 <gammafunk> yeah, millimarvins are a measure of removed lines
20:08:12 <gammafunk> but "net millimarvins" is not a consistant use of the unit
20:08:33 <gammafunk> I don't think millimarvins are defined in terms of net lines removed, just lines removed
20:08:39 <gammafunk> I could be wrong though
20:08:50 <johnstein> ah. I was just using |amethyst's nomenclature above
20:09:12 <Jorgrell> Unstable branch on crawl.jorgrun.rocks updated to: 0.21-a0-228-g14ccbf3 (34)
20:09:14 <gammafunk> maybe it is done with net removed, since that's probably more useful as a concept
20:09:26 <gammafunk> right, it is net
20:09:50 <gammafunk> he has 68900 removed lines and 46231 added lines
20:09:51 <gammafunk> so
20:10:05 <gammafunk> !calc (68900 - 46231.0) / 1000
20:10:06 <Sequell> 22.67
20:10:12 <johnstein> so if I remove 20 lines by refactoring code to a new function, but add 22 lines for that new function. do I get credit for those 20 lines? or have I sinned since my net is +2?
20:10:20 <johnstein> this is super important obviously
20:10:20 <gammafunk> the latter
20:10:29 <gammafunk> it is net lines, I was mistaken
20:10:45 <johnstein> ah I see
20:10:48 <gammafunk> !calc (68900 - 46231.0) / 1000 * 10
20:10:49 <Sequell> 226.69
20:11:06 <johnstein> I better avoid refactoring then
20:11:34 <gammafunk> !learn set millimarvin 22.669 lines of code removed (as of 2017-08-30)
20:11:35 <Sequell> millimarvin[1/1]: 22.669 lines of code removed (as of 2017-08-30)
20:12:02 <gammafunk> good news is that if you do get some net removed lines
20:12:11 <gammafunk> the newest value is more favorable to your millimarvins
20:13:37 <Cheibriados> prompted for cursed ring twice 13https://crawl.develz.org/mantis/view.php?id=11204 by geekosaur
20:14:09 <|amethyst> johnstein:  you can make up for it by removing all comments from the code
20:26:05 <johnstein> is it even possible to make it even doing that? the new function alone is 30+ lines
20:27:50 <|amethyst> johnstein:  *all* comments, not just the ones on those functions :)
20:28:36 <johnstein> ah yes. that I can do
20:32:55 -!- amalloy_ is now known as amalloy
20:34:03 <amalloy> johnstein: net positive millimarvins simply requires net negative lines; it doesn't matter how large a millimarvin is
20:34:28 <amalloy> if you net removal of one line, you'll gain less than one millimarvin, but more than zero
20:34:33 <johnstein> yea. I finally made that connection on my wall back to my car from work
20:34:44 <johnstein> yea. exactly that pout
20:34:47 <johnstein> point
20:34:48 <|amethyst> assuming MarvinPA doesn't suddenly go on a code-adding binge
20:35:06 <|amethyst> in which case we might have to redefine the unit
20:35:15 <amalloy> protomarvins
20:35:41 <johnstein> depending on how much code I the invis ghost handle has and how I can implement it to the function, I might be able to get close if I ignore my comments
20:35:43 <amalloy> using as a reference point the amount of code he'd removed before transcending to his next stage
20:35:53 <geekosaur> marvinalia?
20:43:24 <gammafunk> if johnstein doesn't get millimarvins from this commit I say we talk to the cbro admin about getting his account there banned
20:43:30 <gammafunk> the stakes are just too high
21:04:21 -!- amalloy is now known as amalloy_
21:20:13 <gammafunk> @?draconian annihilator
21:20:13 <Gretell> red draconian annihilator (12q) | Spd: 10 | HD: 16 | HP: 66-97 | AC/EV: 9/10 | Dam: 15 | 10weapons, 10items, 10doors, spellcaster, cold-blooded | Res: 06magic(40), 05fire | XP: 1490 | Sp: b.lightning (3d21), poison arrow (3d23), iron shot (3d29), crystal spear (3d35), blink | Sz: Medium | Int: human.
21:20:25 <gammafunk> !source drac_color_incompatible
21:20:26 <Sequell> Can't find drac_color_incompatible.
21:20:32 <gammafunk> !source drac_colour_incompatible
21:20:32 <Sequell> 1/1. https://github.com/crawl/crawl/blob/master/crawl-ref/source/mon-place.cc#L509
21:20:45 <gammafunk> and yet grey dracs can cast b.lighting?
21:20:59 <gammafunk> ah, those don't spawn
21:21:04 <gammafunk> I was wondering about that too
21:21:21 <gammafunk> nor do pale, as it turns out
21:21:34 <gammafunk> ok, well black draconians can cast iron shot
22:55:39 -!- n1k is now known as n1
23:09:47 <espais> any chance i can get a higher access level for mantis?  i wanted to close out the captcha issue if possible
23:42:05 <espais> !tell napkin can you either grant me a status higher than reporter on mantis to close out the captcha issue or close it yourself?  unless if there are any user complaints i think it should be good to go
23:42:05 <Sequell> espais: OK, I'll let napkin know.
23:42:16 <espais> !tell napkin though i would be interested to see if the number of bot registrations increases
23:42:17 <Sequell> espais: OK, I'll let napkin know.