00:12:05 -!- upsy has joined ##crawl-dev
01:43:58 -!- Textmode has quit [Ping timeout: 240 seconds]
02:09:22 <Gretell> Pseudonut (L6 DrPa)  (D:1 (ZotDef))
02:35:39 <greensnark> !lm * zotdef crash -log
02:35:39 <Sequell> 4. Pseudonut, XL6 DrPa, T:1654 (milestone): Can't find crash dump.
02:35:44 <greensnark> Hm
02:36:56 <sorear> Why do grettel reports have no data
02:37:58 <greensnark> Segfaults produce those weird-looking crash milestones
02:38:15 <greensnark> Whereas with asserts you see the assert condition
02:39:04 <greensnark> !lm * zotdef crash -log
02:39:05 <Sequell> 4. Pseudonut, XL6 DrPa, T:1654 (milestone): http://crawl.develz.org/morgues/trunk/Pseudonut/crash-Pseudonut-20101207-080922.txt
02:39:19 <greensnark> Or a floating-point exception in this case
02:40:25 * sorear can't remember the last time he saw a true SIGFPE
02:40:38 <greensnark> Yeah, this is just a divbyzero
02:40:44 <sorear> every one I've seen has come from the integer unit
02:47:18 <bhaak> per definition a SIGFPE is SIGFPE. there doesn't need to be floating points involved :-)
02:48:44 * bhaak is in stickler mode
02:50:00 <greensnark> !lg * zotdef win
02:50:00 <Sequell> No games for * (win).
02:50:11 <greensnark> !hs * zotdef
02:50:11 <Sequell> 864. 78291 the Champion of Chaos (L19 GhCK), worshipper of Makhleb, rotted away on D:1 (zotvault) on 2010-12-07, with 199178 points after 11283 turns and 2:00:20.
02:50:16 <greensnark> !hs * zotdef -tv
02:50:16 <Sequell> 864. 78291, XL19 GhCK, T:11283 requested for FooTV.
02:51:13 <greensnark> Looks crowded
03:16:19 -!- jld has quit [Ping timeout: 260 seconds]
03:22:56 -!- jld has joined ##crawl-dev
04:01:49 -!- monky has quit [Quit: hello]
04:53:22 -!- Mu_ has joined ##crawl-dev
05:13:39 -!- dpeg has joined ##crawl-dev
05:13:42 <dpeg> cheers
05:20:14 -!- elliptic has quit [Ping timeout: 245 seconds]
05:23:17 -!- Zaba has quit [Ping timeout: 265 seconds]
05:24:27 <dpeg> Haha, Johanna finally made MentalMouse (the guy with the save points campaign) shut up.
05:31:43 -!- Zaba has joined ##crawl-dev
05:56:34 -!- galehar has joined ##crawl-dev
05:57:09 <galehar> hey
05:59:18 -!- galehar has quit [Remote host closed the connection]
05:59:21 <kilobyte> <3 Johanna
05:59:56 -!- raphael has joined ##crawl-dev
06:00:46 -!- raphael has quit [Client Quit]
06:03:27 <dpeg> Hi
06:04:40 <due> hi dpeg
06:04:40 <Henzell> due: You have 2 messages. Use !messages to read them.
06:04:43 <due> !messages
06:04:43 <Henzell> (1/2) dpeg said (19h 5m 3s ago): The new mimics are fun. But shouldn't the old ones also follow you?
06:04:46 <due> !messages
06:04:46 <Henzell> (1/1) dpeg said (19h 4m 4s ago): And I think I remember the bug from yesterday: I had a Crypt stairs mimic (<3), and it showed on Ctrl-O. But Ctrl-G wouldn't move there!
06:04:57 <due> Yes ,there are lots of things to sort out.
06:05:03 <due> I will set aside some time this weekend to work on them!
06:05:52 <dpeg> due: can you dis-guise mimics by throwing stones at them?
06:08:33 <due> if you can, that affects currnet mimics too
06:08:58 -!- galehar has joined ##crawl-dev
06:09:10 <dpeg> yes, and neither of them should be prone to this
06:09:23 <dpeg> before long, players will be throwing stones at items and stairs
06:10:24 <kilobyte> good point, there's so many things a "smart player" should do but are insanely tedious
06:10:41 <casmith789> dpeg: I doubt people will, most people play to enjoy the game
06:10:46 <dpeg> yes, and we should abolish all of these
06:10:56 <dpeg> casmith789: this is not the correct way of approaching this
06:11:03 <due> if people can then they will
06:11:08 <dpeg> yes, at least some of them
06:11:17 <casmith789> I see what you mean though.
06:11:23 <dpeg> and we will protect them from their doubtful traits
06:11:45 <kilobyte> like, in original rogue, do you wield a bow and shoot every semi-risky monster, switch to the normal weapon at distance 2 and proceed to melee?
06:11:53 <casmith789> people use autoexplore and it is far from optimal
06:12:48 <bhaak> but it isn't tedious
06:12:53 <dpeg> kilobyte: this is not nearly as bad as engraving elbereth all the time or throwing those stones
06:13:27 <dpeg> kilobyte: and with Crawl's interface, this behaviour is very smooth
06:13:30 <bhaak> ah, there is the side blow to nethack. I was wondering how long it would take :)
06:13:42 <dpeg> the point is that we should abolish tedium where it's unwanted and streamline it where it's necessary
06:13:52 <bhaak> are mimics dangerous enough that people would do this?
06:13:54 <dpeg> bhaak: I added a blow to our game in the same line! ):
06:13:55 <kilobyte> Crawl has some other problems of this kind: everyone can weaken up every single monster using darts.  I prefer to splat many times before every game instead -- I hate that so much.
06:13:56 <dpeg> erm, :)
06:14:26 <dpeg> kilobyte: but pressing ffffffff is not tedious...
06:14:32 <dpeg> and later on, darts are useless
06:15:14 <kilobyte> they are strictly better than using that turn to advance towards the monster (assuming both of you do melee)
06:15:47 <dpeg> kilobyte: yes. I don't think a hack'n slash game can avoid this.
06:16:18 <bhaak> but, as dpeg said, darts are useless later on, so you have to find the point to change your strategy. It would be bad if you could use one strategy for the whole game
06:16:50 <dpeg> bhaak: later on, you can use javelins or large rocks (as Og/Tr).
06:17:15 <dpeg> But you can also learn a bit of stabbing and approach sleeping monsters differently.
06:17:18 <kilobyte> javelins are pretty rare and heavy... but lugging them can still give you a significant edge
06:17:47 <kilobyte> the problem with early darts (and javelins if you're lucky) is that they don't require any skill investment
06:17:57 <dpeg> kilobyte: if we think that's bad, then we should abolish several forms of RC
06:17:57 <kilobyte> so you pay _solely_ with tedium
06:18:05 <bhaak> and you lose javelins, they are not a fixed resource
06:18:11 <dpeg> we can make them need more skill, no problem
06:18:45 <dpeg> bhaak: you can reduce chances of losing by enchanting them (straight from NH! :), but then you lose those scrolls for your main weapon...
06:18:52 <kilobyte> that would probably need rebalancing the early game to still allow streaks, but it would be a nice change
06:19:16 * bhaak goes source diving
06:19:20 <dpeg> kilobyte: we have this problem with needles anyway... curare procs very often, the others not. All of this should depend a lot more on skill.
06:19:28 <kilobyte> dpeg: enchanting is a _good_ thing: you pay a cost!
06:19:33 <dpeg> yes
06:23:50 <bhaak> dpeg: yes, straight from nh, but I didn't know the non realistic justification: "we were breaking 2/3 of everything unconditionally. we still don't want anything to survive unconditionally, but we need ammo to stay around longer on average." http://nethackwiki.com/wiki/Dothrow.c#line1322
06:24:35 <dpeg> bhaak: I would suspect the formulas are different... we only took the basic idea.
06:25:27 <dpeg> bhaak: I like it how somebody in the devteam thought about balance (like here that ammo should perish eventually), only to have others tear everything down with later "additions".
06:25:29 <casmith789> the downside of javelins is that they are exceedingly heavy, apart from that they are very good.
06:26:35 <bhaak> dpeg: I'm just surprised that they added such a far fetched mechanism: the objects hit, do damage and disappear, instead of a something more realistically rationalization like "the darts might hit armor or a wall and break"
06:28:04 <dpeg> bhaak: then they would have to add broken darts :)
06:28:58 <bhaak> dpeg: no :) of course the dungeons of doom are full of broken stuff, it's the reason you can indefinitely write Elbereth into the "dust"
06:33:09 <dpeg> "We don't keep track of dungeon rubbish.", I know.
06:33:52 -!- Zaba has quit [Ping timeout: 265 seconds]
06:37:07 -!- Zaba has joined ##crawl-dev
06:50:23 -!- galehar has quit [Quit: disconnect]
06:52:03 <dpeg> Eino is good, too.
06:52:27 <dpeg> Just closed the early shop mimics complaint, asking for feedback to go to the wiki.
06:52:33 <dpeg> He's correct, of course.
06:57:51 <Ashenzari> stair mimic not useable with autotravel (https://crawl.develz.org/mantis/view.php?id=2934) by dpeg
07:00:41 <dpeg> treasure trove with 9 potions of agility :)
07:07:05 <due> :D
07:10:53 -!- galehar has joined ##crawl-dev
07:18:44 <greensnark> dpeg wants to travel into stair mimics? Kinky.
07:20:04 -!- Pseudonut has quit [Remote host closed the connection]
07:21:26 <dpeg> greensnark: does the usage history on akrasiac also count Sprint and ZotDef games?
07:23:20 <dpeg> watching ZD games is so much fun ):
07:23:22 <dpeg> erm, :)
07:25:29 <greensnark> Usage history does not check Sprint and ZotDef
07:25:34 <dpeg> ok
07:26:29 -!- syllogism has joined ##crawl-dev
07:27:27 <Zaba> hmm
07:27:44 * greensnark polymorphs Zaba into a turnip.
07:28:10 <dpeg> An undecided turnip, it seems.
07:29:22 <casmith789> but greensnark, it must become something of similar HD... oh never mind, my mistake.
07:29:55 <dpeg> HD can always go up under polymorph!
07:30:22 <dpeg> Poor Zaba, bogged down in the dungeon builder and then all this derision.
07:31:44 <casmith789> defining turnips to have 27 HD will make it all better again
07:44:40 <greensnark> Hellturnips
07:44:50 <greensnark> Zot Defence could have a hellvegetable wave
07:45:15 <greensnark> Hellpineapples and hellturnips escorting a few hell watermelons and a helljalapeno
07:46:15 <dpeg> greensnark: your way of forming new words makes your very well suited for the German language.
07:47:08 <casmith789> who made new zot defence by the way?
07:47:23 <greensnark> Atomjack
07:47:35 <greensnark> dpeg: I am not so easily provoked :P
07:47:41 <greensnark> dpeg: You have to work harder!
07:49:03 <dpeg> Heizölrückstoßabdämpfung
07:49:05 <dpeg> that should do it
07:49:47 <dpeg> three umlauts, one ß, no duplicates, more sex appeal than Marilyn
07:50:15 <greensnark> You're referring to Marilyn Manson?
07:50:23 <dpeg> Who is that?
07:50:40 <due> He's hot.
07:51:03 <dpeg> Is he dead?
07:51:04 <bhaak> it is hot
07:59:15 <greensnark> We should get Atomjack on IRC
07:59:48 <dpeg> yes
08:00:27 <greensnark> !seen Atomjack
08:00:27 <Henzell> Sorry greensnark, I haven't seen Atomjack.
08:02:31 <Vandal> Does Crawl use nobreak mode in curses?
08:02:56 <Vandal> Rather, which file in the source deals with the display? I'm trying to learn how to use curses
08:03:22 <due> Vandal: Hahaha, a) from Crawl? b) why?
08:03:58 <Vandal> Trying to learn how to program roguelikes
08:04:56 <greensnark> libunix.cc has the curses code
08:05:09 <due> curses is :(
08:05:14 <due> it makes grown mencry.
08:05:19 <bhaak> but the display isn't really the important part of roguelikes
08:05:28 <greensnark> curses is a rotten API, but what bhaak said
08:05:45 <greensnark> If you want to program a RL, curses is the least of your problems :)
08:06:15 <Vandal> I'm just starting at the beginning, a display
08:06:41 -!- Zaba has quit [Ping timeout: 264 seconds]
08:07:20 <Vandal> thanks greensnark
08:12:37 -!- Zaba has joined ##crawl-dev
08:13:44 <Gretell> Ragdoll (L8 DSSt)  (D:1 (ZotDef))
08:18:40 <Zaba> how odd
08:21:51 <Zaba> hmm..
08:22:09 * galehar pull out his wand of polymorph other
08:25:08 <Vandal> you guys have a good point, and I kind of wanted to make an RL for my Android anyway
08:25:50 <Vandal> I just wanted an online high score list to be central to the game, and that made me think of a server and SSH, and that meant curses
08:26:00 <Vandal> but I don't really need a game server to display high scores
08:26:09 <Vandal> The game can just upload the score when you end
08:26:47 <bhaak> yes, you only need a server that somehow gets the high score list and display it somehow (preferably in a browser)
08:26:58 <Zaba> greensnark, ping?
08:27:07 <greensnark> Yo
08:27:18 <Zao> Vandal: Not everyone are constantly connected either, nor have flat-rate internet.
08:27:22 <Zaba> greensnark, I've been pondering what to do about DNGN_BUILDER_SPECIAL_{WALL,FLOOR}
08:28:26 <Zaba> greensnark, at the moment, it's kind of very damn ugly.. it seems to be consistently set for special rooms when they're generated.  Other than that, some things repsect it, others don't, and the purpose seems to overlap with that of masking (i.e., MMT_*).
08:28:51 <greensnark> So remove it!
08:28:58 <Zaba> greensnark, but replacing it with masking doesn't seem to be very straightforward
08:29:06 <Zaba> which seems to be the most reasonable option..
08:29:39 <Zaba> greensnark, for example, how do I do the equivalent of _find_in_area(sx, sy, ex, ey, DNGN_BUILDER_SPECIAL_WALL) but for a specific map mask?
08:30:21 <Zaba> or count_feature_in_box(sr.tl, sr.br, DNGN_BUILDER_SPECIAL_WALL)
08:30:25 <greensnark> Write a find_in_area replacement that takes a mask to look for?
08:31:08 <greensnark> The count_feature call is anyway just a find_in_area in fancy dress
08:31:21 <Zaba> hmm..
08:32:24 <Zaba> greensnark, no, it isn't..
08:32:36 <Zaba> it's almost the same code, but it's duplicated
08:32:53 <greensnark> I mean the way it's used for special walls
08:32:59 <Zaba> well, yes..
08:33:28 <Zaba> only it uses the version that takes coord_defs.  Damn, what a mess.
08:33:52 <Zaba> but that's at least just two ways to pass the same information (a pair of ints vs. coord_def)
08:33:57 <greensnark> I think you're overanalysing this, just rip it out :P
08:34:16 <greensnark> Get rid of the int pairs functions while you're at it
08:34:22 <Zaba> it's worse with areas: There, you have two absolute coord_defs, absolute coord_def + relative coord_def, dgn_region, four ints
08:34:38 <Zaba> where in four ints, there are also relative and absolute variants
08:36:16 <Zaba> greensnark, another thing I wanted to ask you about, is all those functions that work on 'the whole level', but sometimes work starting with 0 or 1, and ending with GXM/GYM or GXM-1/GYM-1, with no apparent consistency
08:36:28 <Zaba> greensnark, what's the deal, what's correct?
08:37:03 <greensnark> The grid is 0-(G[XY]M-1)
08:37:12 <greensnark> If anything is going up to GXM, it's out of bounds
08:37:14 <Zaba> for example, _check_doors is int x = 1; x < GXM - 1
08:37:34 <Zaba> but _fixup_hell_stairs works with int i = 0; i < GXM;
08:37:36 <greensnark> A lot of functions will skip the mandatory one-thick-border around the level
08:37:50 <Zaba> greensnark, they should use rectangle_iterator anyway, no?
08:37:56 <greensnark> fixup_hell_stairs can totally skip the edge
08:38:18 <greensnark> Zaba: Yeah, I'd use rectangle_iterator for new code
08:38:47 <Zaba> greensnark, is not skipping the border potentially harmful?
08:39:13 <greensnark> Zaba: There are amusing things involving shop items that happen if you don't skip the map edge
08:39:25 <Zaba> humm.
08:39:36 <Zaba> greensnark, so things that just work on the whole map should use rectangle_iterator ri(1)?
08:39:42 <greensnark> In general the map edge should be touched only for laying down the edge walls
08:39:50 <kilobyte> and for display
08:39:59 <greensnark> Well, dungeon.cc does not do display
08:40:21 <greensnark> Shop code will also mess with the edge wall, but I think it's time we fixed that
08:41:29 <Zaba> greensnark, or is there a more conventional way to initialize a rectangle_iterator that skips the edge? Some constant, perhaps?
08:42:31 <kilobyte> wget http://crawl.develz.org/morgues/trunk/KiloByte/morgue-KiloByte-20101119-134939.txt -O -|grep 'Papyrus of Earthen'
08:42:57 <kilobyte> that's a sample what the shop hack can do when something goes even slightly awry
08:43:44 <greensnark> Only moderately tricky part of getting rid of the shop hack is a shim for save compatibility
08:43:51 <kilobyte> every use of ^F in that game somehow found an item from a shop in Shoals, even when called from Zig or some place
08:43:51 <greensnark> If it were up to me I'd just bump the major version :P
08:44:03 <kilobyte> greensnark: not just that
08:44:32 <kilobyte> what would you do to all the 647863489739 consistency checks that assume that no item can be outside of map bounds?
08:45:32 <greensnark> So many, are there? :P
08:46:48 <greensnark> I'd make the change and catch any hard-to-find consistency checks when they trigger :P
08:47:02 <galehar> are you saying that the shop items are stored on the map edge?
08:47:12 <greensnark> Zaba: rectangle_iterator(1) or rectangle_iterator(BOUNDARY_BORDER)
08:47:18 <greensnark> galehar: Yes, welcome to Crawl development
08:47:23 <galehar> awesome :)
08:47:36 <Zaba> greensnark, hmm.. fine..
08:47:59 <Zaba> I'll try converting DNGN_BUILDER_SPECIAL_* into a map mask or something
08:49:35 <kilobyte> greensnark: could you do that on a branch or something?  I keep getting distracted from finally doing that curses redirector, I'd need a kick or something to concentrate :p
08:49:59 <greensnark> kilobyte: Not Crawl hacking any time soon
08:50:00 <kilobyte> greensnark: or, to distract you: nearly every time I run -test, I get a failure on a test you added
08:50:10 <kilobyte> beh :(
08:50:21 <greensnark> kilobyte: Yes, I know. It's for a monster movement bug that I meant to fix circa 0.7 and never got to :P
08:50:44 <greensnark> Essentially a monster in that position will oscillate back and forth instead of chopping down the plant
08:50:47 <kilobyte> ah, so it _never_ worked correctly in the first place :p
08:50:51 <greensnark> Aye
08:56:34 <Zaba> _make_room, _make_box, _box_room
08:56:39 <Zaba> *head explosion*
08:57:45 <greensnark> What, no make_box_room?
08:57:58 <greensnark> Or make_room_box
08:58:13 <Zaba> afraid so
08:58:42 <greensnark> CL and Scheme have variants of car and cdr like cadr, cddr, cdadr, etc. We could have make_box_room, make_room_box, make_room_room_box, make_box_room_box_room, etc.
08:58:48 <greensnark> Think of the possibilities
08:59:14 <Zaba> I'd rather not :P
08:59:22 <greensnark> If the player visits enough box_room_box_rooms we could generate a free glaive-glaive-guisarme-glaive
08:59:45 <galehar> :D
08:59:49 <Zaba> greensnark, what is the easiest way to iterate from (room_x1; room_y1) to (room_x2; room_y2), inclusively, and set a map mask flag for each coordinate?
09:00:17 <greensnark> rectangle iterator on the corners :P
09:08:16 <greensnark> Heh, I forgot that we increased the shop limit for Sprint. We can crash Crawl just by generating lots of shops
09:16:30 <kilobyte> heh, 100 not 64
09:34:03 <galehar> so, anyone using an IDE? Or just gvim, ctags and ddd?
09:34:42 <Zaba> I use nvi
09:34:50 <Zaba> what's ddd?
09:34:56 <greensnark> Zaba also uses GNU screen
09:35:00 <Zaba> no, I don't
09:35:07 <greensnark> Although he keeps denying it for some reason
09:35:55 <galehar> ddd is a graphical debugger
09:36:07 <Zaba> ah, I just use gdb when I need a debugger
09:37:07 <galehar> a graphical debugger can be great sometimes
09:37:09 <felirx> from what i've heard, eclipse seems to be mature enough to use these days too
09:37:14 <galehar> it can display complex structures
09:37:38 <felirx> then again, it's Java
09:37:51 <Keskitalo> Good evening!
09:37:54 <Zaba> moin Keskitalo
09:38:31 <galehar> felirx: I think there's a c++ plugin
09:38:35 <Keskitalo> Hey Zaba! Moving the knife guarantee from code to a vault was a nice move.
09:38:42 <Ashenzari> Incorrect message when attacking an enslaved unique (https://crawl.develz.org/mantis/view.php?id=2935) by MarvintheParanoidAndroid
09:39:16 <Zaba> Keskitalo, only I've managed to screw it up.  Kudos to kilobyte for fixing it.
09:39:45 <Keskitalo> That was the "uncursed"->"not_cursed" thing? Looked like it was documented incorrectly.
09:40:06 <Zaba> well, he made the vault use that flag, too, while I've been distracted :P
09:40:16 <Keskitalo> Oh, I see. :)
09:40:36 <galehar> Zaba: you shouldn't be so afraid to break things. It's always a good move to replace ugly code by clean one, even if buggy. We can always debug later.
09:40:45 <Zaba> I should really test things properly.. but it's hard to do so when so much randomness is involved
09:41:04 <Zaba> galehar, if I was afraid to break things... would I venture into dungeon.cc at all?
09:41:16 <galehar> good point :)
09:41:25 <Zaba> galehar, I'm just giving credit for fixes where it's due
09:44:43 <Zaba> hmm..
09:51:43 <kilobyte> I guess he's in bed
09:52:27 <kilobyte> why would you bother with eclipse?
09:52:46 <kilobyte> all it can do is moving you to a given function with a click
09:52:56 <kilobyte> when that works, which is basically never
09:53:54 <galehar> so, what do you use kilobyte? gvim and ctags, like due?
09:54:21 <kilobyte> vi is the work of... well, let's not insult the Devil
09:54:24 <kilobyte> so is emacs
09:55:02 <galehar> ah, ok
09:55:08 <Zaba> emacs is actually an elisp VM with a data buffer-oriented interface.  In other words, a religion.  You can barely use it merely as a text editor :P
09:55:10 <greensnark> I'm not sure whether this discussion is more pointless than the coding convention discussion involving return statements and parens, but it's pretty close :P
09:55:14 <kilobyte> I use just a regular editor... and no ctags, either
09:55:31 <kilobyte> you can do the same with just a search, it takes about the same time
09:55:41 <greensnark> Zaba: We'll discuss that right after you explain why you use GNU screen and are ashamed to mention it :P
09:55:54 <kilobyte> what's wrong with screen?
09:56:08 <greensnark> The question is why Zaba is ashamed to admit he uses it :P
09:56:16 <Zaba> I don't use it
09:56:17 <greensnark> And insists that he uses only dtach
09:56:23 <greensnark> See, there we go again :P
09:56:26 <kilobyte> the only downsides of it I know are: 1. failing to handle Unicode above 16 bits, and 2. blocking scrollback
09:56:28 <Zaba> and yes, I use dtach
09:58:10 <greensnark> kilobyte: How do you jump to function definitions as fast as ctags+editor support?
09:58:34 <greensnark> Also I use emacs and am not ashamed to admit it :P
09:58:39 <Zaba> greensnark, so.. I kind of have converted DNGN_BUILDER_SPECIAL_{WALL,FLOOR} to a map mask
09:58:46 <greensnark> Zaba: w00t
09:58:57 <Zaba> I've no idea whether it actually works, though :P
09:59:03 <kilobyte> the same file?  search.  Another file?  A command line alias.
09:59:33 <Zaba> and I'm not sure how to properly test
09:59:45 <greensnark> kilobyte: Question is, how is it as fast as using editor support? Do you have a key trigger to search and jump to that location automatically?
10:00:02 <greensnark> And do you have a key trigger to jump back to your old location?
10:00:49 <Zaba> hmm..
10:01:03 <Zaba> I guess I'll have to turn on special room debugging and do some sanity checking, at least
10:01:20 <Zaba> because, the cases where checks for DNGN_BUILDER_SPECIAL_* were done are best defined as 'corner'
10:01:26 <Zaba> or rather, the cases where they *matter*
10:03:00 <Zaba> greensnark, the problem is, it just translates the existing mechanism to use MMT_SPECIAL map mask
10:03:05 <Zaba> greensnark, it doesn't make any code any more clear :P
10:03:43 <greensnark> Zaba: Will it avoid special rooms clobbering vaults and vice versa?
10:03:51 <Zaba> hmm.
10:03:57 <Zaba> greensnark, does the current mechanism prevent that?
10:04:05 <greensnark> Zaba: No
10:04:15 <Zaba> greensnark, it could be easily done.. if we just mark special rooms with MMT_VAULT!
10:04:36 <Zaba> oh hey, I wonder whether that would actually work.
10:04:36 <greensnark> The map builder may consider the special-walls to be non-overwriteable, dunno
10:04:54 <Zaba> greensnark, only _builder_basic and its subroutines check for DNGN_BUILDER_SPECIAL_*, AFAIK.
10:04:59 <Zaba> the vault-related code does not care
10:05:51 <Zaba> greensnark, but, I have code that marks special rooms with MMT_SPECIAL.  Could change that to MMT_VAULT easily.
10:06:03 -!- eith has joined ##crawl-dev
10:06:14 <Zaba> it would have effect on vaults that are placed after the special room, for sure
10:06:15 <greensnark> Wasn't there some special bazaar code that used the special_floor thing
10:07:01 <Zaba> sounds like something that should be killed with fire
10:07:06 <greensnark> Yes
10:08:09 <Zaba> I don't see anything, though
10:09:11 <Zaba> uh..
10:09:23 <greensnark> Ah, I think it was replaced with floor_halo, so that's fine
10:09:31 <Zaba> greensnark, can special rooms be generated with hidden doors?
10:09:39 <Zaba> or with doors at all :P
10:09:40 <greensnark> No idea
10:09:56 <Zaba> I've just had one with such and I wonder whether it's my breakage :P
10:12:17 <greensnark> Zaba: Looks like the code would put a door in front of the room and there's a fixup routine somewhere that randomly converts some doors to secret doors
10:12:43 <Zaba> yes..
10:12:50 <Zaba> but, often, rooms have no doors.
10:13:19 <Zaba> ah
10:13:23 <greensnark> You're the expert, analyse it!
10:13:36 * greensnark has a meeting in 2 min
10:13:42 <Zaba> greensnark, _hide_doors only hides doors surrounded by DNGN_ROCK_WALL, it seems
10:13:44 <Zaba> ...or something.
10:15:13 <Zaba> and.. I guess normal special rooms have DNGN_BUILDER_SPECIAL_WALL instead
10:15:16 <Zaba> or..
10:15:43 <Zaba> yes, that's right
10:15:52 <Zaba> and so they don't get hidden.. makes sense.
11:24:15 <Ashenzari> Somewhat reducing spam of "foo comes into view" messages. (https://crawl.develz.org/mantis/view.php?id=2936) by Arxale
11:37:09 <Zaba> oh god
11:39:43 <galehar> Arxale is making progress. His latest patch seems quite good :)
11:40:08 <Zaba> I don't think I can do this getting-rid-of-DNGN_BUILDER_SPECIAL_* thing right now
11:40:17 <Zaba> it has too many unexpected consequences
12:20:35 -!- valrus has joined ##crawl-dev
12:20:43 -!- valrus has quit [Remote host closed the connection]
12:20:54 -!- valrus has joined ##crawl-dev
12:26:07 -!- jld has quit [Ping timeout: 272 seconds]
12:28:21 -!- elliptic has joined ##crawl-dev
12:33:01 <Gretell> vampiricplant (L6 DSWr)  (D:1 (ZotDef))
12:33:10 <Gretell> vampiricplant (L4 DSWr)  (D:1 (ZotDef))
12:33:39 <Gretell> vampiricplant (L4 DSWr)  (D:1 (ZotDef))
12:34:26 -!- eith has quit [Ping timeout: 265 seconds]
12:40:23 <CIA-1> 03zaba * rddcef1a79bc4 10/crawl-ref/source/dungeon.cc: Use rectangle_iterator in _fixup_{hell,pandemonium}_stairs.
12:40:24 <CIA-1> 03zaba * r6642fecf5017 10/crawl-ref/source/dungeon.cc: Use rectangle_iterator in a few more places in dungeon.cc.
12:40:26 <CIA-1> 03zaba * rece483f5c9f6 10/crawl-ref/source/ (dungeon.cc dungeon.h): Remove dungeon.cc:smear_feature.
12:40:28 <CIA-1> 03zaba * raa9d1eb65be5 10/crawl-ref/source/dungeon.cc: Split the stair placement code from spotty_level into _place_spotty_stairs.
12:40:29 <CIA-1> 03zaba * rd283860156d0 10/crawl-ref/source/ (dungeon.cc dungeon.h): Remove the 'seeded' parameter from spotty_level.
12:45:51 <Zaba> !tell greensnark Is there any reason why travel_distance_grid_t is not a FixedArray? Being accessible using a coord_def would be useful.
12:45:51 <Henzell> Zaba: OK, I'll let greensnark know.
13:28:23 -!- Textmode has joined ##crawl-dev
13:53:29 -!- monky has joined ##crawl-dev
13:55:09 <Ashenzari> got abyssed, game didn't notice change in terrain (https://crawl.develz.org/mantis/view.php?id=2937) by Maha
13:59:12 <Gretell> Ragdoll (L7 CeHu)  (D:1 (ZotDef))
14:09:38 -!- TGWi has joined ##crawl-dev
14:10:14 <Ashenzari> Improve Static Discharge (https://crawl.develz.org/mantis/view.php?id=2938) by MarvintheParanoidAndroid
14:22:11 -!- Zaba has quit [Ping timeout: 255 seconds]
14:26:00 -!- jld has joined ##crawl-dev
14:28:34 -!- Zaba has joined ##crawl-dev
14:33:31 <CIA-1> 03kilobyte * rf56a503ed324 10/crawl-ref/source/defines.h: Limit MAX_SHOPS, spawning 65th one would cause a crash.
14:34:11 <CIA-1> 03kilobyte * rc4e880069b88 10/crawl-ref/source/ouch.cc: Felids: make corpses and zombies named after you.
14:37:43 <sorear> zombies, eh?
14:38:30 <galehar> you can animate your own corpse? :)
14:48:00 -!- syllogism has quit []
14:58:18 <kilobyte> that's the whole point in removing corpses
14:58:31 <kilobyte> jpeg already protested against the possibility of eating them
14:58:38 <kilobyte> s/removing/leaving/!
14:58:47 <kilobyte> bad braino
14:58:47 <TGWi> eating cats is pretty gross too thoguh
14:58:49 <TGWi> though
14:59:02 <bhaak> but dpeg likes eating cats
14:59:13 <kilobyte> TSO, Zin and Ely will of course get mad at you
15:01:23 <sorear> jpeg protested, eh?
15:04:04 -!- blackpenguin has quit [Ping timeout: 240 seconds]
15:05:14 <kilobyte> or rather, seems unhappy about it
15:06:09 <kilobyte> nothing of the kind "remove it or I'll kill you, dance on your grave and put a gravestone in Comic Sans"
15:10:34 <Ashenzari> Aizul and Xtuaha corpse tiles are different color than the living creatures (https://crawl.develz.org/mantis/view.php?id=2939) by MrPerson
15:15:35 <Ashenzari> Bizarre behavior of vaults wtih lava in Shoals (https://crawl.develz.org/mantis/view.php?id=2940) by minmay
15:32:25 -!- upsy has quit [Quit: Leaving]
15:59:51 -!- galehar has quit [Quit: Page closed]
16:30:18 -!- paxed has quit [Ping timeout: 265 seconds]
16:31:23 <kilobyte> You die...
16:31:23 <kilobyte> _You feel sick.
16:31:23 <kilobyte> You rejoin the land of the living...
16:33:06 -!- jooosa has joined ##crawl-dev
16:35:39 -!- joosa has quit [Ping timeout: 272 seconds]
16:35:39 -!- jooosa is now known as joosa
17:01:08 <Ashenzari> kraken crashing game (https://crawl.develz.org/mantis/view.php?id=2941) by Maha
17:27:02 <Gretell> Unstable branch on CDO updated to: 0.8.0-a0-3636-gc4e8800 (31.16)
17:36:35 -!- casmith789 has quit [Remote host closed the connection]
17:56:01 <dpeg> Would you believe that there's no ego robe after fetching golden, barnacled and silver runes?
17:56:01 <Henzell> dpeg: You have 1 message. Use !messages to read it.
17:56:05 <dpeg> !messages
17:56:05 <Henzell> (1/1) greensnark said (6h 54m 30s ago): cbus asked how your taekwondo is going and said additionally: "? shows that 0 stat = instadeath in CDO trunk"
17:56:41 <dpeg> hm, I am not gonna try that on my Sp right now :)
17:57:00 <Gretell> larasium (L6 HOFi)  (D:1 (ZotDef))
17:58:22 -!- Vandal| has joined ##crawl-dev
18:23:19 -!- MarvinPA|2 has joined ##crawl-dev
18:26:47 -!- MarvinPA has quit [Ping timeout: 265 seconds]
19:16:51 <Ashenzari> Ashenzari not recognizing armor as "bound" (https://crawl.develz.org/mantis/view.php?id=2942) by pixnaps
19:19:14 -!- valrus_ has joined ##crawl-dev
19:23:24 <CIA-1> 03dolorous * r9c7caac744ac 10/crawl-ref/source/traps.cc: Add missing period.
19:33:59 -!- Zaba has quit [Ping timeout: 260 seconds]
19:41:13 -!- Mu_ has quit [Quit: Defecator, may everything turn out okay so that you can leave this place.]
19:44:27 -!- Zaba has joined ##crawl-dev
19:51:48 -!- dpeg has quit [Quit: Lost terminal]
20:13:24 <CIA-1> 03kilobyte * rdaaf8a7a4a49 10/crawl-ref/source/ (ouch.cc spl-summoning.cc): Felids: preserve EV/AC for skeletons/zombies.
20:13:26 <CIA-1> 03kilobyte * ra929315a7cb6 10/crawl-ref/source/mislead.cc: Don't mislead into dummy monsters (and a few random but also quite inappropriate ones).
20:13:27 <CIA-1> 03kilobyte * r8619226f1b9a 10/crawl-ref/source/abl-show.cc: ZotDef: scale up the cost of plants and fungus when there's hundreds of them.
20:13:31 <CIA-1> 03kilobyte * r101e117d9b68 10/crawl-ref/source/ (abl-show.cc mon-stuff.cc mon-stuff.h zotdef.cc): ZotDef: introduce a hard cap of 350 allies.
20:18:09 <Gretell> Unstable branch on CDO updated to: 0.8.0-a0-3641-g101e117 (31.16)
21:15:13 -!- MarvinPA|2 has quit []
21:42:18 -!- Pseudonut has joined ##crawl-dev
21:48:05 <monky> holy word isn't marked as dangerous for undead (and presumably demonspawn)
21:59:34 <greensnark> kilobyte: Maybe ZotDef rotting deaths could pin the blame on the monster on the orb instead of generic "rotted away"?
21:59:34 <Henzell> greensnark: You have 1 message. Use !messages to read it.
21:59:38 <greensnark> !messages
21:59:38 <Henzell> (1/1) Zaba said (9h 13m 47s ago): Is there any reason why travel_distance_grid_t is not a FixedArray? Being accessible using a coord_def would be useful.
22:00:25 <greensnark> Zaba: I don't recall, why travel grid isn't a FixedVector; maybe it was just to skip the array bounds checks
22:00:36 <greensnark> Damn superfluous commas
22:02:43 <Ashenzari> weirdness on interacting with shop items (https://crawl.develz.org/mantis/view.php?id=2943) by Maha
22:33:57 -!- TGWi has quit [Read error: Connection reset by peer]
22:46:43 -!- paxed has joined ##crawl-dev
22:58:00 <Ashenzari> You are throwing a -4,-2 orcish hand axe. (https://crawl.develz.org/mantis/view.php?id=2944) by June
23:01:48 -!- paxed has quit [Ping timeout: 265 seconds]
23:03:22 -!- paxed has joined ##crawl-dev
23:10:45 -!- elliptic has quit [Quit: Leaving]
23:45:11 -!- valrus_ has quit [Remote host closed the connection]
23:55:51 -!- Pseudonut has quit [Remote host closed the connection]