00:06:25 -!- Twinge has quit [Read error: Connection reset by peer]
00:06:35 -!- Twinge has joined ##crawl-dev
00:07:17 <Ashenzari> Jiyva constantly removing mutations, other comments (https://crawl.develz.org/mantis/view.php?id=1798) by mwoody
01:05:01 -!- ortoslon has joined ##crawl-dev
01:22:42 <Ashenzari> Altars shown in passive mapping display incorrectly (https://crawl.develz.org/mantis/view.php?id=1799) by Seriyu
01:29:21 -!- Niccus has joined ##crawl-dev
02:28:43 <due> greensnark: I remembered: did you manage to fix the out-of-bounds thing?
02:59:20 -!- ortoslon has quit [Quit: bye]
03:21:38 -!- syllogism- has joined ##crawl-dev
03:31:06 <Gretell> Soadreqm (L9 SETm) ASSERT(!monster_at(p) || monster_at(p)->submerged() || fedhas_passthrough(monster_at(p))) in 'player.cc' at line 347 failed. (D:7)
03:31:14 <Gretell> Soadreqm (L9 SETm) ASSERT(!monster_at(p) || monster_at(p)->submerged() || fedhas_passthrough(monster_at(p))) in 'player.cc' at line 347 failed. (D:7)
03:31:21 <Gretell> Soadreqm (L9 SETm) ASSERT(!monster_at(p) || monster_at(p)->submerged() || fedhas_passthrough(monster_at(p))) in 'player.cc' at line 347 failed. (D:7)
03:38:12 <Napkin> moin, moin
03:39:12 <Napkin> has anyone upgraded from lenny backport git 1.6.5 to git 1.7.1 yet? ran into any trouble?
03:56:44 <KiloByte> I'm running unstable's git (currently 1.7.1) from day one.
03:58:10 <Napkin> i felt more comfortable with backports packages when i installed git
03:58:31 <Napkin> anyways, I upgraded.. please let me know if there is any error
04:02:28 <Gretell> Soadreqm (L10 SETm) ASSERT(!monster_at(p) || monster_at(p)->submerged() || fedhas_passthrough(monster_at(p))) in 'player.cc' at line 347 failed. (Orc:2)
04:03:31 -!- casmith789 has joined ##crawl-dev
04:06:00 <Gretell> Soadreqm (L10 SETm) ASSERT(!monster_at(p) || monster_at(p)->submerged() || fedhas_passthrough(monster_at(p))) in 'player.cc' at line 347 failed. (Orc:2)
04:14:29 -!- ogaz has quit [Remote host closed the connection]
04:24:47 -!- syllogism- has quit [Read error: Connection reset by peer]
04:24:51 -!- syllogism- has joined ##crawl-dev
04:26:47 -!- syllogism- has quit [Client Quit]
04:29:19 <Gretell> Soadreqm (L10 SETm) ASSERT(!monster_at(p) || monster_at(p)->submerged() || fedhas_passthrough(monster_at(p))) in 'player.cc' at line 347 failed. (Orc:2)
04:32:15 -!- syllogism- has joined ##crawl-dev
04:33:38 -!- Amonchakad has joined ##crawl-dev
04:40:54 <Gretell> Soadreqm (L10 SETm) ASSERT(!monster_at(p) || monster_at(p)->submerged() || fedhas_passthrough(monster_at(p))) in 'player.cc' at line 347 failed. (Orc:1)
04:49:12 <Gretell> Soadreqm (L10 SETm) ASSERT(!monster_at(p) || monster_at(p)->submerged() || fedhas_passthrough(monster_at(p))) in 'player.cc' at line 347 failed. (Orc:2)
04:51:55 -!- xardas-3 has joined ##crawl-dev
04:59:50 -!- Mu_ has joined ##crawl-dev
05:04:26 -!- Textmode has quit [Quit: Over the ages, man has been certain of a great many things. some of these things were even true.]
05:06:14 -!- ortoslon has joined ##crawl-dev
05:11:54 -!- xardas-3 has quit [Quit: ChatZilla 0.9.86 [Firefox 3.5.9/20100401213457]]
05:21:08 -!- Niccus has quit [Quit: Leaving]
05:28:54 <greensnark> due: Can you pastebin the map that is causing the problem again? I can't seem to find the right one
05:29:25 <greensnark> !lm Soadreqm x=god
05:29:27 <Henzell> 648. [2010-06-20] [god=Fedhas] Soadreqm the Nimble (L10 SETm) killed Edmund on turn 15880. (D:9)
05:29:43 <greensnark> !lm Soadreqm type=crash -game
05:29:44 <Henzell> Soadreqm, XL10 SETm, T:13745 (milestone) has no matching game.
05:36:28 <due> Sure
05:37:08 <due> greensnark: http://sprunge.us/YYPd
05:44:46 <CIA-41> 03greensnark * r26565b35e548 10/crawl-ref/source/dungeon.cc: Fix join_the_dots crashing on vault exits close to the edge of the map. (due)
05:44:47 <greensnark> due: Fixed!
05:44:55 * greensnark will be back in 2 hours.
05:46:28 -!- ortoslon has quit [Quit: bye]
05:51:19 <due> Yay <3
06:11:45 -!- purge has quit [Ping timeout: 240 seconds]
06:17:29 <due> Okay,need threat now.
06:24:49 -!- ortoslon has joined ##crawl-dev
06:49:09 -!- ORYLY has joined ##crawl-dev
06:58:50 -!- ORYLY has quit [Quit: ORYLY]
06:59:25 <due> @?hellwing
06:59:25 <Gretell> hellwing (154) | Speed: 10 | HD: 7 | Health: 28-63 | AC/EV: 8/10 | Damage: 17, 10 | Flags: 05demonic, evil, fly | Res: 06magic(56), 03poison | XP: 388 | Sp: teleport other, animate dead, teleport self.
07:06:51 -!- Enne has joined ##crawl-dev
07:48:40 <due> @?wizard
07:48:40 <Gretell> wizard (06@) | Speed: 10 | HD: 10 | Health: 20-60 | AC/EV: 0/13 | Damage: 6 | Flags: spellcaster | Res: 06magic(53) | Chunks: 07contaminated | XP: 710 | Sp: b.venom, mystic blast, invisibility, confuse, slow, teleport self / magic dart, paralyse, haste, b.lightning, confuse, teleport self / paralyse, crystal spear, blink, b.fire, b.cold, minor healing / paralyse, throw flame, invisibility, teleport other, fireb..
07:49:13 <greensnark> So this is literally going to be the Lair vault from Hell? :P
07:53:23 <due> Yes.
07:53:26 <due> Just going to commit it.
07:53:46 <due> Would appreciate comments? :)
07:54:41 <due> Pushed.
07:55:11 <CIA-41> 03due * rbb38f8e67d26 10/crawl-ref/source/dat/des/branches/lair.des: New lair ending: forest from hell. Literally.
07:57:20 <greensnark> Hehe
07:57:52 <due> Like?
07:57:59 <Madtrixr> Lovely.
07:58:01 <due> I really, *really* like the effect in the hell room.
07:58:15 <Madtrixr> oh due, you wily aussie.
07:58:25 * Madtrixr shakes fist in a joking way
07:58:55 <due> Don't shake it around unless you're offering to use it.
07:59:11 <due> No hellephant, though.
07:59:30 <greensnark> due: I expected something to zap me when stepping between the bloody statues :)
08:00:42 <due> They'rejust "creepy scenery".
08:00:49 <due> Hey, has dpeg's email changed?
08:01:34 <greensnark> Not that I know
08:01:46 <due> Excelent.
08:01:54 <greensnark> due: How do I get from the double-door entry chamber to the inner chambers?
08:02:03 * greensnark is too lazy to look at .des file
08:02:28 <due> You don't :)
08:02:37 <due> Unless you have shatter.
08:02:42 <due> Go back to the corner of stonewalls,there's a hidden door.
08:04:19 <greensnark> Ah, I see it :)
08:04:22 * greensnark peeked at .des :P
08:04:26 <due> Cheat.
08:04:40 <due> I was nice and put the demons oon door restrict.
08:04:45 <due> They won't bother you wile you fight the wizard.
08:04:55 <greensnark> I wiztamed the dire elephant and it's off rampaging through the level :D
08:05:06 <due> You suck :)
08:05:32 <due> I wonder what an 8 would give you in Lair:8.
08:05:43 <greensnark> Let's see whether the wizard can fight off this group of demons :)
08:05:57 <greensnark> ...
08:05:58 <due> It's mostly hellwings, but there's a few nasties.
08:06:00 <greensnark> He's using magic dart
08:06:04 <due> "lol"
08:06:29 <greensnark> I'll just recall my dire elephant!
08:06:31 <due> I must admit, I make some good maps. ;)
08:07:06 <due> I'm happier with this one than with the jungle_book, not that I dislike the jungle_book...
08:07:06 <greensnark> Your dire elephant trunk-slaps the hellwing for 9!
08:07:16 <due> Poor hellwing.
08:07:25 <greensnark> You should colour the hell-chamber door fiery or something
08:07:27 <due> Just wait 'til the smokedemon starts smiting.
08:07:33 <due> Why? It's secret.
08:07:39 <greensnark> Oh
08:07:42 <due> How about the entire wall?
08:07:46 <greensnark> Maybe I shouldn't have set all skills to 27 :)
08:07:51 <due> :)
08:07:53 <greensnark> I didn't even notice it was secret :D
08:08:00 <due> Do you like the effect?
08:08:10 <greensnark> The chamber is cool
08:08:11 <due> (Endless lava, portal, demons, red floor, etc.)
08:08:17 <due> :D
08:08:23 * due chuffed.
08:08:49 <greensnark> Hey, my dire elephant is at half health :P
08:08:55 <greensnark> And it's only killed half the demons :P
08:09:07 <due> :o
08:09:31 <greensnark> Aw, the smoke demon killed it :'(
08:09:38 <due> :(
08:09:44 <due> WHORE, SMOKEDEMON WHORe.
08:10:48 <greensnark> due: The statue-filled chamber could use some foreboding effects
08:10:59 <greensnark> It looks very clean and inviting
08:11:03 <greensnark> Like a laundromat :)
08:11:16 <greensnark> The hellish chamber is great
08:11:21 <due> Pools of blood and disembodied screams?
08:11:28 <greensnark> Yes!
08:11:42 <greensnark> Maybe some zombies
08:11:56 <ortoslon> Rupert just palaryzed me and then berserked me, i didn't have a chance :( c'est la crawl
08:12:10 <greensnark> ortoslon: <3
08:12:14 <due> <3 rupert
08:12:17 <due> zombies, hmm.
08:12:21 <due> Maybe the statues could turn into zombies.
08:12:40 <greensnark> We need to be able to place corpses in .des files
08:12:42 <due> Would having a trigger for a monster noticing the player be too much?
08:12:46 <greensnark> And a statue that raises them when the player enters
08:13:01 <due> Corpses would be good :)
08:13:05 <due> TGWi would have your children.
08:13:40 * greensnark adds to todo.
08:13:57 <due> My todo... greensnark, nixor...
08:14:00 <due> No, wait, wrongtodo list.
08:16:47 <greensnark> :P
08:35:26 -!- MarvinPA has joined ##crawl-dev
08:44:58 <CIA-41> 03greensnark * r38d84fd88be6 10/crawl-ref/source/ (main.cc mon-stuff.cc): Fix ASSERT when a Fedhas worshipper on a plant or fungus steps onto a wandering mushroom.
08:57:32 -!- MarvinPA has quit [Quit: Page closed]
08:57:58 <Gretell> Voker57 (L14 DSGl) ASSERT(smc) in 'mon-util.cc' at line 754 failed. (Shoals:3)
08:59:45 <greensnark> Anybody here knows whether Fedhas mushrooms are supposed to attack fungi?
09:00:11 <greensnark> !lm * type=crash -log
09:00:12 <Henzell> 46. Voker57, XL14 DSGl, T:35764 (milestone): http://crawl.develz.org/morgues/trunk/Voker57/crash-Voker57-20100620-135758.txt
09:03:31 <syllogism-> sounds unlikely
09:14:56 <greensnark> Ok
09:15:13 <CIA-41> 03greensnark * r244cd68d0984 10/crawl-ref/source/ (describe.cc mon-act.cc): Don't allow Fedhas mushrooms to attack plants and fungi to reach player.
09:20:21 -!- Amonchakad has quit [Quit: When two people dream the same dream, it ceases to be an illusion. KVIrc 3.4.2 Shiny http://www.kvirc.net]
09:23:20 -!- pointless_ has joined ##crawl-dev
09:35:44 <CIA-41> 03greensnark * r173c23167b3d 10/crawl-ref/source/ (dbg-maps.cc dungeon.cc dungeon.h): Fix -mapstat aborting early because of broken isolated-zone check; fixed to use dgn_count_disconnected_zones.
09:37:18 <greensnark> Apparently plants have deflect missiles active in my games
09:38:23 <greensnark> KiloByte: I don't think g++ likes 64-bit enum constants
09:39:02 <greensnark> Since we don't actually use mons_class_flags as an enum, I think we should just replace the lot by const uint64_ts
09:39:05 * greensnark does so.
09:45:10 <Enne> Thanks, C++, for having no cross-platform way to specify the size of an enum.
09:50:19 <greensnark> It would be nice if C++0x were around already
09:50:40 <greensnark> But apparently we must be tormented by hearing about all the things that are fixed in 0x but not usable in current compilers yet
09:51:42 <CIA-41> 03greensnark * r7cec2d55cd5b 10/crawl-ref/source/ (mon-class-flags.h mon-enum.h): Replace mons_class_flags with a set of const uint64_ts, since C++ doesn't guarantee support for 64-bit enum constants, producing bizarre effects such as plants having deflect missiles.
09:52:04 <casmith789> greensnark: oh that was an actual bug?
09:52:12 <casmith789> I thought you were just missing plants a lot
09:52:19 <greensnark> :P
09:52:33 <bhaak> you don't have the right to complain if you're using a language from the 80ies ;-)
09:53:07 <greensnark> I believe NH has to use multiple flags members in their monster structs :D
09:53:23 <Enne> That's a bug waiting to happen.
09:53:29 <greensnark> Whenever I think Crawl dev sucks, I remind myself of NH and feel better :)
09:53:36 <bhaak> yeah and do you see me complaining about a language from the 70ies?
09:53:46 <greensnark> bhaak: You never complain!
09:54:00 <greensnark> That would be unnatural in normal people, but you're Swiss, so that's understandable
09:54:01 -!- Zaba has quit [Ping timeout: 276 seconds]
09:54:03 <greensnark> ;p
09:54:09 <casmith789> UNH has a good active devteam
09:54:27 <greensnark> bhaak is cool
09:54:33 <greensnark> Does Un have other developers?
09:54:44 <greensnark> I think Spork is lorimer + paxed, right?
09:54:54 <bhaak> greensnark: you just don't decipher HOW swiss people are complaining
09:55:31 <bhaak> greensnark: "couldn't we think about doing it differently?" == "your proposal is bullshit and my idea is worthy of a nobel price"
09:55:45 <greensnark> :)
09:57:26 <greensnark> I see miscreants have been crashing FightClub :P
09:57:34 * greensnark stabs casmith789 with a demon trident of flaming.
09:57:37 <casmith789> haha
09:57:41 <casmith789> that was yesterday
09:57:43 <greensnark> Interesting crash :)
09:57:44 <casmith789> I said then :(
09:58:00 <casmith789> what, 99 loms v 20 cerebovs?
09:58:23 <casmith789> also... you stabbed me with a demon trident?
09:58:24 <greensnark> The crash dump rather carelessly neglects to mention command-line args
09:58:53 <greensnark> Crash appears to be in good old CrawlStoreValue
09:59:01 <greensnark> casmith789: Negligible stabbing bonus!
09:59:06 <casmith789> exactly
09:59:08 <greensnark> I reserve the demon tridents for my friends
09:59:16 <greensnark> If I really dislike a person I'd use a dagger
09:59:20 <casmith789> I feel honoured
09:59:24 <greensnark> :D
10:00:20 -!- ortoslon has quit [Quit: bye]
10:04:33 -!- Zaba has joined ##crawl-dev
10:09:38 <CIA-41> 03greensnark * r37516bb29136 10/crawl-ref/source/ (dbg-asrt.cc initfile.cc state.h): Include Crawl command-line arguments in crash dump.
10:13:13 <greensnark> Hmm, the arena is not a happy place in HEAD
10:15:22 -!- Zaba has quit [Ping timeout: 264 seconds]
10:18:04 <CIA-41> 03dolorous * r8c2721b182af 10/crawl-ref/source/items.cc: Add spacing fixes.
10:18:11 -!- rax has quit [Ping timeout: 265 seconds]
10:19:48 -!- Zaba has joined ##crawl-dev
10:30:41 -!- phyphor has quit [Read error: Connection reset by peer]
10:30:55 -!- phyphor has joined ##crawl-dev
10:49:13 -!- rax has joined ##crawl-dev
10:50:52 <Poor_Yurik> numpad 5 is wait one turn, but shift+numpad 5 doesn't initiate resting
11:00:01 -!- ortoslon has joined ##crawl-dev
11:04:57 <Zao> If you're using putty, the Nethack numpad option is supposedly sane.
11:19:47 <Poor_Yurik> Not sure what you mean.
11:20:05 <Poor_Yurik> Enne, forgot to send you door tiles.
11:34:41 <CIA-41> 03pointless_ 07articulated_kraken * r867f45f6847a 10/crawl-ref/source/mon-abil.cc: Switch kraken tentacle pathing to a weighted distance function
11:35:51 <CIA-41> 03greensnark * rb96501a03be5 10/crawl-ref/source/arena.cc: Force view_lock=true for arena (else arena crashes).
11:41:27 -!- ogaz has joined ##crawl-dev
11:46:19 <CIA-41> 03dolorous * r6c83f2ce2f80 10/crawl-ref/source/itemname.cc: Add missing parentheses.
11:59:19 -!- Siber has joined ##crawl-dev
12:06:48 <syllogism-> "Well that sure was fast, accidentally closed the game and now my save is corrupt and crashes at loading" doesn't that just save the game
12:12:43 <greensnark> No idea if tiles saves on closing game
12:12:55 <greensnark> Console does not save if you close the terminal window
12:13:52 <Enne> Tiles does.
12:14:19 <syllogism-> he was playing tiles yes
12:20:52 <greensnark> syllogism-: Should ask him to report a bug :P
12:20:57 <greensnark> And attach save
12:30:43 -!- Textmode has joined ##crawl-dev
12:32:49 <Textmode> morning all
12:33:09 <greensnark> Crazy yanks wake up in the middle of the night!
12:33:34 <greensnark> Although your hostname says syd and au, so I'm not sure where you are :P
12:33:48 <syllogism-> :P
12:34:34 <greensnark> eliotn wants instructions on writing a Crawl bot
12:34:40 <greensnark> Someone drop into ##crawl and enlighten them
12:34:54 <greensnark> syllogism-: You should post the link to MrPeeps' flowchart :P
12:37:05 <ogaz> but it's not accurate anymore!
12:45:11 <greensnark> Maddeningly obscure crash in CrawlStoreValue :(
12:45:42 <greensnark> Or quite possibly not in CrawlStoreValue at all
12:46:01 <greensnark> Everything goes south once the game gets the impression that one Cerebov firestorm  fire vortex has summoned another fire vortex.
12:46:15 <greensnark> I'm a bit hazy on how it gets this idea
12:48:03 <pointless_> Cerebov kills himself with firestorm, a vortex is placed at his mindex?
12:49:52 <greensnark> Looks like it
12:50:04 <greensnark> Still not sure why it's crashing though
12:51:45 -!- Amonchakad has joined ##crawl-dev
12:58:46 <KiloByte> greensnark: heh, about specifying enum's type, the very uncommited change I wrote before leaving for an aunt's party messes heavily with compiler support because of object_class_type inside item_def
12:59:21 <KiloByte> we waste 8 bytes per item because enums are ints
13:00:14 <greensnark> padding menace
13:01:53 <KiloByte> without the base type (which specifically uses 8-bit masks), we have 3 bytes left in no-waste 7*8 bytes structure
13:02:19 <KiloByte> if it's defined as char, it's ok -- if not, item_def jumps to 8*8 bytes
13:06:14 <greensnark> Ok, looks like the crash was just a case of trying to call a monster its own summoner
13:06:57 <KiloByte> GCC accepts typed enums but spams warnings if you don't specify --std=c++0x, clang gets a wonderful compiler crash
13:07:36 <greensnark> Let's not even talk about MSVC, eh :)
13:07:50 <pineapple> msvc prolly works,going by that
13:08:14 <greensnark> New MSVC might
13:08:20 <greensnark> They do claim to support a lot of 0x
13:08:29 -!- eith has joined ##crawl-dev
13:11:50 <KiloByte> it does
13:12:42 <greensnark> So all we need is for clang to catch up and we can use typed enums
13:12:43 <KiloByte> the problem with gcc is that if you do specify --std=c++0x it in turns spams warnings about auto_ptr being deprecated
13:12:49 <greensnark> Ack
13:12:53 <greensnark> Can't we turn that off?
13:13:38 <KiloByte> I guess we can, or convert auto_ptrs to whatever the replacement is called
13:13:48 <greensnark> Do we save much space if we use uint16_t for coord_def, or does padding destroy that anyway?
13:13:57 <Zao> Depends on usage, but shared_ptr tends to be a decent match.
13:14:23 <greensnark> Er, int16_t
13:18:22 <CIA-41> 03greensnark * r239738eefd7d 10/crawl-ref/source/mon-place.cc: Fix crash when Cerebov's firestorm kills him and creates a fire vortex on his mindex (casmith789).
13:18:35 <casmith789> :D
13:20:48 <greensnark> casmith789: Should have waited until I rebuilt the FIghtClub crawl :P
13:21:06 <casmith789> ;)
13:21:34 <greensnark> WTF :P
13:25:43 -!- varmin has quit [Remote host closed the connection]
13:25:55 -!- varmin has joined ##crawl-dev
13:26:15 <greensnark> Well, that was surprising
13:26:26 <greensnark> I had to kill the bot to make it stop using the bloody stale Crawl :P
13:33:27 <CIA-41> 03kilobyte * ra201159a83c1 10/crawl-ref/source/ (19 files): s/marshallLong/marshallInt/, since this confusion tends to cause bugs.
13:33:31 <CIA-41> 03kilobyte * r1f34e725521a 10/crawl-ref/source/ (9 files): Kill some longs from player.h
13:33:31 <CIA-41> 03kilobyte * rf4f8547af102 10/crawl-ref/source/clua.cc: Disallow longs in the Lua interface.
13:33:34 <CIA-41> 03kilobyte * rf1f190d3e3f4 10/crawl-ref/source/ (store.cc store.h): Remove CrawlTableWrapper, unused since times long ago.
13:33:38 <CIA-41> 03kilobyte * r0fe9a9e87802 10/crawl-ref/source/ (8 files): Make a few flag types use non-ambiguous storage.
13:33:42 <CIA-41> 03kilobyte * r0fa67c59d6d7 10/crawl-ref/source/ (17 files): Get rid of ambiguous longs from CrawlHashTable and a lot of misc places.
13:33:43 <CIA-41> 03kilobyte * recfbc8411e56 10/crawl-ref/source/ (externs.h itemname.cc itemname.h): Define make_name()'s seed as 32 bit -- the rest was unused anyway.
13:33:50 <CIA-41> 03kilobyte * ree530f001a95 10/crawl-ref/source/ (AppHdr.h enum.h): Use typed enums -- or a non-enum field -- in item_def to save memory.
13:34:54 <KiloByte> greensnark: the last commit uses that typed enums thing, I wonder if we should use that or look for some other way
13:35:58 <greensnark> Nice
13:36:32 <greensnark> I'd never even heard of CrawlTableWrapper
13:37:58 <greensnark> KiloByte: Seems fine to me
13:38:21 <greensnark> Although we do lose a little type safety without typed enums now
13:38:43 <KiloByte> in that single place, yeah
13:39:02 <greensnark> !lm * type=crash
13:39:02 <Henzell> 46. [2010-06-20] Voker57 the Necromancer (L14 DSGl) ASSERT(smc) in 'mon-util.cc' at line 754 failed on turn 35764. (Shoals:3)
13:39:04 <greensnark> !lm * type=crash -tv
13:39:04 <Henzell> 46. Voker57, XL14 DSGl, T:35764 (milestone) requested for FooTV.
13:39:52 <greensnark> Mmm, reaching attack + spines = kerboom
13:39:59 <Zao> http://codepad.org/muE53kQi
13:40:18 <Zao> That code snippet looks quite odd.
13:41:19 -!- ais523 has joined ##crawl-dev
13:41:40 <Zao> Oh never mind, subtle.
13:42:54 <KiloByte> gcc handles this correctly, clang spams warnings
13:43:48 <KiloByte> it's a Perlish idiom some people frown at, we can fix that in our code but ncurses use that as well
13:46:04 <KiloByte> a different question: could someone run the Mac builds?  Neunon won't have any time until midway next week, and people complain about no Mac binaries.
13:46:39 <Zao> The only Mac I've got access to is PPC, which I assume is rather uninteresting.
13:49:03 <greensnark> Someone's been asking for mac binaries?
13:50:32 <KiloByte> http://crawl.develz.org/wordpress/0-6-1#comments
13:50:58 <greensnark> Mac users :(
13:51:11 * greensnark fears the Mac Xcode project.
13:51:51 <CIA-41> 03greensnark * r8b086d5ec5b8 10/crawl-ref/source/spells3.cc: Fix lingering ref to unsigned long breaking 32-bit compile.
13:51:52 <Zao> Can't you use use the makefiles?
13:51:52 <CIA-41> 03greensnark * r2efa7e6bfc70 10/crawl-ref/source/mon-act.cc: Fix crash when snapping turtle is killed by reaching attack hitting spines.
13:52:06 <Zao> non-tiles built with them as soon as I mangled them to work properly with my 10.4 SDK.
13:52:12 <greensnark> Can't use the makefile, no
13:52:27 <greensnark> The Xcode project does a bunch of extra stuff to create fancy disk images
13:52:31 <greensnark> With all deps included in the dmg
13:52:57 <greensnark> And usually the only way I can get it to work is by massive hand hacking that removes a lot of stuff that looks like it's important to neunon :P
13:52:59 <Zao> Ah.
13:53:03 <Zao> Fancypants.
13:55:05 <greensnark> I suppose I can give it a shot and hope it works this time
13:55:06 <KiloByte> greensnark: ugh, sorry for 8b086d5e :p  Didn't test 32 bit after every step (and the crusade against ambiguous longs is not complete).
13:55:29 <greensnark> KiloByte: No issues, I make worse mistakes all the time :P
13:55:38 <KiloByte> if it costs you nothing but CPU time, then that can be cool
13:55:49 <KiloByte> (trying the build I mean)
13:56:21 * greensnark gives it a whirl
13:56:49 <CIA-41> 03dolorous * r4b241fe7de48 10/crawl-ref/source/spells3.cc: Add missing parentheses.
13:57:11 <greensnark> Wow, no errors so far :P
13:57:37 <greensnark> Maybe neunon fixed it post 0.5?
13:57:45 <greensnark> Ok, at last the errors :D
13:58:01 <greensnark> Oh, just -pedantic whinging
13:58:31 <KiloByte> nuke it, it's gone from Unix makefiles
13:58:36 <greensnark> Yes
13:59:21 <greensnark> Oh hang on, I'm building 0.7 :P
14:00:27 <greensnark> But the Xcode project seems in much better shape
14:00:33 <greensnark> I guess neunon fixed it
14:00:58 <KiloByte> cool
14:01:20 <KiloByte> I don't get dolorous talking about "missing parentheses":
14:01:21 <KiloByte> -        name_type = name_flags & MF_NAME_MASK;
14:01:21 <KiloByte> +        name_type = (name_flags & MF_NAME_MASK);
14:02:03 <greensnark> If it makes him happy :P
14:02:15 -!- ortoslon has quit [Quit: bye]
14:02:24 <greensnark> It was amusing at one point with some folks removing parentheses and others adding them back
14:02:53 * greensnark is in the laissez faire camp. If it compiles and works okay and is not heinously evil, let it be.
14:03:48 <KiloByte> yeah, unless they get into the Lisp land, it's harmless
14:05:19 <bhaak> hmm, this example looks unnecessary to me, but something like this is ugly:  if (bool1 && bool2 || bool3) { [...]
14:05:34 <greensnark> bhaak: jpeg liked that style for some reason :)
14:05:45 <bhaak> ieks!
14:05:56 <greensnark> I prefer parenthesising the x && y, but whatever
14:06:24 <greensnark> Xcode is crying about missing tiledef-floor wall and feat
14:06:59 <bhaak> IMO the code should be clear to somebody who doesn't dream in C :)
14:07:05 <greensnark> The problem with the Xcode project is that it can never stay in sync with the makefiles
14:12:23 <KiloByte> couldn't the packaging be done from the makefiles?
14:12:51 <KiloByte> heck, I've ran Inno Setup through wine in makefiles before
14:13:45 <greensnark> Everything can be done from the command-line
14:13:48 <greensnark> So can be in makefile
14:16:53 <greensnark> Ok, I'll call it quits on this
14:17:09 <greensnark> I'm not particularly in the mood to figure out why the sdl submodule wants to build the X11 frontend on Mac
14:17:30 <greensnark> The Mac binary users will have to wait
14:17:53 <Mu_> sprint.des:802: Ego 'returning' is incompatible with item 'large rock'.
14:18:02 <Mu_> my stone giant gimmick ;_;
14:18:14 <Mu_> i don't suppose there's any way around this is there? :p
14:19:55 <greensnark> Mu_: Just give them extra rocks? :P
14:20:26 <Mu_> not as amusing but ok :p
14:20:30 <KiloByte> Mu_: we can just make the checks less nazi, doing that would break "any missile ego:returning", but I don't think that's used in any place
14:21:07 <greensnark> Does the code handle large rocks of returning correctly
14:21:09 <KiloByte> only armours in a Volcano and an Ice Cave do that
14:21:19 <greensnark> If so yeah, make an exception for vaults
14:21:46 <syllogism-> The +0 large rock of returning returns to your pack!
14:21:49 <ogaz> greensnark: they were possible for a couple of weeks(before the returning ammunition spell was removed) and I don't remember anything going wrong
14:22:02 <Mu_> cool :)
14:22:04 <greensnark> Ok
14:22:11 <greensnark> KiloByte: You making that change? :)
14:22:16 <sorear> Why are the checks so nazi to begin with
14:22:21 <KiloByte> change makeitem.cc : is_missile_brand_ok
14:22:31 <greensnark> I think zelgadis was very concerned about inappropriate brands
14:22:47 <greensnark> It used to be possible to make melee weapons with missile brands, etc.
14:22:48 <sorear> They take all the fun out of fightclub
14:22:51 <greensnark> So he got rigrous
14:23:01 <greensnark> And when zelgadis gets rigorous, he gets pretty rigorous :)
14:23:03 <syllogism-> silver large rock <3
14:23:15 <ogaz> steel boulders please
14:23:21 <greensnark> BALLS OF STEEL?
14:23:21 <KiloByte> they were added later, we (with due) needed a way to select any item which can legally get a given brand
14:23:41 <greensnark> I'm surprised due hasn't enabled custom item names already
14:24:19 <greensnark> KiloByte: Oh, you added the checks? :P
14:24:22 <KiloByte> I added these checks but indeed the asserts are not that needed
14:24:28 <greensnark> Heh
14:24:42 <KiloByte> or we could give it an argument "strict"
14:25:12 <greensnark> Oh crap, all this time I was building an oold version of 0.6
14:25:15 * greensnark needs sleep.
14:27:47 <greensnark> !lm * crash -2
14:27:48 <Henzell> Malformed argument: crash
14:27:53 <greensnark> !lm * type=crash -2
14:27:53 <Henzell> 45. [2010-06-20] Soadreqm the Nimble (L10 SETm) ASSERT(!monster at(p) || monster at(p)->submerged() || fedhas passthrough(monster at(p))) in 'player.cc' at line 347 failed on turn 13745. (Orc:2)
14:29:14 <greensnark> Looks like process_decription crashes on randarts with too many properties
14:29:21 <greensnark> Joy
14:29:42 <greensnark> Finally a real advantage of large terminals :P
14:32:42 <CIA-41> 03pointless_ 07articulated_kraken * rd3be3a242ad3 10/crawl-ref/source/mon-abil.cc: Do a better job of tracking tentacle connections
14:32:45 <CIA-41> 03pointless_ 07articulated_kraken * re4c84a0d7f9a 10/crawl-ref/source/ (mon-abil.cc mon-cast.cc): get_long -> get_int function name changes.
14:32:46 <CIA-41> 03pointless_ 07articulated_kraken * r19238515a4e7 10/crawl-ref/source/ (69 files in 4 dirs): Merge branch 'master' of ssh://crawl-ref.git.sourceforge.net/gitroot/crawl-ref/crawl-ref into articulated_kraken
14:33:10 <sorear> articulated kraken sound like a lot of fun
14:33:17 <greensnark> Yes!
14:33:20 <sorear> (is kraken its own plural?)
14:33:21 * greensnark wants them in master
14:33:40 <pointless_> it's basically functional now, but I've caused some messaging/death handling regressions
14:34:23 <bhaak> sorear: yes
14:34:25 -!- Amonchakad has quit [Quit: When two people dream the same dream, it ceases to be an illusion. KVIrc 3.4.2 Shiny http://www.kvirc.net]
14:35:33 * bhaak didn't know that crawl commit messages were scrutinized for their grammar :-)
14:36:22 <sorear> from #angband, on a randarts game: 14:33 < Derakon> The Mattock 'Dildo' (1d8) (+8,+13) (+4 CON/tunnel), resist  fear/nether.
14:36:32 <pointless_> they are also scrutinized for whitespace useage...
14:37:12 <greensnark> Nice mattock :P
14:40:20 <Mu_> is there a maximum number of shops that a map can have
14:40:23 <greensnark> resist nether is a particularly nice touch
14:40:35 <greensnark> Mu_: 25
14:40:44 <sorear> I thought it was 5
14:40:51 <sorear> or is that only for non-vaults
14:40:55 <greensnark> That's the max randomly generatable shops
14:41:08 <Mu_> oki, that explains the feature bugs in this sprint map
14:41:40 <ais523> sprint keeps hitting the items-per-limit level too, or at least used to; often some of the guaranteed gold didn't generate
14:41:46 <ais523> but that's memory from months ago
14:42:03 <greensnark> We could up the limits
14:42:23 <Mu_> my map is a lot larger than the original sprint map too
14:42:25 <greensnark> We should also switch to a better way of saving levels so upping limits doesn't bloat levelfiles
14:43:36 <sorear> The server people are very resistant to large changes in save formats
14:44:31 <greensnark> SQLite saves are the future :)
14:45:29 <KiloByte> SQLite won't help with this issue
14:46:02 <greensnark> No, but I'm reerring to "large changes in formats"
14:46:38 <KiloByte> hmm... actually, upping the items limit shouldn't break anything, there's no point in saving entries past the last used index
14:46:53 <greensnark> For instance marshallMonster saves the entire monster whether it's alive or not
14:46:58 <bhaak> a database structure alone doesn't ease the problem of changing the savegame structure?
14:47:23 <greensnark> bhaak: We're not proposing to use extraordinarily detailed tables here :P
14:47:27 <greensnark> That would be too much work :P
14:47:37 <greensnark> We're just proposing to put each level's data into one blob :P
14:47:59 <greensnark> So the inefficient code that marshalls level data is still a problem
14:48:22 <greensnark> But using SQLite gives us one clean save file
14:48:26 <bhaak> erm, one blob? that would solve anything?
14:48:29 <greensnark> Without the custom massacre of NH's format
14:48:44 <greensnark> bhaak: Currently Crawl saves each level's data into its own file on disk
14:48:49 <greensnark> One file for D:1, one for D:2 etc.
14:49:09 <greensnark> As a concession to presenting it as one savefile, it will tar or zip it if you want :P
14:49:20 <bhaak> oh, nice :)
14:49:33 <greensnark> NH has that save archive code, right?
14:49:44 <greensnark> Although NH's save loading got exceedingly slow the further you go into the game
14:50:04 <greensnark> Using SQLite will give us instant access to level data, with SQLite ensuring safe writing of levels, etc.
14:50:21 <greensnark> For instance, if you run out of disk space, Crawl will smash level files :P
14:50:26 <greensnark> That kind of problem will go away with SQLite
14:50:29 <bhaak> while you're playing, each level is saved as one file by NH, too
14:50:34 <greensnark> And we can add all kinds of juicy metadata
14:50:43 <KiloByte> SQLite isn't very efficient for blobs, and since we would store nothing but blobs, it's not such a hot idea
14:50:51 <greensnark> KiloByte: How inefficient?
14:51:11 <greensnark> I plan to try it and see
14:51:12 <bhaak> there are SQL db that are efficient with blobs? :)
14:51:16 <KiloByte> it can't handle fragmentation, for one
14:51:18 <greensnark> Our blobs are not particularly huge anyway
14:51:19 <KiloByte> bhaak: no
14:51:30 <bhaak> KiloByte: thought so
14:52:07 <greensnark> bhaak: Isn't there some NH code that combines everything into one save file at some point?
14:52:39 <bhaak> greensnark: yes, when you're saving. but when you're playing every level you leave is swapped out to disk as a separate file
14:53:41 <bhaak> I'm surprised it works as good as it does
14:54:22 <greensnark> There is a long unpack time for a game in Gehennom, right?
14:54:30 <greensnark> Possibly much better now with modern hardware
14:54:43 -!- Niccus has joined ##crawl-dev
14:55:16 <bhaak> on the amiga, you had 1 or 2 seconds delay every time you left a level
14:55:17 <KiloByte> heh, saving items only until the last used one is a low hanging fruit for reducing save files
14:55:44 <KiloByte> and it will crash if the limit is ever reduced
14:55:58 <bhaak> nowadays, you don't notice it. although I've heard, the iphone version had some problems when the OS sent the suspending command to the game and it couldn't save in time.
14:56:45 <greensnark> Ah, the iphone is like 1990's hardware :)
14:56:58 <CIA-41> 03kilobyte * r7ed7668902bf 10/crawl-ref/source/effects.cc: Fix a part of armour acq code being ignored.
14:57:01 <CIA-41> 03kilobyte * rfb5802231e98 10/crawl-ref/source/ (9 files): Placate clang's warnings about shadowing class variables in class functions.
14:57:02 <CIA-41> 03kilobyte * r8ca9b209fbb1 10/crawl-ref/source/exclude.cc: Fix an invalid return value.
14:57:02 <greensnark> But the iphone version probably doesn't even need to pack the save, no?
14:58:05 <bhaak> even if you don't zip the save, it takes some time to copy the level data into the savegame
14:58:29 <greensnark> bhaak: Why not leave the level data as individual level files on iphone?
14:58:41 <greensnark> There's no compelling advantage to having it in one file there that I can see
14:58:44 <bhaak> AND the savegames are rather large. it might even be faster to run a gzip (not bzip2 :-) process
14:58:57 <greensnark> Oh, I'm not refering to compression :)
14:59:05 <bhaak> but you have to touch the savegame routines for that1 :-)
14:59:19 <greensnark> bhaak: You have to touch a lot of code to make it run on iphone, no? :P
14:59:38 <ais523> you have to rewrite all the Lua as something else
14:59:48 <greensnark> ais523: We're talking about NH iphone port
14:59:53 <ais523> ah
14:59:56 <bhaak> greensnark: surprisingly few code. I could drop unnethack into the code base and it just worked
14:59:58 <ais523> sorry, the channel confused me
15:00:03 <greensnark> Apple will probably allow Lua anyway, they're more interested in killing Flash
15:00:14 <greensnark> They've made some concessional change to their agreement
15:00:28 <bhaak> yeah, what greensnark said
15:00:36 <ais523> greensnark: they changed it from "this is banned" to "this is allowed but only if we agree to it in writing"
15:00:42 <greensnark> Yes
15:00:47 <ais523> which is effectively the same, failing a method of predicting the way Apple will behave
15:00:52 <greensnark> So if you're a prominent game dev you'll probably get approval
15:01:01 <bhaak> but the whole appstore process is like that!
15:01:12 <bhaak> you never know if your program gets through or not
15:01:12 <greensnark> If you're a poor schmuck apple will give you the perm in writing in a few years or so
15:02:37 <greensnark> Crawl will probably need at least a 3GS to run anyway :P
15:02:47 <greensnark> Unlike NH, it is not at all worried about conserving system resources
15:03:09 <greensnark> The more recent iPod touches with 256M of RAM should also cut it
15:03:25 <greensnark> But they're likely to be like playing the game in valgrind :P
15:04:01 <bhaak> I'll could donate an optimization book to the crawl devs :)
15:06:22 * greensnark slaps bhaak with a trout.
15:08:19 <ais523> greensnark: Crawl's playable inside valgrind
15:08:22 <ais523> or was last I checked
15:08:27 <ais523> even if it is a bit annoyingly slow
15:08:37 <greensnark> You are waay more patient than I am, it seems :P
15:08:53 <greensnark> Ok, I also use debug builds with valgrind, so that adds a lot of overhead
15:08:55 <greensnark> But still!
15:10:01 <ais523> hmm, I should go back to working on my ttyrec player
15:10:12 <ais523> it's been handling Crawl just fine for a while now
15:10:42 <bhaak> \o/
15:11:01 <ais523> I want to get it handling things like termcast and dgamelaunch streams
15:11:01 <bhaak> nethack-de's website needs a new ttyrec player!
15:11:08 <sorear> crawl cares abount conserving system resources once a year
15:11:22 <ais523> it can already handle streaming, but atm only over TCP (with or without timestamps embedded)
15:11:35 <sorear> in about two months Crawl needs to support running 50 instances on a 1GB virtual server
15:11:54 <greensnark> I think 2009 had 70+ active on first day
15:11:56 <greensnark> on cao
15:12:05 <greensnark> 200908 I mean
15:12:14 <ais523> that's 20MB each; that doesn't seem too ridiculous
15:12:17 <greensnark> Load average was galloping into double digits :P
15:12:19 <ais523> it's not like Crawl is Dwarf Fortress
15:12:36 <bhaak> easy, token system with e-mail notification when your CAO place is free :-)
15:12:57 <greensnark> The way players complain about lag normally, I'm surprised so many players managed to play without complaints first day  :)
15:13:00 <ais523> bhaak: nah, use 20 or so distributed servers with an rsync-based method of registering users that takes about 15 minutes to work
15:13:03 <sorear> ais523: and our 20MB .text can be shared...
15:13:08 <Ashenzari> Withered plant awkwardly described as "It is missile deflection." (https://crawl.develz.org/mantis/view.php?id=1800) by Twilight
15:13:09 <ais523> some of which crash at random
15:13:19 <greensnark> Heh, it is missile deflection
15:13:37 <syllogism-> but it is :P
15:13:40 * sorear wants to know who came up with the name .text
15:13:55 <ais523> probably the same people who wrote the first ld
15:13:58 -!- eith has quit [Ping timeout: 245 seconds]
15:14:25 <CIA-41> 03kilobyte * r76c433efdaab 10/crawl-ref/source/tags.cc: Conserve disk space by not saving items and traps past the last used slot.
15:15:39 <greensnark> <3
15:15:50 <greensnark> KiloByte: Can you also add code not to save stuff in empty slots?
15:15:58 <greensnark> Such as dead monsters
15:16:49 <greensnark> If monster type is non-monser, don't save the rest of it, and similar for items
15:20:01 <KiloByte> reordering that would require a fair bit of code or breaking save compat, but yeah, it can be done
15:20:34 <greensnark> Well, a minor version bump would fix that, no?
15:20:46 <greensnark> It's not as if we ever pretended that older version can load newer versions' saves :)
15:21:27 <KiloByte> monsters, clouds and shops do suffer from the same thing as items too
15:21:31 <greensnark> Yes
15:21:50 <greensnark> I think clouds and shops at least have the decency to be small structures
15:22:14 <greensnark> Well, smallish :P
15:22:21 <greensnark> Small by Crawl standards :P
15:22:27 <KiloByte> I wonder how safe it would be to renumber clouds and shops, that would let us not save empty slots at all
15:22:48 <greensnark> That sounds complicated
15:23:02 <ogaz> !time
15:23:02 <Henzell> Time: Jun 20, 2010, 08:23:02 PM, UTC. The 2010 tournament starts in 41 days, 3 hours, 36 minutes and 57 seconds.
15:23:05 <greensnark> You you'll have to check what code is hanging on to indexes for arbitrary crap
15:23:38 <greensnark> But most of these structs have a well-defined type; if that type is invalid, just skip the rest still saves us a fair bit
15:23:49 <greensnark> ogaz: Can't wait to get cracking? :D
15:23:59 <KiloByte> monsters have foe indexes, kraken tentacles keep their head's numbers, other things like summon chains may use that too
15:24:22 <greensnark> Summons just keep a string referencing their summoner, I think
15:24:27 <KiloByte> ah
15:24:46 <greensnark> I think poison source is also just a string, right?
15:25:00 <KiloByte> two strings as a prop, but yeah
15:25:00 <greensnark> But anyway, I wouldn't mess with renumbering :)
15:25:24 <ogaz> greensnark: mostly just curious
15:26:00 <KiloByte> actually, delayed kill blaming should be reworked, probably as a structure with all the fields like ikiller, so they are consistent
15:27:28 <KiloByte> if we use your idea of saving type first and skipping the rest if the type is empty, together with skipping indices past the last one used should be good enough
15:29:14 -!- Amonchakad has joined ##crawl-dev
15:30:16 <sorear> Actually I've been experimenting with keeping our saves in some form of JSON
15:30:21 <sorear> But don't tell anyone
15:30:30 <greensnark> :)
15:32:13 <KiloByte> heh, env.grid_colours are stored compressed already :p
15:32:23 <KiloByte> just run length encoding, but still...
15:42:48 -!- Xiberia has quit [Ping timeout: 265 seconds]
15:46:39 -!- syllogism- has quit [Ping timeout: 240 seconds]
15:46:42 -!- syllogism_ has joined ##crawl-dev
15:52:56 <KiloByte> absolute freaking "yay": the monsters array is saved as two separate chunks, one for all fields but attitude and foe, another for those three bytes per monster...
15:53:28 <ais523> so monsters are slightly too big to compress well?
15:54:05 <KiloByte> uhm no, they have loads of fields and are of variable length already
15:56:35 <sorear> it makes sense in context
15:56:59 <sorear> if you read the comments, you'll see that tag definitions are never supposed to be changed, ever
15:57:12 <sorear> if new data needs to be saved, you're supposed to create a new tag for it
15:57:18 <sorear> this has... not been followed well
15:58:12 <sorear> which is why I want to move Crawl to a more sane compatibility scheme, but I haven't worked out all the details yet
15:58:12 <KiloByte> minor save versions handle that in a far better way
16:02:31 <greensnark> Expecting people to create a new tag for every minor change was ridiculous anyway
16:02:31 <KiloByte> but... will anyone complain if I get lazy and screw save compat while messing with saving the arrays?  I mean reordering them so type is always first (and the rest of empty slots is skipped), and now dropping that attitude tag?
16:02:50 <greensnark> KiloByte: Just bump major version, this is master branch :P
16:03:09 <KiloByte> .... the count of monsters is saved as a char...
16:03:19 <KiloByte> (MAX_MONSTERS is 350)
16:04:06 <KiloByte> if you wanted to get rid of a ballisto infestation, here's how :p
16:04:20 <greensnark> Haha
16:04:30 <ais523> KiloByte: genius
16:04:37 <greensnark> I increased MAX_MONSTERS, and must have forgotten that
16:04:41 <greensnark> That bug has been around for a while :P
16:04:52 <ais523> so just get exactly 256 monsters on the level, save, come back
16:05:01 <ais523> hmm, I wonder if you could set that off deliberately in Zot?
16:05:23 <greensnark> Fedhas probably can
16:06:11 <ais523> or just spammals or something
16:06:32 <greensnark> What was the fix for sqlite wanting gethostuuid on Mac...
16:06:44 <greensnark> 10.4 SDK doesn't have gethostuuid
16:09:00 <greensnark> Ah I just have to define SQLITE_ENABLE_LOCKING_STYLE 0
16:09:07 <Amonchakad> an electric eel in a fountain - is that intended?
16:09:13 <Amonchakad> especially inside the hive?
16:09:13 <greensnark> Yes
16:09:16 <greensnark> Yes
16:09:18 <Amonchakad> kk
16:15:53 -!- TGWi has joined ##crawl-dev
16:17:24 <CIA-41> 03kilobyte * rf95b706f5344 10/crawl-ref/source/tags.cc: Don't save clouds and shops past the last used slot.
16:17:29 <CIA-41> 03kilobyte * r5fb26ded0bd1 10/crawl-ref/source/tags.cc: Don't save monsters past the last used slot.
16:17:30 <CIA-41> 03kilobyte * r1eab4f807822 10/crawl-ref/source/tags.cc: Clear the slots that were not saved.
16:18:20 <greensnark> KiloByte: Approx how much saving do we get for a newly generated level?
16:20:25 <KiloByte> what do you mean?  Most monsters and items are already there, so it shouldn't be much different from saving the same level later.
16:22:03 <greensnark> Yes, how much space saving from the old code :)
16:22:27 <greensnark> I think levels used to be >160k, right?
16:23:27 <KiloByte> oh, there's no reason to save the cgrid, 80*70*2 bytes that can be reconstructed from already saved data
16:23:37 <greensnark> cgrid?
16:23:42 <greensnark> Oh, clouds
16:23:46 <greensnark> Yeah, good point
16:40:52 <greensnark> KiloByte: Actually Xcode project works without modifications :P
16:41:12 <greensnark> I just had a dirty directory and wasn't paying attention :P
16:41:40 <greensnark> Watching world cup replays with Alt-Tab apparently messes with my concentration
16:48:40 <KiloByte> greensnark: both console and tiles?
16:49:16 <KiloByte> if so, upload time!
16:52:23 <greensnark> Yes, just verifying and cleaning up
16:57:23 <greensnark> Wretched macro save is broken :P
16:59:27 <CIA-41> 03greensnark 07stone_soup-0.6 * rb39e47933df1 10/crawl-ref/source/ (3 files in 2 dirs): Rename Mac ASCII .app wrapper script to the app name, fix its handling of spacey paths.
17:00:29 <greensnark> So.. should we fix the macro bug :P
17:00:56 <greensnark> Who even uses macros :P
17:01:17 -!- syllogism_ has quit [Ping timeout: 240 seconds]
17:01:18 -!- syllogism- has joined ##crawl-dev
17:01:31 <greensnark> It's the return of syllogism
17:11:20 -!- itsmu has joined ##crawl-dev
17:11:31 -!- Mu_ has quit [Ping timeout: 248 seconds]
17:11:39 -!- itsmu is now known as Mu_
17:13:10 -!- TGWi has left ##crawl-dev
17:17:58 <KiloByte> oops, sorry for breaking current master.  But thanks for "git rebase -i HEAD^^^" and "git push sf HEAD^^:master" (before the first save break)
17:18:05 <CIA-41> 03kilobyte * r05999fbde4b1 10/crawl-ref/source/ (defines.h env.h tags.cc): Use a #define instead of hardcoded 20 for the size of Pan mons_alloc, reduce that to 10.
17:18:06 <CIA-41> 03kilobyte * rad130f4461bf 10/crawl-ref/source/tags.cc: Fix an infinite loop while reading shops.
17:28:52 <Ashenzari> Draconian scales not getting harder (https://crawl.develz.org/mantis/view.php?id=1801) by Happylisk
17:29:46 <CIA-41> 03greensnark 07stone_soup-0.6 * r002299d37b23 10/crawl-ref/ (settings/init.txt source/initfile.cc): Don't use a default macro_dir line in init.txt, since initfile.cc sets it anyway + set a sane default macro_dir on Mac OS.
17:42:26 -!- Amonchakad has quit [Quit: When two people dream the same dream, it ceases to be an illusion. KVIrc 3.4.2 Shiny http://www.kvirc.net]
17:45:51 <greensnark> Ok, Mac binaries finally uploaded
17:51:55 <KiloByte> great!
17:57:31 <CIA-41> 03greensnark * r956fc093bdd8 10/crawl-ref/ (settings/init.txt source/initfile.cc): Don't use a default macro_dir line in init.txt, since initfile.cc sets it anyway + set a sane default macro_dir on Mac OS.
18:03:45 <KiloByte> the download page is updated
18:04:17 <greensnark> Thanks
18:05:04 <greensnark> Heh, I don't think I've seen the downloads page before
18:05:06 <greensnark> It looks pretty neat
18:10:49 <CIA-41> 03kilobyte * r96b03ebe2bdc 10/crawl-ref/source/ (tags.cc tags.h): Don't store monster attitudes as a separate tag. Breaks save compat.
18:10:50 <CIA-41> 03kilobyte * rec498f277f55 10/crawl-ref/source/tags.cc: When saving an unused item/trap/shop/monster/cloud slot, write just the type.
18:10:51 <CIA-41> 03kilobyte * r84d759255764 10/crawl-ref/source/tags.cc: Don't store the cgrid, reconstruct it from cloud data.
18:10:53 <CIA-41> 03kilobyte * r1196655e8bb6 10/crawl-ref/source/ (tag-version.h tags.cc): Bump the major save version to 26.
18:12:55 <CIA-41> 03kilobyte * rf4475de022a6 10/crawl-ref/source/tags.cc: Drop a now-useless #ifdef for save major 25.
18:13:46 -!- neunon has joined ##crawl-dev
18:17:31 <KiloByte> hi Neunon!
18:17:37 <neunon> Hi there.
18:18:04 <KiloByte> greensnark just managed to do the Mac builds, so that's ok
18:18:10 <neunon> Ah, neat
18:18:19 <neunon> one less thing to do this week :3
18:20:28 <greensnark> Hey, thanks for fixing the Xcode project, the Build button just worked <3
18:27:23 <greensnark> KiloByte: When I killed a kobold just now I got "a  corpse"
18:27:32 <greensnark> Would that be something to do with the corpse name changes?
18:30:20 <KiloByte> these should do anything only for renamed monsters
18:30:57 <greensnark> all corpses seem to have broken names atm
18:35:45 <KiloByte> hmm, can't seem to reproduce that
18:35:57 <greensnark> Oh
18:36:23 <greensnark> I'll look into it right after I've sorted out this vampire feed crash
18:37:51 <KiloByte> a kobold corpse, an orc corpse, the human corpse of Sigmund
18:38:04 <greensnark> Must be something specific to my setup
18:38:20 <greensnark> The Mac build is i386 by default, and it doesn't have typed enums
18:38:30 <greensnark> Could be one or the other
18:38:38 <greensnark> I'll look into it in a minute
18:39:21 <KiloByte> default gcc compile doesn't use them either (unless you add --std=c++0x)
18:43:09 <CIA-41> 03greensnark * r0b962904c2b3 10/crawl-ref/source/delay.cc: Fix crash when vampire feeding on a poisonous corpse is interrupted by thirst decreasing and removing PR.
18:45:37 -!- Mu_ has quit [Quit: Defecator, may everything turn out okay so that you can leave this place.]
18:50:59 <KiloByte> on a real i386 machine it works ok, too (Debian gcc 4.4.4)
18:51:57 <greensnark> http://sprunge.us/QcUG fixes it for me
18:52:04 <greensnark> It's not clear to me why nobody noticed this before :)
18:52:50 <greensnark> I don't normally test trunk much locally
18:55:15 <due> morning
18:55:29 <greensnark> It's due the great and terrible
18:55:31 <ogaz> afternoon
18:56:24 <due> yup :)
19:02:23 <CIA-41> 03kilobyte * re098adb7a0f5 10/crawl-ref/source/itemname.cc: Fix uninitialized corpse flags (Greensnark).
19:04:13 <KiloByte> due: elephants need less tramplage and more goring, their tusks are unused
19:04:47 <due> Hm.
19:04:57 <due> This is true.
19:05:02 * KiloByte tosses due a freshly poached chunk of ivory, then runs to sleep.
19:05:47 <greensnark> Elephant poachers cannot be tolerated
19:05:56 <greensnark> !send KiloByte swat hellephants
19:05:56 <Henzell> Sending swat hellephants to KiloByte.
19:11:19 -!- casmith789 has quit [Remote host closed the connection]
19:28:32 -!- syllogism- has quit []
19:51:13 -!- purge has joined ##crawl-dev
19:51:23 <due> heya purge :D
20:19:56 <CIA-41> 03greensnark * r3fa0fd0436ef 10/crawl-ref/source/describe.cc: Fix ASSERT when describing items with $LINES+1 description lines (such as the unrand "Leech" in an 80x24 term).
20:29:14 <purge> hi due :)
20:29:30 <purge> just got done reading backlog
20:30:02 <greensnark> Don't tell due what we said about his dire elephant plushie
20:30:04 <CIA-41> 03greensnark * r0ea2284c1655 10/crawl-ref/source/test/ (des/monster-plant-pathfind.des snake-rune.lua): Fix test failures -- debug map that looked unplaceable, Snake test failure if the Snake branch was disabled.
20:30:30 <purge> lol
20:35:49 <due> What did you say aboutmy dire elephant plushie?
20:36:15 <greensnark> Nothing at all
20:36:22 <due> :|
20:36:25 <pointless_> "Why is it shaped like a Y"
20:51:41 <CIA-41> 03greensnark * r32a1d487574a 10/crawl-ref/source/tags.cc: Fix non-debug build.
21:23:23 -!- daftfad has joined ##crawl-dev
21:23:56 <CIA-41> 03greensnark * r5fe3a2c87e79 10/crawl-ref/source/mon-util.cc: [1414] Fix crash when killing kraken zombie by destroying a tentacle.
21:24:47 <CIA-41> 03greensnark 07stone_soup-0.6 * r44fea5b81929 10/crawl-ref/source/mon-util.cc: [1414] Fix crash when killing kraken zombie by destroying a tentacle.
21:34:31 <Poor_Yurik> I was thinking Hellephants should have a bite attack that is slow, fairly bad against EV, but nasty.
21:34:37 <Poor_Yurik> obviously no trunk slap, as they have no trunk.
21:42:57 <purge> Personally i think anything with hell in its name needs to have a maw located somewhere on its body :)
21:47:57 <CIA-41> 03greensnark * r5b01a0268f0f 10/crawl-ref/source/directn.cc: [1485] Allow normal targeting of kraken tentacles.
21:53:18 <due> Maws are good.
21:55:37 <CIA-41> 03greensnark * raa7dfaa938f4 10/crawl-ref/source/dat/des/variable/altar.des: [1430] fedhas_altar_4 was a minivault without surrounding floor and without a mini_float tag to request exits to be generated, fixed.
21:55:54 <due> Someone's on firetonight.
21:56:56 <due> ??slime
21:56:56 <Henzell> slime pits[1/3]: Accessed from the Lair somewhere between levels 6 and 8. Six levels deep, with a rune in the loot vault at the bottom. Killing the royal jelly turns the walls around the loot transparent, makes them diggable, and allows teleport control.
21:57:10 <due> CanI force Slimeto appearonLair:8 using anentry map?
21:57:47 <greensnark> Sticky spacebar? :P
21:58:03 <due> Yeah.
21:58:08 <due> I've been trying to fix it all week, driving me crazy.
21:58:14 <greensnark> You can explicitly place a Slime entrance wherever you want, *but* that will mean there are duplicate entrances
21:58:15 <due> I want to have a Llair:8 ending that is also the slime entry.
21:58:27 <due> Even is I tag it slime_entry?
21:58:49 <greensnark> Well, the question is do you want it to be a Lair:8 primary vault
21:59:00 <greensnark> If you use PLACE: Lair:8, that will force it to be placed on Lair:8
21:59:03 <due> Yes
21:59:09 <greensnark> And if the slime entry was already placed, you will get a dupe
21:59:13 <due> Hm
21:59:15 <due> Damn.
21:59:37 <greensnark> If you just want to say: "*If* you are placing a slime entry vault on Lair:8, consider this vault" you can do that with a slime_entry and a DEPTH: Lair:8
21:59:48 <due> Ooh, that would work.
21:59:51 <greensnark> But then you will get that vault and the regular Lair:8 vault
21:59:54 <due> Yeah.
21:59:58 <greensnark> Assuming the level has room for both
22:00:04 <due> I coulld fix that with tags thoguh, couldn't I?
22:00:09 <due> l_uniq_foo?
22:00:24 <greensnark> If you use luniq_foo, the primary Lair:8 vault will always win
22:00:29 <due> Bugger
22:00:32 <greensnark> Because it's placed first
22:00:35 <due> Not fair!
22:00:43 <greensnark> Why not just accept both? :P
22:00:49 <greensnark> That'll be a fun Lair:8 for anyone :P
22:00:56 <due> BecauseI'm not quite that evil.
22:01:03 <greensnark> Practise!
22:01:09 <greensnark> Becoming evil is quite easy
22:01:17 <greensnark> It just takes a little work, and it's endlessly rewarding
22:01:34 <due> :)
22:02:01 <due> Okay, I'mrunning out of ideas for Lair:$ vaults.
22:04:45 <greensnark> due: Wormcave
22:04:56 <greensnark> Starring: spiny worms in writhing masses
22:05:16 <greensnark> A few lindwurms
22:05:30 <due> Hm.
22:05:31 <due> I like it.
22:05:50 <greensnark> And some regular worms
22:05:52 <st_> that would be as tedious as those horrible ant hive vaults
22:06:04 <due> But fun.
22:06:12 <greensnark> Depends on how many spinies are used
22:06:23 <due> @?spiny worm
22:06:23 <Gretell> spiny worm (09w) | Speed: 9 | HD: 12 | Health: 36-96 | AC/EV: 10/6 | Damage: 3208(acid) | Res: 06magic(48), 08acid+++ | Vul: 09poison | Chunks: 09poisonous | XP: 541.
22:06:32 <greensnark> I wasn't serious about writhing masses of spiny worms :)
22:06:55 <greensnark> They're fully paid-up Lair citizens anyway, just vanishingly rare
22:12:21 <ogaz> due: one thing that didn't come up in ##crawl are the dragon/ice dragon vaults that can show up in lair
22:14:44 -!- purge has quit [Ping timeout: 240 seconds]
22:25:25 <Ashenzari> Shouldn't Death Channel be extensible? (https://crawl.develz.org/mantis/view.php?id=1802) by pixnaps
22:27:12 -!- eith has joined ##crawl-dev
22:31:45 <greensnark> Riiight
22:36:41 <CIA-41> 03greensnark * r78a66cebf91b 10/crawl-ref/source/dat/ (clua/lm_monst.lua des/branches/hells.des): [1375] Fix statues in dis_mu coming to life even after being destroyed.
22:45:32 <Ashenzari> phase shift missing from status display (https://crawl.develz.org/mantis/view.php?id=1803) by pixnaps
22:49:00 <due> ogaz: Yes, aren't they awesmoe?
22:49:38 <ogaz> due: wasn't a criticism by any means, just another monster that could fit "toughest enemy to show up on Lair:8
22:49:50 <due> :D
22:49:51 <due> True!
22:50:02 <due> Easy worms, then suddenly! BAM! A DRAGNO.
22:50:37 <ogaz> due: also, instead of basing acid blobs off of slime entrance vaults, could they be based off jelly pits?
22:51:05 <due> Well, yeah.
22:51:13 <due> But I like the idea of having a Lair:8 slime city entrance.
22:51:20 <due> Because dpeg won't letme do a slime city Slime:6.
22:51:33 <ogaz> alright, fair enough
22:51:54 <sorear> spiny worms need to replicate
22:52:01 <CIA-41> 03greensnark * r88e245f4198d 10/crawl-ref/source/ (main.cc stairs.cc terrain.cc terrain.h): [1339] Fix wasted turn when trying to go downstairs on vestibule-of-hell exit.
22:52:10 <sorear> and make ominous popping and burping noises in the process
22:53:13 <greensnark> And when they reach critical mass the level explodes
22:54:42 <sorear> No, it just causes monster generation to slow down
22:55:05 <sorear> Base 40% per turn monster gen chance -1% per monster -1% per 1000 env.turns_on_level
22:55:32 -!- Vandal has quit [Read error: Connection reset by peer]
22:56:33 -!- Vandal has joined ##crawl-dev
22:58:28 * due eats greensnark.
22:59:43 -!- Niccus has left ##crawl-dev
23:04:50 -!- neunon has quit [Quit: neunon]
23:04:57 <CIA-41> 03greensnark * r433a2ef94521 10/crawl-ref/source/monster.cc: [1780] Fix missing drowning messages when a monster falls into deep water.
23:14:03 -!- pointless_ has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539]]
23:54:20 -!- Enne has quit [Quit: Enne]