00:01:35 -!- Henzell has quit [Ping timeout: 258 seconds]
00:02:19 -!- Henzell has joined ##crawl-dev
00:39:55 -!- ogaz has quit [Ping timeout: 248 seconds]
00:40:54 -!- Enne has quit [Quit: zzz]
00:49:22 -!- ortoslon has joined ##crawl-dev
01:11:40 -!- ogaz has joined ##crawl-dev
01:14:28 -!- ortoslon has quit [Read error: Connection reset by peer]
02:01:07 -!- ortoslon has joined ##crawl-dev
02:29:07 -!- by has joined ##crawl-dev
02:32:46 -!- morik_ has quit [*.net *.split]
02:32:47 -!- rax has quit [*.net *.split]
02:36:03 -!- by has quit [Read error: No route to host]
02:38:55 -!- morik_ has joined ##crawl-dev
02:38:55 -!- rax has joined ##crawl-dev
02:49:05 -!- ais523 has joined ##crawl-dev
02:53:56 -!- syllogism has joined ##crawl-dev
03:01:42 -!- Amonchakad has joined ##crawl-dev
03:09:36 -!- ais523 has quit [Remote host closed the connection]
03:10:45 -!- ogaz has quit [Ping timeout: 240 seconds]
03:13:41 -!- ogaz has joined ##crawl-dev
03:14:26 -!- ortoslon has quit [Read error: Connection reset by peer]
03:49:00 <Ashenzari> Hell effect explosion causes allied angels to turn hostile (https://crawl.develz.org/mantis/view.php?id=1671) by Tenaya
03:49:00 <Ashenzari> When cockroaches are confused they have trouble with 8 legs, but they only have 6 (https://crawl.develz.org/mantis/view.php?id=1670) by Tenaya
04:10:13 -!- ogaz has quit [Remote host closed the connection]
04:19:22 -!- xale has joined ##crawl-dev
04:22:30 -!- xale has quit [Client Quit]
04:22:53 -!- xale has joined ##crawl-dev
04:33:06 -!- Niccus has quit [Quit: out]
04:40:29 -!- Textmode has joined ##crawl-dev
04:41:04 <Textmode> morning all
05:08:57 <ixtli> Textmode: hey :D
05:09:11 <Textmode> meep!
05:09:15 * Textmode huggles ixtli
05:09:26 <ixtli> yay!
05:09:30 <ixtli> how have you been?
05:09:37 <ixtli> havent gotten a chance to work on ipad port in a few days
05:13:23 -!- by has joined ##crawl-dev
05:13:33 <by> hello
05:23:26 <Textmode> excuses
05:23:31 * Textmode huggles by
05:28:57 <by> hi textmode
05:29:12 <by> btw, there's a couple of bugs open in mantis re the spell colouring
05:58:28 -!- casmith789 has quit [Remote host closed the connection]
06:00:26 <Textmode> yay, as iff I didn't have enough thigs to do :/
06:00:41 <Textmode> I guess I'll take a look later.
06:26:58 <xale> by: i finished the save/load code for mrefs.
06:27:05 <xale> by: it's in /tmp/mrefs-save+load/
06:38:41 <Napkin> wow, google "Dungeon Crawl" finally has our website as first result \o/
06:40:59 <Textmode> cool, how long did that take?
06:41:32 <Napkin> no clue - just checked again today
06:41:39 <Napkin> 550k hits last month, btw
06:42:17 <by> xale: will take a look in a moment
06:42:47 <by> does anyone here know how stack_iterators interact with item destruction?
06:43:26 <by> from the comment in stuff.h, it looks like calling destroy_item(si.link()) may be a bad idea
06:45:12 <by> on the other hand, ely weapon sacrifice appears to do just that, so let's hope it's ok
06:54:46 <due> Napkin: yay!
06:55:50 <CIA-39> 03dolorous * r7cdeb113759a 10/crawl-ref/source/godprayer.cc: Fix Mantis 1662: Kiku shouldn't shield from torment while pain branding.
06:57:22 <Napkin> due :D
06:58:28 <due> by: that seems slightly dangerous, but oh well.
07:15:43 <CIA-39> 03N78291 * raf8ac9f24226 10/crawl-ref/source/effects.cc: Jiyva worshippers are not affected by slime walls.
07:15:43 <CIA-39> 03N78291 * r26eb56db2188 10/crawl-ref/source/ (ouch.cc player.cc): Conservation effect for Jiyva worshippers.
07:15:43 <CIA-39> 03N78291 * r5eadbac3d1c2 10/crawl-ref/source/ (godconduct.cc misc.cc): Allow attacking slime shaped shapeshifters with Jiyva.
07:15:43 <CIA-39> 03N78291 * r6295587c2b2c 10/crawl-ref/source/ (command.cc effects.cc godpassive.cc godpassive.h main.cc): Off level eating for Jiyva.
07:16:29 -!- Zaba has quit [Ping timeout: 240 seconds]
07:20:34 -!- Zaba has joined ##crawl-dev
07:32:33 <Ashenzari> Jiyva gives double resist corrosion and conservation (https://crawl.develz.org/mantis/view.php?id=1672) by rob
07:33:47 <Textmode> by: any idea how many issues I'm looking for?
07:38:33 <Textmode> okay...I can't log in, it says I'm not typig in a username (er, pretty sure I am...)
07:40:24 <by> I've tagged issues I ran across with "spell colouring"
07:40:39 -!- ais523 has joined ##crawl-dev
07:43:15 <Textmode> okay.
07:47:28 <by> xale: with these patches, I guess we just rely on all monsters being restored together, right?
07:48:19 <Textmode> no idea what happened there, but I'm logged in now.
07:48:31 <by> one other thing I ran into: some monsters store other monsters as a property (eg, mon.props[ORIG_MONSTER_KEY] for returning from hog form)
07:48:44 <by> this should be safe re refcounting
07:49:25 <by> though I'm not sure -- are there problems when a monster instance is not actually in env.mons?
07:51:35 <xale> there was a problem with fake draconians, but i fixed that too
07:52:23 <by> (it could also have been an option to just store the source monsters as mon.props[SOURCE_MONSTER_KEY], but it appears that CrawlStoreValue doesn't handle ghost_demon monsters properly)
07:55:13 <xale> what should i do to test hogs?
07:55:47 <xale> or anything else that you think might be a problem
07:56:38 <by> could just commit it and see if problems show up
07:56:59 <xale> i think it's good, but obviously there still might be bugs
07:57:24 <xale> if you find something broken let me know
07:58:23 <xale> email me or whatever.
07:59:30 <by> yeah, I think it's fine -- non menv-monsters will never be retained/released anyway
07:59:46 <xale> yes, they shouldn't be
08:00:28 <xale> there was one instance of that when monsters' spells are initialized, but that's gone
08:02:59 <by> will it mess things up for you if I rebase some of the patches together a bit?
08:03:16 <xale> mess up how? probably not.
08:10:13 <by> a normal living monster will have refcount == 1, right?
08:10:18 <xale> yes
08:10:42 <by> so setting it to 1 on restore for old save-games should be fine
08:11:16 <xale> there might be problems with bolts/orbs in the old games
08:11:49 <xale> although they won't be connected to any monsters there, so maybe not.
08:11:55 <xale> not sure.
08:12:04 <by> yes, I don't think orbs should be a problem, and bolts aren't saved as far as I know
08:12:15 <by> also, it's unlikely there'll actually be a save game with a living orb
08:12:25 <xale> true
08:40:51 -!- ortoslon has joined ##crawl-dev
08:44:16 <CIA-39> 03dolorous * rfeee1cfcd9f7 10/crawl-ref/docs/changelog.txt: Update the Jiyva entries in the 0.7 changelog.
08:49:07 <CIA-39> 03dolorous * r989e27ea284d 10/crawl-ref/source/delay.cc: Add spacing fix.
08:49:33 <by> xale: the mgrd at (%d,%d) points at dead monster %s warning should go, right?
08:50:41 <xale> dead monsters are removed from the grid
08:51:04 <by> ah, then there's a bug with save/loading
08:51:24 <xale> what bug?
08:51:29 <by> restoring dead monsters probably links them back into the grid
08:51:49 <xale> let me try that
08:53:13 <by> probably just check alive() instead of type in tag_read_level_monsters
08:53:39 <xale> yes
08:54:05 <xale> i mean, yes it does give a warning
08:56:08 <xale> yeah, I think that'll do it, check whether it's alive
08:57:17 <xale> you'll fix this, or should i add a patch?
08:57:22 <by> I'll fix it
09:08:53 -!- by has quit [Ping timeout: 260 seconds]
09:10:33 -!- by has joined ##crawl-dev
09:11:35 -!- Cryp71c has joined ##crawl-dev
09:12:32 <Cryp71c> Rar
09:14:25 <CIA-39> 03code * rbbc7f4b11b7e 10/crawl-ref/source/ (dbg-scan.cc mon-place.cc mon-stuff.cc monster.cc monster.h): refcount menv entries
09:14:26 <CIA-39> 03code * r7c1760d1b82f 10/crawl-ref/source/mon-cast.cc: monsters that try to cast spells must have a valid mindex
09:14:28 <CIA-39> 03code * r70e85ace771b 10/crawl-ref/source/ (makefile.obj mon-index.cc mon-index.h): add auto_mindex to simplify management of menv indices
09:14:34 <CIA-39> 03code * r8ee523426f4b 10/crawl-ref/source/ (tag-version.h tags.cc): save/load bits for the refcount and source
09:14:34 <CIA-39> 03code * r3584ca8eae3f 10/crawl-ref/source/ (mon-place.cc monster.cc monster.h): keep track of a monster's source (i.e. what monster caused it appear)
09:14:35 <CIA-39> 03by * r2a84601031d7 10/crawl-ref/source/tags.cc: Don't place dead monsters on restore.
09:14:37 <CIA-39> 03code * rf43d924016a1 10/crawl-ref/source/beam.h: make bolts track their sources
09:14:38 <CIA-39> 03code * r95e0957e0dd3 10/crawl-ref/source/mon-project.cc: connect iood bolt with its source
09:14:39 <CIA-39> 03code * rd0a27c8086ea 10/crawl-ref/source/hiscores.cc: dead monsters can still kill you
09:14:41 <CIA-39> 03code * rd27df762a37e 10/crawl-ref/source/beam.cc: properly attribute deaths by beams
09:15:30 <Cryp71c> ahhhhh, crazy commits.
09:18:00 <xale> great
09:18:12 <by> yes, thanks!
09:18:17 <xale> you're welcome!
09:18:40 <xale> i think there was a mantis issue for the iood
09:19:29 <Adeon> what, you code here?
09:23:22 <xale> by: https://crawl.develz.org/mantis/view.php?id=1627
09:24:44 <by> the morgue and death message are good now, but there's still bugginess with the score display
09:25:45 <xale> alright so i'll take a look at other death messages then
09:26:03 <xale> also should clouds be fixed too, or do we not consider them broken?
09:26:12 <by> I think I've got the disintegration one
09:26:21 <by> what are cloud messages like?
09:26:34 <xale> they're okay, but they don't say what created a cloud
09:26:47 <xale> so it's the same whether you killed yourself or something else did
09:35:14 <by> if you'd like to do that, you're welcome
09:35:28 <xale> i dunno, perhaps there is something more important?
09:35:43 <xale> it should be trivial to connect clouds to a monster now.
09:36:44 <xale> re: "a an disintegration bolt" and such. i think aux (or whatever it's called) should include an article.
09:37:35 <xale> or maybe they shouldn't, but it must be consistent.
09:43:53 -!- ogaz has joined ##crawl-dev
09:47:34 -!- eith has joined ##crawl-dev
09:55:48 <xale> by: interesting, a wand of disintegration and the orb result in the same ktyp
09:56:29 <by> yes
09:58:31 <xale> perhaps the 'killer' field should be empty for self-inflicted deaths?
09:58:51 <xale> that way we can disambiguate without changing much in the score file
10:10:04 <by> I was going for "you", which is apparently also used for poison
10:18:24 <Ashenzari> can't inscribe artefact with a long description (https://crawl.develz.org/mantis/view.php?id=1674) by olo
10:18:24 <Ashenzari> Swamp crash (something to do with tiledef-feat.cc) (https://crawl.develz.org/mantis/view.php?id=1673) by pixnaps
10:23:52 <CIA-39> 03dolorous * r077f3445237b 10/crawl-ref/source/ (delay.cc delay.h): Make Recite effects available in a public function.
10:23:53 <CIA-39> 03dolorous * r8dd0b671f30b 10/crawl-ref/source/ (abl-show.cc delay.cc delay.h): Reorder the Recite parameters so that it can calculate its own power.
10:23:54 <CIA-39> 03dolorous * r794cbe6a3c47 10/crawl-ref/source/monster.cc: Add Recite effects to imprisoned monsters just after the tomb drops.
10:23:57 <CIA-39> 03dolorous * r8c8afdd824d7 10/crawl-ref/source/ (delay.cc delay.h godabil.cc godabil.h): Move non-delay-related Recite functions from delay.cc to godabil.cc.
10:23:58 <CIA-39> 03dolorous * r4220d1785e7b 10/crawl-ref/source/ (abl-show.cc delay.cc godabil.cc godabil.h monster.cc): Rename various Recite-related functions to explicitly mention Zin.
10:23:59 <CIA-39> 03dolorous * r9643666ea079 10/crawl-ref/source/ (godabil.cc godabil.h): Reorder functions in godabil.cc yet again.
10:24:01 <CIA-39> 03dolorous * rda3266f2ed8a 10/crawl-ref/source/godabil.cc: Add spacing fix.
10:26:46 -!- Enne has joined ##crawl-dev
10:33:54 <xale> by: is there any place i can get a score file with various deaths?
10:34:52 <xale> i think there is a simple fix for both the wand and the orb, but i'm not sure i'm not breaking something else.
10:40:49 <CIA-39> 03dolorous * r6f8dd14f3143 10/crawl-ref/source/delay.cc: Revert erroneous "recite" delay name change.
10:41:52 -!- ais523 has quit [Remote host closed the connection]
10:53:31 <CIA-39> 03by * r772a6efee446 10/crawl-ref/source/hiscores.cc: Improved disintegration death source handling.
10:53:33 <CIA-39> 03by * r1027d1581da9 10/crawl-ref/source/hiscores.cc: Don't convert death_source to monster_type in scorefile_entry.
10:53:37 <CIA-39> 03by * r508f2e281442 10/crawl-ref/source/item_use.cc: Set bolt::aux_source for player-fired wands.
10:53:38 <CIA-39> 03by * r04404a941aa8 10/crawl-ref/source/beam.cc: Reorder cases in bolt::internal_ouch, improving IOOD attribution.
10:54:11 <by> xale: oh, I thought I'd said I was working on this one
10:55:50 <xale> maybe
10:56:25 <by> now to check if I broke mara clone attribution..
10:57:14 -!- ortoslon has quit [Quit: bye]
11:02:41 <CIA-39> 03by * r172df9011ff4 10/crawl-ref/source/initfile.cc: 8 color
11:02:42 <CIA-39> 03by * r2ae7433a0a26 10/crawl-ref/source/ouch.cc: scoring
11:02:43 <CIA-39> 03by * r0c6af8be9e6d 10/crawl-ref/source/hiscores.cc: Restore kill attribution for illusions of Mara.
11:02:47 <by> bah
11:04:02 <CIA-39> 03by * rbd5cb1ba89f6 10/crawl-ref/source/ (initfile.cc ouch.cc): Revert "scoring" and "8 color", accidentally committed.
11:05:04 <dpeg> by: what is "scoring" about?
11:05:30 <by> just temporary commits for easier testing
11:05:37 <dpeg> ah
11:05:52 <dpeg> by: thought you might have secret tweaks to the scoring formula :)
11:06:36 <xale> by: yeah, the orb and the wands work now
11:06:50 <xale> by: so what other messages are broken?
11:07:14 <by> I'm not aware of any
11:08:56 -!- Mu_ has joined ##crawl-dev
11:10:48 -!- DrPraetor has quit [Ping timeout: 260 seconds]
11:18:48 <xale> by: https://crawl.develz.org/mantis/view.php?id=1068 this doesn't seem to happen anymore as well
11:19:32 <by> nice
11:28:45 <Ashenzari> Breaking out of a full tomb prematurely should make it disappear early (https://crawl.develz.org/mantis/view.php?id=1675) by dolorous
11:31:54 <xale> Enne: do you happen to know why show_inventory_weights does not work with tile_menu_icons?
11:33:29 -!- DrPraetor has joined ##crawl-dev
11:38:52 -!- 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]
11:46:16 -!- ortoslon has joined ##crawl-dev
11:54:56 <dpeg> by: powered by pain is a great idea!
11:58:41 <Textmode> dpeg: ?
12:02:28 <dpeg> Textmode: ??
12:02:35 <Textmode> ??!
12:02:35 <Henzell> I don't have a page labeled ! in my learndb.
12:02:42 * Textmode huggles Henzell
12:03:41 <dpeg> Textmode: by proposed a new DS mutation, "powered by pain".
12:04:25 <Textmode> ah
12:04:43 <elly> powered... BY PAIN
12:05:02 <elly> I like it
12:06:24 -!- ais523 has joined ##crawl-dev
12:07:05 <dpeg> not to be confused with "powered by propane"
12:07:25 * elly is powered by IRC.
12:14:24 <Cryp71c> lol
12:18:24 <xale> anybody wants to help me with the tiles?
12:19:12 <by> don't know too much about tiles, but go ahead and ask
12:19:15 <xale> it looks like the menu entries for inventory items with weight are formatted using a bunch of spaces.
12:19:28 <xale> presumable that doesn't work very well with a graphical display?
12:19:47 <by> I think the issue is that the tiles inventory uses a different two-column menu
12:20:19 <by> felirx' new menu class would probably be up to unifying the two -- not sure if anything's planned there
12:22:00 <by> it may just be that with Options.tile_menu_icons, there's not enough space
12:22:27 <xale> ' ' <repeats 107 times>, "(12.0 aum)"
12:22:37 <xale> there would be enough space if it didn't repeat 107 times! :-)
12:23:24 <xale> my guess is that spaces are not a very good layout method unless you're using a fixed-width font.
12:25:11 <felirx> slowly my plan is to convert everything to it
12:25:34 <felirx> that was a weird sentence structure
12:26:03 <CIA-39> 03dolorous * r8edd3fa6f1fb 10/crawl-ref/source/main.cc: Add spacing fix.
12:26:13 <dpeg> dolorous to the rescue :)
12:26:20 <felirx> shops is next in my plan after the tutorial tweaks and some fix to the code repetition
12:26:37 <dpeg> felirx: would you also be willing to improve the shop interface?
12:26:42 <dpeg> I have some ideas...
12:26:43 <xale> convert everything to what?
12:26:50 <dpeg> xale: Christianity
12:26:52 <felirx> it will be mouse operable and so on when I'm done
12:27:11 <felirx> haven't started that though
12:27:54 <dpeg> felirx: I know. My ideas are about something else: the shopping and viewing modes should be better distinguished, for example.
12:28:40 <felirx> probably the best bet is to write those down to wiki
12:31:21 <by> xale: convert to menu.h:PrecisionMenu, I think
12:32:02 <dpeg> felirx: they are, of course.
12:33:06 <CIA-39> 03by * r58d53562f6d5 10/crawl-ref/source/ (show.cc terrain.cc terrain.h travel.cc): Slime wall polish: prevent travel, colour adjacent floor.
12:37:38 <xale> so what's the best way right now to make weights work?
12:37:57 <dpeg> by: thank you!
12:38:41 <by> it's not a particularly nice solution, but I hope it'll do for now
12:40:06 <xale> is crawl supposed to support proportional fonts?
12:40:18 <xale> because if it isn't we could just use a different padding.
12:56:01 <by> not completely sure about the tiles menus, but in general crawl assumes a fixed width font
12:56:21 <by> i.e., even in tiles mode, you can cgotoxy and cprintf
12:58:39 -!- sorear has quit [Read error: Connection reset by peer]
12:58:42 <felirx> tiles fonts have the same width and height per unit
12:58:50 <felirx> or per letter
13:04:17 -!- sorear has joined ##crawl-dev
13:09:50 <xale> and the shop menu also uses spaces to align the prices?
13:10:13 <felirx> the shop menu uses no menu structure at all
13:10:19 <felirx> it has it's own system of something
13:10:33 <xale> oh, i see
13:11:48 <xale> maybe menu entries need another text variable that will be right aligned?
13:11:51 <xale> or would that be too ad hoc
13:12:28 <xale> it could be used in shops too.
13:22:00 -!- casmith789 has joined ##crawl-dev
13:38:04 -!- Enne has quit [Ping timeout: 258 seconds]
13:42:27 <CIA-39> 03by * rec76b6c0e3cb 10/crawl-ref/source/dbg-asrt.cc: Mark milestones for crashes. (#1545)
13:44:52 -!- Enne has joined ##crawl-dev
13:48:25 <by> Master branch on CDO updated to: 0.7.0-a0-2129-gec76b6c (24.2)
13:57:06 <by> unfortunately I don't know a reliable way to crash console crawl currently, to test whether the crash milestones work
13:57:41 <Cryp71c> by, I've had experiences where mashing ` (command repeat) really quickly will crash.
13:57:51 <by> but recently?
13:57:52 <Cryp71c> but that was a couple weeks ago and may have been fixed / not be as reliable as I think.
13:58:44 <xale> you could add a deliberate error somewhere for testing.
13:59:01 <by> yes, but I think I won't :) there'll be crashes yet
14:09:40 <Ashenzari> Mindless is a little extreme, for allies. (https://crawl.develz.org/mantis/view.php?id=1676) by Kyrris
14:10:39 <xale> Enne: aroundp?
14:13:38 -!- by has quit [Quit: leaving]
14:13:44 <Enne> Sort of?
14:13:48 <Enne> I'm painting my bedroom.  ;)
14:19:23 <xale> i forgot what i was going to ask
14:24:45 -!- dpeg has quit [Quit: away]
14:26:26 <Enne> Glad to help, xale.  :)
14:29:33 <felirx> mmmm
14:30:02 <felirx> How should I get upstairs command "<" from the tutorial.des
14:30:10 <felirx> since < breaks </white>
14:31:37 -!- MarvinPA has joined ##crawl-dev
14:35:17 -!- DrPraetor has quit [Ping timeout: 240 seconds]
14:45:12 -!- Enne has quit [Read error: No route to host]
14:47:00 -!- by has joined ##crawl-dev
14:47:01 -!- Enne has joined ##crawl-dev
14:48:20 <by> felirx: <<
14:49:55 <Ashenzari> everything is gray outside LOS (https://crawl.develz.org/mantis/view.php?id=1677) by rob
14:50:21 <felirx> I know that
14:50:22 -!- Niccus has joined ##crawl-dev
14:50:32 <felirx> it's just that crawl.get_command() outputs single < :D
14:50:51 <felirx> since that's what the bound command button is
14:58:30 -!- by has quit [Ping timeout: 276 seconds]
15:03:52 -!- by has joined ##crawl-dev
15:04:59 <xale> yay, it almost works
15:10:56 <felirx> err what, my local build runs by default
15:11:02 <felirx> I have to hold shift to walk
15:11:07 <felirx> with both caps on and off
15:11:19 <felirx> or the repetition is just out of whack
15:11:24 <felirx> win console
15:12:38 <felirx> turning numpad on and off fixed that, weird
15:17:51 <CIA-39> 03felirx * r03a1f7ccf046 10/crawl-ref/source/dat/des/tutorial/tutorial.des: Rename Tutorial message functions
15:17:52 <CIA-39> 03felirx * r79c0c31103f3 10/crawl-ref/source/dat/des/tutorial/tutorial.des: Change tutorial_basic_2 commands to be tiles sensitive
15:31:46 <xale> Enne: is there any particular reason why menu entries that do not fit into a single line
15:32:00 <xale> Enne: are split into two lines, rather than, say, three or as many as necessary.
15:32:19 <felirx> the new Preciniosmenu::textitems do that
15:32:31 <xale> is this done already?
15:32:35 <felirx> split to as many lines as there is space available
15:32:47 <Enne> I think I just didn't want to solve the problem of having menus that displayed variable amounts of items on the screen, depending on their length.
15:32:49 <felirx> space reserved even
15:33:15 <xale> Enne: oh right, i haven't thought about that
15:33:58 <xale> felirx: are these new precision menus are done/commited or is it something you're working on?
15:34:06 <felirx> see menu.h/cc
15:34:14 <felirx> the startup menus use that code
15:34:20 <Enne> (It'd be awesome to handle that properly, but it was much easier to just use a fixed number of items per page at the time.)
15:34:22 <xale> hmm
15:36:54 <xale> so the plan is to eventually replace all menus with precision menus?
15:37:22 <felirx> or inherited versions of it
15:45:07 <xale> and the reason it isn't is because these menus are not finished yet?
15:45:15 <xale> like they don't seem to support pagination?
15:47:06 <by> felirx implemented them to allow redoing the starting menus with mouse support, and I guess they just implement those functions that were required there
15:47:59 <felirx> technically the only thing that is really missing is paging support
15:48:34 <by> xale: we have a nice bug: monsters don't survive following stairs
15:48:54 <felirx> the original menu classes did not support free placement of different objects like the first  screen you see when you start the game
15:49:56 <felirx> instead it uses a vector like list approach which works nicely for things like inventory and such
15:50:08 <st_> is there no difference between z and Z now?
15:50:28 <by> st_: shouldn't have changed
15:51:29 <st_> no enemies in LOS, casting magic dart with z lets me target instead of "no enemies in view" but maybe it was always like that
15:51:38 <xale> by: looking at it
15:51:48 <by> mon-transit.{cc,h} is relevant
15:51:58 <by> thanks
15:52:10 <xale> felirx: okay, i think i'm going to finish hacking the old menus to support inventory weight even though it will eventually be scrapped
15:52:36 <xale> but at least now i understand why there are so many menu classes!
15:53:05 <felirx> eventualy I'll write more inherited classes from freeform or menuobject
15:53:07 <felirx> and menuitem
15:53:16 <felirx> like inventoryitem and shopitem
15:53:27 <xale> yeah, it makes sense
15:54:21 <by> st_: looks like a bug
15:54:31 <st_> should I file?
15:54:41 <by> no, I'll try to do it now
15:54:44 <by> thanks
15:58:12 <CIA-39> 03by * rdcba8ecbf264 10/crawl-ref/source/mon-util.cc: Fix no-hostile-in-range check for range 8 spells.
15:59:35 <pineapple> does anyone here understand how crawl handles activity on levels taht you're not currently on?
15:59:48 <by> sort of, yes
15:59:54 <pineapple> how?
15:59:55 <by> it simulates it when you return
16:00:44 <pineapple> you mean it does a burst of movement when you enter the level again, based on how long you were away for?
16:00:44 <by> effects.cc:update_level
16:03:44 <Cryp71c> by, major bug happening on CDO..sometimes monsters will disappear entirely a turn or two after they become aware of the player.
16:03:51 <xale> heh
16:03:56 <pineapple> oops?
16:03:56 <Cryp71c> by, doesnt happen all the time, but enough that it became conspicuous pretty quickly.
16:04:07 <by> read the scroll-back?
16:04:12 <Cryp71c> by, nothing there.
16:04:32 <by> xale is already working on it
16:04:34 <casmith789> about 16 mins ago
16:04:46 <Cryp71c> by, ah, k. Wanted to make sure it was mentioned.
16:09:57 -!- MarvinPA_ has joined ##crawl-dev
16:10:39 -!- MarvinPA has quit [Ping timeout: 276 seconds]
16:10:46 -!- MarvinPA_ is now known as MarvinPA
16:11:18 <xale> oh I see
16:12:29 <xale> copying a monster into the list of transiting monsters and back clears the refcount
16:16:09 <xale> also follower::place uses it's own method to find an unused menv entry.
16:16:26 <xale> s/'//
16:18:30 <by> monsters::source could be invalid after transit, so probably followers should get refcount=1, and follower sources should be reset
16:19:25 <by> (or we could try making a dead copy of the source follow as well, but that sounds like overkill)
16:20:52 <by> can you take an orb downstairs?
16:20:55 <xale> i had a dream where source could be used to track summon chains
16:21:05 <by> :)
16:23:35 <by> maybe we should fix things better; would it be realistic to give every monster a unique id?
16:24:14 <by> env.mons could then be a map<monster_id,monsters>
16:29:35 <xale> by: http://xale.me/tmp/0001-properly-handle-followers.patch
16:29:39 <xale> this seems to fix it for me
16:33:01 <xale> however, this is not exactly what cryptic said is happening?
16:33:39 <xale> there is a bug with the followers, but perhaps there is something else?
16:34:09 -!- ortoslon has quit [Quit: bye]
16:35:06 <by> ah, the "aware"?
16:35:22 <xale> yes
16:36:11 <Enne> by: I'm not sure that a map solves the problem.  You'd have exactly the same incorrect refcount issues of "deleted too early" or "left around too long" in a map as you would having an array with the index as the monster id.
16:36:44 <Enne> At best, the "delete too early" case for a map would just result in no monster found rather than an incorrect monster found, but that's still wrong behavior.
16:37:35 <by> I was thinking about the issue of source getting invalid when taking a monster downstairs, there
16:38:13 <xale> i actually thought that env.mons is just written on disk as is
16:38:20 <xale> but since it isn't, it can actually be anything
16:38:36 <st_> is it intentional you can't use reaching weapons diagonally?
16:41:04 <Enne> by: Oh, that's an excellent point.
16:43:32 <by> Cryp71c: did monsters disappear for you unrelated to taking stairs?
16:44:24 <st_> I had a ghost disappear the turn it noticed me
16:44:24 <casmith789> by: I've just been trying to abuse it. If you go down/up stairs with monsters, they disappear the next turn, but I think one at a time
16:45:17 <Cryp71c> by, stairs were no where near
16:46:04 <Cryp71c> by, just a second ago I had a hound in melee range, hit him once, he disappeared, no messages, nothing.
16:46:14 <Cryp71c> by, also...I can't be certain, but I think I might've gotten xp for it.
16:46:26 <xale> haha
16:46:46 <xale> by: is there any reason not to keep all monsters in memory all the time?
16:47:05 <by> memory usage would seem to be a reason
16:47:17 <st_> well I can tell you that I got no milestone for defeating the ghost that disappeared so I'm not so sure
16:47:41 <by> I definitely got no xp or milestone for azrael, which is how I noticed the stairs bug
16:47:50 <Cryp71c> st_, same for Prince Ribbit, but I kept noticing my xp was quite high for having kill nothing else of notoriety
16:48:57 <Enne> We could keep the refcounts separate from the monsters, and keep all of those in memory at once.
16:50:22 -!- MarvinPA has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539]]
16:50:28 <Cryp71c> by, BirdofPrey is having every monster on his current level disappear.
16:50:33 <Cryp71c> You might spec him to see.
16:50:38 <Cryp71c> Maybe you'll notice something of significance?
16:52:26 <CIA-39> 03by * r50eadc8f20bc 10/crawl-ref/source/wiz-mon.cc: Fix bug in wizard_spawn_control, restoring accidentally deleted statement.
16:52:36 <CIA-39> 03code * r80a6a7e5be19 10/crawl-ref/source/ (mon-place.cc mon-place.h mon-transit.cc): properly handle followers
16:53:11 <xale> oh shafts too
16:54:34 <xale> what's the 'space' for?
16:56:02 <xale> i can't get anything to disappear.
16:56:30 -!- Luca__ has quit [Remote host closed the connection]
16:56:46 -!- Luca__ has joined ##crawl-dev
16:58:08 <by> xale: that's what it was like before, if I understand correctly
16:58:31 <by> it's probably so that you can't take hordes downstairs and prevent hostile monsters from summoning or something
16:59:56 <by> 24cef5c20d doesn't really shed light on the issue
17:00:45 -!- Enne_ has joined ##crawl-dev
17:02:16 <by> it's greensnark's code, so we can ask him some time
17:02:25 <xale> you mean it was removed some time ago?
17:02:39 -!- Enne has quit [Ping timeout: 276 seconds]
17:02:39 -!- Enne_ is now known as Enne
17:03:29 <by> I'm not sure I understand your question
17:05:04 <xale> oh, you moved the - 5 from follower::place to get_free_monster() i see now
17:06:59 <xale> by: did you get anything else to disappear?
17:07:07 <by> no, unsuccessful so far
17:07:20 -!- Cryp71c has quit [Quit: rar!]
17:07:42 <by> I wonder if we can somehow catch monsters getting reset accidentally
17:08:10 <by> usually, they should have hit_points <= 0 when not saving/restoring, right?
17:09:13 <xale> hm? monsters usually have hp > 0 and 0 only when dead or reset.
17:10:15 <by> I'd like to put some warning message or ASSERT in monsters::release or monsters::reset that checks that only dead monsters get reset
17:11:45 <xale> yeah that's a good idea
17:12:40 <xale> you assert that refcount is strictly positive in release and non-negative in retain.
17:12:43 <xale> s/you/also/
17:13:19 <xale> by: but monsters get reset when unloading a level, don't they?
17:13:33 <by> yes, so it's not that easy
17:14:11 <xale> maybe it's better to rollback cdo so that people can play?
17:14:22 <xale> and there is less hurry?
17:16:33 <xale> <nht> this bug is not very reliable
17:24:20 -!- MarvinPA has joined ##crawl-dev
17:29:48 <xale> by: shouldn't the version be checked when loading loading a monster?
17:30:21 <xale> by: since we decided to allow older save games.
17:30:59 <xale> oh, it is checked already, never mind.
17:37:36 <st_> !lg
17:37:39 <Henzell> 1934. clouded the Conjurer (L12 HuWz), killed by wand of fire in a Labyrinth on 2010-06-02, with 11641 points after 22420 turns and 1:20:17.
17:38:10 <st_> is this a buggy message
17:38:35 <by> !lg st_ x=ktyp
17:38:36 <Henzell> 1934. [ktyp=wild magic] clouded the Conjurer (L12 HuWz), killed by wand of fire in a Labyrinth on 2010-06-02, with 11641 points after 22420 turns and 1:20:17.
17:38:55 <by> !lg st_ x=killer
17:38:55 <Henzell> 1934. [killer=] clouded the Conjurer (L12 HuWz), killed by wand of fire in a Labyrinth on 2010-06-02, with 11641 points after 22420 turns and 1:20:17.
17:39:03 <by> how were you killed?
17:39:03 <st_> I shot myself with it
17:39:18 <by> ah, that used to work better...
17:42:12 <by> !lg * ktyp=wild magic s=kaux
17:42:18 <Henzell> 1511 games for * (ktyp=wild magic): 152x scroll of immolation, 150x reading a scroll of immolation, 145x a spatial distortion, 141x spell miscasting, 132x a distortion effect, 121x the fury of Makhleb, 95x , 66x the capriciousness of Xom, 47x distortion unwield, 43x the severe capriciousness of Xom, 39x your miscasting Sticky Flame, 36x a magical explosion, 26x your miscasting Mephitic Cloud, 17x ...
17:43:56 <by> bolt::internal_ouch decides whether or not it's "wild magic" by checking aux_source.empty()
17:44:18 <by> and I set aux_source for wands to make the disintegration message look nicers
17:50:01 <xale> handle_monster_move creates a beam
17:50:03 <xale> for no good reason
17:51:48 <xale> this is how monsters that followed you disappeared.
17:53:40 <by> managed to get a monster with refcount -1 now
17:53:52 <xale> after all the patches?
17:54:11 <xale> what did you do?
17:54:46 <by> I have an extra patch here, that might have caused it
18:10:26 -!- Vandal has quit [Read error: Connection reset by peer]
18:19:29 <by> _jelly_divide causes refcount inconsistencies, though that's unlikely to be what people are seeing
18:22:55 <xale> http://sprunge.us/WgfD that might help, should've done it in the first place.
18:27:04 <xale> yay it fails.
18:27:46 <CIA-39> 03by * rf5b6923f5738 10/crawl-ref/source/dbg-scan.cc: Improved refcount scanning.
18:27:48 <CIA-39> 03by * r851ced10f6c2 10/crawl-ref/source/spl-mis.cc: Fix god miscasts setting bad beam_source.
18:27:52 <CIA-39> 03by * ra61e760fa660 10/crawl-ref/source/monster.h: Refcount asserts (xale).
18:27:56 <by> xale: in what cases?
18:28:07 <xale> taking summons downstairs
18:29:27 <by> I'll update cdo -- that way I can test the milestone code :)
18:29:54 -!- ogaz has quit [Ping timeout: 240 seconds]
18:33:01 <by> Master branch on CDO updated to: 0.7.0-a0-2137-ga61e760 (24.2)
18:35:08 <xale> when summoner is reset before its minions their source becomes invalid
18:35:19 <xale> and when they disappear they can remove something that was placed there.
18:35:32 <xale> wait.
18:35:36 <xale> that cannot happen!
18:36:13 -!- syllogism has quit []
18:38:15 <by> got a crash milestone, yay
18:38:37 <xale> by: so the monsters are moved to the transit list and then menv is fully reset?
18:41:44 <xale> so they get released twice, once in add_monster_to_transit(), then in reset()
18:42:11 <by> files.cc:_grab_follower_at seems to be where they go
18:42:52 <by> which calls add_monster_to_transit, and the monster_cleanup
18:47:06 <xale> also, if you take summoner downstairs but leave its things upstairs
18:47:32 <xale> and then come back, the summoner is not going to be placed into the same entry.
18:47:50 <xale> i think we need to remove all this logic and use it only for the orb, for now
18:48:53 <xale> i haven't thought about movement between the levels at all when writing this.
18:49:24 <xale> but.. but, you can make boris conjure the orb and then take him downstairs.
18:51:11 <Ashenzari> Give Nemelex an on-death message (https://crawl.develz.org/mantis/view.php?id=1678) by OG17
18:52:06 <by> if the refcounting worked properly, shouldn't the orb keep a dead copy of boris around upstairs?
18:53:29 <xale> let me try that, maybe you'll end up with two borises
18:54:26 <by> the monster_cleanup call should make sure one of them is dead
18:54:36 <xale> why would it be dead?
18:56:07 <by> when you go downstairs, grab_follower_at places boris on the transit list, and calls monster_cleanup for the one in menv
18:58:15 <xale> then he's going to be saved to a file because the entry was not reset.
18:58:29 <xale> and you will have a boris downstairs too
19:00:08 <xale> i don't know how to setup things to make it actually happen, but i'm pretty sure something will break.
19:01:54 <xale> i think you're better off reverting these patches for now, or rolling back cdo
19:02:17 <by> hmm
19:04:36 <xale> i didn't think about interlevel travel at all, so it's back to the drawing board
19:04:47 <xale> and that might take several days.
19:04:55 <by> ok, I didn't either
19:12:26 -!- Textmode has quit [Quit: Over the ages, man has been certain of a great many things. some of these things were even true.]
19:14:21 -!- MarvinPA has quit [Quit: Page closed]
19:17:16 -!- Mu_ has quit [Quit: Defecator, may everything turn out okay so that you can leave this place.]
19:18:47 <xale> by: so what do you think?
19:18:49 <CIA-39> 03neunon * r52b2c4e64d09 10/crawl-ref/source/makefile: makefile: use 'strip -x' on Mac OS X
19:20:36 <by> I'm just working on the reverts
19:21:17 -!- MarvinPA has joined ##crawl-dev
19:32:37 <CIA-39> 03by * r2de784baf71d 10/crawl-ref/source/ (14 files): Revert monster refcounting.
19:32:38 <CIA-39> 03by * r30a71f85be6d 10/crawl-ref/source/ (tag-version.h tags.cc): Update save code for reverted refcounting.
19:32:39 <CIA-39> 03by * rdaac3443b800 10/crawl-ref/source/ (beam.cc item_use.cc): Don't set aux_source for player-fired wands.
19:40:18 -!- casmith789 has quit [Remote host closed the connection]
19:40:48 <CIA-39> 03neunon * r0c7e4dd93a3e 10/crawl-ref/source/ouch.cc: ouch.cc: expose _wizard_restore_life to DEBUG builds
19:40:50 <CIA-39> 03neunon * r68b228d03db2 10/crawl-ref/source/makefile: makefile: don't check .cflags on 'make test'
19:42:20 -!- Vandal has joined ##crawl-dev
20:00:34 -!- by has quit [Quit: leaving]
20:18:00 -!- xale has quit [Quit: xale]
20:53:45 <CIA-39> 03neunon * r196361ac9894 10/crawl-ref/source/makefile: makefile: use 10.5 SDK on x86_64 non-TILES builds
20:53:45 <CIA-39> 03neunon * rb28d9d123ed7 10/crawl-ref/source/makefile: makefile: use 'uname -m' instead of 'arch' for finding host architecture
21:25:53 -!- ais523 has quit [*.net *.split]
21:25:53 -!- ivan has quit [*.net *.split]
21:25:53 -!- joosa has quit [*.net *.split]
21:28:15 -!- ais523 has joined ##crawl-dev
21:28:15 -!- ivan has joined ##crawl-dev
21:28:15 -!- joosa has joined ##crawl-dev
21:35:42 -!- Enne has quit [Quit: zzz]
21:38:52 -!- DrPraetor has joined ##crawl-dev
21:59:46 -!- ogaz has joined ##crawl-dev
22:24:06 <CIA-39> 03dolorous * r5e23b53ca954 10/crawl-ref/source/dat/database/randname.txt: Add another name for ponderous randarts.
22:26:34 -!- elly is now known as Qelly
22:27:02 -!- Qelly is now known as ellyQ
22:27:30 -!- ellyQ is now known as elly
22:37:24 -!- eith has quit [Ping timeout: 276 seconds]
23:02:30 <Ashenzari> Crash: ASSERT(in_bounds(src_pos) || src_pos.origin()) in 'mon-behv.cc' at line 716 failed. (https://crawl.develz.org/mantis/view.php?id=1679) by MarvintheParanoidAndroid
23:02:35 -!- ais523 has quit [Remote host closed the connection]