02:18:51 -!- Cheibriados has joined ##crawl-dev 02:18:51 -!- The topic of ##crawl-dev is: Crawl Development | Logs: http://s-z.org/crawl-dev/ | People with +v have commit access. | Please keep general Crawl-related chit-chat to ##crawl. | Dev wiki: http://crawl.develz.org/wiki | Long stuff to pastebin.ca please | Immortal Warwalrus and Crazy Yiuf forever. 02:19:09 pre-lair/pre-orc xom vault with ghost-moth capable shapeshifters (https://crawl.develz.org/mantis/view.php?id=5596) by Pacra 02:19:09 Overlapping lines in High Scores list (https://crawl.develz.org/mantis/view.php?id=5595) by spacewolf009 02:29:49 inept feature mimic (12X) | Speed: 10 | HD: 10 | Health: 39-73 | AC/EV: 5/1 | Damage: 14 | Flags: 11non-living | Res: 06magic(40), 12drown, 04rot+++, 13neg+++, 13torm | XP: 559. 02:29:49 <|amethyst> %??inept feature mimic hd:10 02:29:54 feature mimic (02X) | Speed: 10 | HD: 27 | Health: 126-171 | AC/EV: 5/1 | Damage: 31, 3108(poison), 3103(constrict)(crush) | Flags: 11non-living | Res: 06magic(108), 05fire, 02cold, 10elec, 03poison, 12drown, 04rot+++, 13neg+++, 13torm | XP: 6378. 02:29:54 <|amethyst> %??feature mimic hd:27 02:30:00 <|amethyst> yay 02:34:32 -!- Mottie has quit [Ping timeout: 256 seconds] 02:35:00 -!- Textmode has quit [Ping timeout: 255 seconds] 02:37:32 Is (crush) still accurate? AF_CRUSH was removed a while ago 02:47:23 -!- absolutego has joined ##crawl-dev 02:50:22 -!- Textmode has joined ##crawl-dev 02:55:04 evilmike: I think it's just flavour. All constricting attacks are AF_CRUSH 02:56:11 |amethyst: I can remove the assert that makes querying item mimic crash 02:56:32 a mimic without an item would just be called a mimic 02:56:59 they are quite stable now, so removing the assert should be fine 02:57:15 <|amethyst> hm 02:58:20 <|amethyst> I guess that works, though I'm not too worried about it, especially now that item and feature mimics have the same stats 03:02:43 feature mimic (16X) | Speed: 10 | HD: 8 | Health: 30-58 | AC/EV: 5/1 | Damage: 12, 1208(poison), 1203(constrict) | Flags: 11non-living | Res: 06magic(32), 05fire, 02cold, 10elec, 03poison, 12drown, 04rot+++, 13neg+++, 13torm | XP: 399. 03:02:43 <|amethyst> %?? feature mimic 03:02:50 <|amethyst> got rid of (crush) 03:03:49 -!- Textmode has quit [Ping timeout: 255 seconds] 03:05:08 looks like AF_CRUSH is in mon-enum.h and mon-data.h, but was removed from melee-attack.cc recently 03:07:16 maybe it has no more reason to exists with constriction 03:07:22 <|amethyst> it's always used with AT_CONSTRICT, and AT_CONSTRICT is the important one 03:09:58 <|amethyst> hm, what would AT_CONSTRICT, AF_FIRE do 03:12:23 <|amethyst> nothing interesting 03:12:24 <|amethyst> FR: 03:13:32 <|amethyst> actually, I'm not sure why constriction is an AT_ rather than an AF_ 03:14:05 <|amethyst> surely one could imagine an alligator that holds you in its jaws (AT_BITE, AF_CONSTRICT) 03:19:07 -!- shirish has quit [Remote host closed the connection] 03:29:15 |amethyst: good point. I'll consider this when I rewrite constriction :) 03:34:32 evilmike: AF_CRUSH was meant to be what constriction is now 03:34:47 with a different design, mostly like Nethack's owlbears 03:51:09 kilobyte: how is it different? Normal damage, not increased? 03:52:07 -!- capablanca has joined ##crawl-dev 03:52:23 -!- mikee_ has quit [Disconnected by services] 03:52:27 -!- capablanca is now known as mikee_ 03:53:59 galehar: AF_CRUSH never went beyond a message, which then has been removed as misleading 04:07:32 kilobyte: there's no way to translate names yet, right? 04:07:58 already have an idea of how you want to implement it? 04:08:34 yeah, although this will require the whole grammar machinery 04:09:20 there's a need for recursion 04:10:20 recursion requires recursion 04:11:15 a .pl example: "You hit XXX." -> "You hit $ACTOR(XXX, accusative)". "XXX zombie" -> "$ACTOR(XXX, genitive)" 04:11:43 tail recursion or head recursion? :) 04:11:45 not sure if French has noun cases, but I bet you'll have similar issues 04:12:18 bhaak: can be either 04:12:44 heck, even English has those: "quick blade of draining" but "vampiric quick blade" 04:13:01 i think that zombie mapping is not correct in the general sense 04:13:13 quick blade of vampiricism 04:13:22 draining quick blade 04:13:26 which probably should be abstracted to a grammar function of this kind, as other languages have different special cases 04:13:28 noun composition is a tricky business and it's not just using the genitive. is it in polish? 04:14:00 so we'd have some $item_brand("quick blade", SPWPN_VAMPIRICISM) 04:14:47 yes, that make sense. the specific language programmer can then decide if "vampiric quick blade" or "quick blade of vampirism" is better suited for this translation 04:15:23 it can't sadly be done by tables alone, there's a need for some actual code 04:15:34 oh yeah 04:15:51 only for a few special cases like zombies, though 04:16:32 especially with german noun composition. there are NO rules. it was regular, ~600 years (or even more) ago. but now, there are none rules left. 04:23:57 -!- alefury has joined ##crawl-dev 04:27:40 Item buggily placed in feature at (45, 35). 04:27:40 kilobyte: yeah, constructing full names would need some work. I was thinking of how to store base name translations 04:27:53 since translated description files have english names 04:29:55 my plan is to use 100% strings, no enums, for consistency, to allow using table lookups outright where no custom code is needed 04:30:16 keys would be English but often in a form with grammar functions 04:30:36 like that $item_brand(quick blade, vampiric) example 04:31:40 an irregular translation like English would need to check for special cases (vampiric and anti-magic here), a regular one can recurse via a table immediately 04:32:45 ie, the rule would be "$item_brand(%s, %s)" -> "$item(%s) of $brand(%s)" 04:34:28 Polish for example has a regular word order here but needs to apply noun genders 04:36:39 ie, it'd be "$brand(%2$s, $gender_of(%1$s)) $item(%1$s)" 04:37:12 but where/how to store the translated names? In dat/database/xx/name.txt? 04:37:22 heh, this clearly shows sprintf-like placeholders for %s are no good, as reordering them is totally unreadable 04:38:29 03elliptic * rfb959798448d 10/crawl-ref/source/beam.cc: Fix IMB explosion giving a message about exploding even if you can't see it. 04:39:14 I wouldn't use sqlite for that -- item names are often used in long loops, a thousand SQL queries isn't a good idea, a thousand hash accesses is hardly worse than an array 04:40:52 i don't know how fast sqlite is in that regard. in nethack-de i used a (althought statically allocated) array that gets search _linearly_ for each token that needs translation 04:41:19 so for a typical sentence 5 to 10 times over an array with ~20000 entries 04:41:58 i was surprised that even when i started several years this didn't put a noticeable lag onto the program 04:42:26 bhaak: is that only for final sentences, or for item names for a lookup? 04:42:58 because it'd be a square time operation in some cases 04:43:50 kilobyte: i have a bit a different approach, in most of the cases the final sentences gets tokenized and then each token gets translated 04:44:17 that's what I intend to do too, for final text 04:44:28 only rarely is stuff being translated before the sentence is sent to the output routune. 04:44:52 you can't look up descriptions by name, etc, though, right? 04:45:17 for the database entries? of course i can :) 04:45:31 otherwise wishing wouldn't be possible 04:46:03 there is a function that searches for a token like NOUN_MAGICBANE from the user input 04:46:29 the database entries just need to have the token as title 04:46:53 ie, it's impossible to look up something that's composite 04:47:13 yes 04:47:24 luckily that's not needed for nethack 04:48:10 why "luckily"? You don't need a linear lookup for strings, either. 04:48:21 for stuff like "Medusa's corpse" there is extra code to parse it 04:48:46 yes, i don't need linear lookup, it's completely unoptimized. 04:49:00 but it is still insanely fast. 04:49:39 O(1) lookup here adds no additional programming time, as C++ has pre-made hashes (ok, in C++90, only O(log n)) 04:50:37 "luckily" because this would be quite a headache to parse noun compositions back to tokens with the framework i've created. although if it were needed, i would have done it differently. i tried at first to get the framework to be able to cope with everything nethack would throw at it 04:50:55 unluckily nethack is programmed in C :-) 04:51:22 luckily writing a hash table takes what, 20 minutes? 04:52:37 writing a hash table where you also can look up parts of a key or value? 04:53:55 i don't say that a large table is the niftiest code there is, but it works extremely well, is fast and is quite robust. there is no need (in nethack-de) to use a more complex solution 04:54:32 back to galehar's question, though: a hash or std::map lookup is fast, bhaak's linear lookup through 20000 entries is good enough, parsing a SQL query and a disk access would be a hog 04:54:56 bhaak: looking up parts is nice, I just have a different concept 04:56:02 you can put sqlite into a memory mode. although I'm not sure if you can do that with data that's been loading from a file, OTOH, the OS will probably load the complete sqlite file into memory anyway when you are using it constantly 04:56:06 I wish gettext had no problems with encoding (which it insists on always doing for you) or multiple domains, it's .mo files are exactly what I need 04:56:26 yeah, but you still need to parse SQL statements 04:56:39 preparing them beforehand can speed it up, not sure if enough, though 04:57:06 off the top of my head i can't remember why i needed parts but it is needed and also synonyme handling is easier in a linear list. 04:57:42 depends a bit on your sql driver (for general sql db), i think sqlite itself doesn't have "prepared statements" 04:59:03 oh, it has. but your queries would probably be quite easy and there would not be a huge performance loss 04:59:38 for (std::map::const_interator i = translations.begin(); i != translations.end(); ++i) if (is_this_a_part_of_the_string(lookie, i->second) return i->first; 04:59:46 (C++ syntax is beyond abysmal) 05:00:06 OSX builds of master branch on crawl.develz.org updated to: 0.11-a0-1273-gef185ba 05:00:10 here you have a linear search 05:00:49 although all of this is quite theoretical. the only way to really know would be to try it out. but i wouldn't use sqlite unless there is a compelling reason for it (for example if the code gets to ugly with simpler data structures) 05:01:39 since Crawl already uses C++, there's no reason to avoid the STL and std::map 05:02:12 the only thing i wouldn't be sure about is what an impact a specific approach would have on a public server. i'm sure 100s concurrent nethack-de's on a public server could have some issues, but that's not a danger for the foreseeable future :) 05:06:26 what do you mean, Nethack 4.3 is not going to have a .de translation?!? 05:07:56 kilobyte: so maybe a simple text file (similar to description files?) and it's loaded in a map at game start 05:08:28 dat/name/xx/names.txt for example 05:08:40 kilobyte: i'm quite sure that by the time NetHack 4.3 is released, we have universal translators tha can translate on-the-fly by manipulating our visual cortex so we don't even see non-native text anymore. 05:09:11 bhaak: ah, so the time between releases isn't going to shorten then 05:09:37 bhaak: it's already on the crawl 1.0 feature list ;) 05:10:29 galehar: but aren't we going for 0.∞ ? 05:11:22 0.9999999..... is equal to 1 :-) 05:12:35 I'd go for something more complex, it'd be a pain to have every single noun case as a different entry. In Polish for example, for every noun, there's 6 cases (7th one no one cares about), gender, and plural. The plural can have cases too, but at least for actors only plural nominatives are ever used. 05:13:45 bhaak: if it ever reaches 0.9999999...., which it won't do as even if we produce twice as many Crawl releases every year as the one before, at the heat death of the universe it will still be a finite number 05:14:30 and also, version numbers are not reals but a vector of dot-separated integers 05:15:11 ok ok, Knuth uses real numbers, but his ones are capped too :) 05:15:29 kilobyte: that's certainly a bug then. you just need to incorporate a feature in crawl that generates a new universe before the old one dies 05:15:47 ... 05:15:47 -!- clouded_ has joined ##crawl-dev 05:15:51 not even emacs has that! 05:16:00 yet! 05:16:01 that's planned for emacs 24 05:16:24 ah, so that's why it isn't out yet 05:16:36 the testing is kind of tricky 05:19:17 -!- mikee_ has quit [Quit: HydraIRC -> http://www.hydrairc.com <- Like it? Visit #hydrairc on EFNet] 05:41:18 elliptic, galehar: any ideas what Shadow Creatures should do in places that have no native monsters? In trunk, Temple and Lab it does nothing but still takes mana and time, in LEVEL_PORTAL_VAULT it creates monsters from the last LEVEL_DUNGEON map you were on. 05:41:57 ie, a Shadow Creatures user wants to enter Pan from a D:27 portal 05:44:25 ideally I think everywhere would have some list of native monsters... there is a similar issue with kiku corpses 05:46:31 portal_branches have no lists for: Temple, Ziggurat (always overridden though), Lab, all portals except Sewer and Volcano (IIRC Ice Caves tend to override, not sure if always) 05:47:11 so I should define a list for all portals then? Sounds mostly ok. 05:47:32 except for Bazaar, Troves and Labs -- what to do there? 05:47:53 IMO labs should have more monsters in them anyway :P 05:48:22 my vague ideas for improving them: make the map smaller, make it shift more frequently, put more monsters wandering around 05:49:02 Makefile not detecting Command-Line Tools only Mac setup (https://crawl.develz.org/mantis/view.php?id=5597) by vickio 05:49:06 more hungry ghosts are a necessity 05:49:20 hungry Donald :p 05:49:25 hah 05:49:47 HUNGRY DONALD YEAH <3 05:49:52 (still waiting for folks to shout at me, and to possibly revert) 05:49:52 not sure about bazaar/trove/temple... might be simplest for shadow creatures and kiku corpses to just not work there 05:49:56 kilobyte: portal vaults should define monster lists 05:50:06 but if due approves, I guess Donald stays :p 05:50:06 I thought most did that already, aside from bazaars 05:50:31 kilobyte: dude you're putting donald in lab occasionall? 05:50:32 kilobyte: you made donald generate in labs already? I missed this 05:51:27 I also think that alongside making labs smaller (thinking like half the size in each dimension), we don't need to stick with the hunger theme... nobody starves in them even now :P 05:52:16 and mythologically I'm not sure people starved in labyrinths either... 05:52:43 -!- Pingas has joined ##crawl-dev 05:53:14 -!- Pingas has quit [Client Quit] 05:53:17 elliptic: a64339f4 05:53:20 -!- Pingas has joined ##crawl-dev 05:54:48 <3 05:54:56 if that commit stays, there's a question whether he should be able to get out if you leave him there. Currently he doesn't, which IMO is consistent with other levels. 05:55:10 yeah, I don't see a problem with leaving him in there 05:55:50 given the donald speech you added, it seems like he's too stupid to find his way out anyway :P 05:57:47 poor donald :D 06:01:17 labyrinth donald best donald 06:01:39 -!- syllogism has joined ##crawl-dev 06:02:03 kilobyte: looks to me like every portal vault (that gets monsters) defines a monster list (with set_random_mon_list() ), except for baileys and ossuaries 06:02:34 -!- ais523 has quit [Remote host closed the connection] 06:06:33 also Trove and Bazaars (both can have hostiles) 06:06:48 and wizlab_golubria 06:07:59 wizlab_doroklohe has "mummy" even for other monster sets 06:09:16 elliptic: I looked into Lab shifts. They are weighted to happen in ummapped areas, which is why you almost never notice them 06:09:35 wizlab_golubria still lacks a monster list? 06:09:53 I thought I gave it one a few days after that vault went in 06:10:22 I'm thinking of removing this so they can happen anywhere, make them more frequent and print a message only when it happens nearby 06:10:25 yeah, it definitely has one (was worried I forgot to push it) 06:10:48 oh, it has in trunk, didn't sync it into portal_branches yet 06:15:11 -!- MarvinPA has joined ##crawl-dev 06:30:59 hmm, in wizlab_doroklohe adding those can be an information leak: you will know what monsters you'll face 06:34:33 this applies to Zigs as well: cast Shadow Creatures and if you'll get daevas or pan lords, wear brown pants and burn resources to prepare 06:35:22 if you cast shadow creatures and get pan lords, you should probably summon more pan lords... 06:41:17 sadly, it seems to limit itself to normal demons 06:47:37 -!- alefury has quit [Quit: KVIrc 4.1.3 Equilibrium http://www.kvirc.net/] 06:50:23 WTF, >40 Zig:27 levels, and not a single Pan lord, holy or even mummy level 06:53:25 had to edit the weight; indeed, can get 1s but not actual lords 07:02:39 galehar: hi! Is there any point in putting my lab ideas on the wiki? 07:19:06 -!- evilmike has quit [Read error: Connection reset by peer] 07:22:49 if you tab-fight while you're flying over deep water, does it skip past the "you are starting to lose your buoyancy" warning 07:24:45 -!- G-Flex has quit [Read error: Connection reset by peer] 07:26:00 -!- G-Flex has joined ##crawl-dev 07:41:35 -!- alefury has joined ##crawl-dev 07:42:10 is it intended for the temple to be safe on the orb ascent? 07:42:11 it seems to be, at least 07:54:14 by the time you've made it to the temple you've pretty much won, anyway 07:54:16 it seems a minor exploit 07:55:28 yeah, but there is only one reason it still exists 07:55:51 namely, gitorious errors out on my pushes for the last 10 minutes :p 07:55:59 (it would be funny if, upon going into the temple on the orb run, a pan lord would spawn on the top of every altar in it) 08:01:23 -!- jeanjacques has joined ##crawl-dev 08:01:48 -!- Mottie has joined ##crawl-dev 08:01:54 dpeg: I think so. Personally, I don't think I'll do more than a couple of minor things, but elliptic was also interesting in improving them 08:02:09 (although I don't know if he actually plans on doing some code) 08:02:55 headzone: I don't see why the warnings wouldn't work. Try it in wizmode and tell us 08:04:22 i meant that the warning might not help much if it's drowned in a flood of tab messages 08:07:42 isn't that a force_more sort of message 08:07:45 or did I do that myself 08:07:52 it really ought to be, by default, if it's not 08:08:51 it's a force more 08:09:12 in that case you aren't about to miss it by tabbing 08:10:05 cool 08:10:44 speaking of force_more: could crawl be made to have lots, lots more force_mores by default? 08:10:52 assuming that wouldn't be too annoying 08:11:06 such as? 08:11:12 !rc 08:11:17 http://crawl.akrasiac.org/rcfiles/crawl-0.10/Patashu.rc 08:11:20 galehar: okay, will do 08:11:20 let's see 08:12:03 hey dpeg! <3 08:12:13 Hi! But no time! 08:12:40 Patashu: what about miscast messages 08:12:45 basically, I think any situation that demands you slow down and play step by step for a while demands a force_more. so things like: item destruction, draining, rot, hit by distortion, hit by electrocution, hit by venom, godly wrath, hell effects 08:12:52 yeah, I think miscast should always be a force_more 08:12:53 I'm trying to think of times when I could use a prompt but don't have one 08:13:02 and the biggest one is when I miscast a spell but think I cast it fine 08:13:07 yes 08:13:10 that was one of the first ones I added 08:13:10 because the effect isn't obvious and I'm not paying close attention 08:13:13 like when extending an effect 08:13:21 if you look at my rc 08:13:24 I have a bunch I took from crate 08:13:26 and a bunch it didn't have 08:13:43 imo overdoing force_more would be bad, because that might train people to just reflexively push through them 08:13:46 but I don't know about that 08:14:11 I do agree, which is why it should be reserved for things that always demand you slow down and look at the situation 08:14:18 yeah 08:14:19 e.g. you or your items or your stats/expa re in immediate danger 08:14:25 what about finding things like abyss/pan exits and runes? 08:14:31 yes, definitely 08:14:33 (that was in crate's rc too) 08:14:49 oh I didn't see that those were there 08:15:48 Abyss rune is a must, yeah 08:15:53 the problem is, some players need more mores, others less 08:15:56 Pan exits/runes: why? 08:16:13 I don't know why I mentioned runes 08:16:21 an i think going with a minimal approach is better in this case 08:16:22 oh, wait, not pan exits/runes 08:16:24 abyss exits/abyss runes 08:16:36 yeah 08:17:08 collecting some useful force mores in a file that can be easily included and shipping that with crawl could be useful 08:17:10 I'll collect what I think are the most important, then 08:17:43 alefury: yes, absolutely -- we already do that with other stuff 08:32:55 hah, finally managed to push\ 08:33:38 probably wasn't dilated enough 08:33:49 galehar, other folks: I guess it's time to ask for some code review or at least a sanity check or two 08:34:03 merging portal_branches? 08:34:41 should be doneish, yeah 08:34:45 yay! 08:34:54 ok, will try to have a look at it later 08:34:57 03kilobyte 07portal_branches * r902463731fcc 10/crawl-ref/source/mon-place.cc: Don't special-case old LEVEL_PORTAL_VAULT wrt Shadow Creatures. 08:34:57 03kilobyte 07portal_branches * r551f343ca87a 10/crawl-ref/source/dat/des/serial/ponds.des: Simplify serial/ponds.des 08:34:58 03kilobyte 07portal_branches * rfee89fcab51d 10/crawl-ref/source/ (branch-data.h mon-pick.cc mon-pick.h): A fallback monster set for Ice Caves. 08:34:58 03kilobyte 07portal_branches * rd562eac1e7b1 10/crawl-ref/source/files.cc: Insult wizmode users instead of crashing when entering a recursed portal. 08:34:58 03kilobyte 07portal_branches * rdaff41667b18 10/crawl-ref/source/dungeon.cc: Don't let people rest in the Temple on the Orb run. 08:34:58 03kilobyte 07portal_branches * rc968e064d0f3 10/crawl-ref/source/dat/des/portals/wizlab.des: Set monster sets for wizlab_doroklohe (for Shadow Creatures). 08:35:04 03kilobyte 07portal_branches * rb15ddc7bf039 10/crawl-ref/source/stairs.cc: Base updating or not the travel cache on whether the old level still exists. 08:35:04 03kilobyte 07portal_branches * r5556f802f8d0 10/crawl-ref/source/ (branch-data.h mon-pick.cc mon-pick.h): Fix Shadow Creatures not doing anything in Baileys and Ossuaries. 08:35:56 (I intentionally left one WTF-level regression as a test :p) 08:38:55 kilobyte: awesome! 08:39:33 kilobyte: merging in trunk is also an efficient way to spot bugs ;) 08:39:53 off to exercises 08:47:57 http://pastebin.com/CinveeYH -> What about these force_mores? 08:53:31 actually, http://pastebin.com/4BxBAvhe 08:54:47 these seem to be very inclusive, I doubt it'd be good to have them as default 08:55:02 some are pretty important, though, of course 08:55:29 there are a lot, yes 08:55:32 but I feel they're all still important 08:58:33 the mutation message I just realized is too inclusive, need to edit it down 08:58:35 "You have reached level" -- how exactly is this relevant? 08:58:53 kilobyte: it's in the default rc :) 08:58:55 that's how it's relevant! 08:59:09 "no visible monsters within range" is especially unfit for a force_more 08:59:13 -!- Ragdoll has joined ##crawl-dev 08:59:26 that's also in the default rc 09:00:09 http://bpaste.net/show/27736/ 09:00:23 oops, without that harpy line 09:02:09 "you feel your flesh rotting away" 09:02:15 you mean, ghouls? 09:02:43 _starting_ to rot obviously warrants a force_more, rot continuing not really 09:02:44 doesn't that also show while on potion of decay or something? 09:02:54 fine with me, yeah. 09:03:43 yeah, probably only starting to rot is needed 09:03:47 i actually have a lot more force_more stuff in my RC than what Patashu posted, but i don't think it's relevant for all players .. most would really lead to everybody hacking on keys whenever force_more shows up, thus ignoring the really important ones 09:06:12 also with new xom you could get several force_more in high tension situations 09:06:20 might want to discuss whether that's desired or not 09:07:10 it's like most users ignoring Windows popups 09:07:27 people actually playing xom probably know what to look for 09:07:35 because most programs ask such vital questions like "Are you sure you want to exit the program?" 09:08:00 I generally don't see programs do that unless there's some specific reason 09:08:19 better windows comparison: when Vista was first release and UAC asked you every time you wanted to look at the screen 09:08:24 released* 09:08:25 i once tried quitting skype 09:08:28 this is nontrivial 09:09:04 ok, how about this? http://pastebin.com/hwc0jm7F 09:09:25 still no fan of the Misc section 09:09:36 the first three lines in misc are in the default rc 09:09:39 shafting is probably worth a pause though 09:09:46 yeah that doesn't mean they can be removed :) 09:10:12 not sure if scales changing needs a force_more, since it always coincides with a level up, right? 09:10:18 why don't you just rip off marvinpa's rcfile? 09:10:23 that's what everyone does :P 09:10:28 G-Flex: a few of programs I often interact with at work use Firebird, their Windows client is a program named "ibExpert". It not only asks whether you really want to exit it, but also repeats the question for every table windows you have open (even though they'll be restored on next start) 09:10:36 !rc marvinPA 09:10:37 http://crawl.akrasiac.org/rcfiles/crawl-0.10/MarvinPA.rc 09:10:42 i added half a dozen or so, but it's good overall 09:10:53 oh, yes 09:10:56 this is the one I ripped off :) 09:11:11 -!- MarvinPA_ has joined ##crawl-dev 09:11:39 stuff like "you miscast" shouldn't be there 09:12:22 the slime stuff neither 09:12:45 "the harpy devours", we all hate it, let's not make us hate it more :P 09:13:03 -!- ToBeFree has joined ##crawl-dev 09:13:04 -!- ToBeFree has quit [Changing host] 09:13:04 -!- ToBeFree has joined ##crawl-dev 09:13:21 yeah :) I started all the ones I thought might be annoying commented out, this time 09:14:03 -!- MarvinPA has quit [Ping timeout: 256 seconds] 09:14:03 -!- MarvinPA_ is now known as MarvinPA 09:14:09 # force_more_message = You don't.* that spell -> this is a good one actually 09:14:15 kilobyte: that's pretty ridiculous, although in any case, that's the application's fault, not Window's 09:14:19 prevents fat fingers 09:19:08 (started all the ones? commented out all the ones...) 09:19:21 -!- syraine has joined ##crawl-dev 09:19:25 I believe I may have just crashed 09:19:39 ASSERT(grd[i][j] < NUM_FEATURES) in 'tags.cc' at line 3528 failed. 09:19:41 Writing crash info to /crawl-svn/morgue/XiN/crash-XiN-20120425-141920.txt 09:25:05 -!- Surr has joined ##crawl-dev 09:26:20 my game seems sort of dead 09:26:28 it's not responsive 09:26:49 So, I came here to report a problem as such: My console (CDO terminal) stopped responding and I closed it manually. Now the game gives me this message when I try to continue my game: Another game is already in progress using this save! 09:27:37 it appears plenty of people have this problem 09:27:58 i get "Connection to crawl.develz.org closed., please contact the admin." when i try to reconnect 09:28:16 03galehar * rcb2d9b160283 10/crawl-ref/settings/init.txt: More prompt for finding abyssal exits and rune. 09:28:17 03galehar * r23ccb3ba91cf 10/crawl-ref/source/util/desc2yaml.pl: A script to convert a description file in yaml format. 09:30:46 absolutego: when casting a wrong spell, you get an error message 09:30:54 that's enough 09:31:15 i appreciate it though, it's a matter of preference 09:31:18 and if not, you really could just force_more=* 10:00:44 -!- ussdefiant has quit [Ping timeout: 250 seconds] 10:01:29 hm my game seems to have frozen too 10:03:51 Just to be 100% sure before I go back to ignoring IRC, this is CDO, not CAO, right? 10:03:57 yes 10:04:30 Cool. Sorry I can't help, good luck 10:10:28 03kilobyte 07portal_branches * re8da5e6c0195 10/crawl-ref/ (264 files in 30 dirs): Merge branch 'master' into portal_branches 10:18:26 -!- Fa has joined ##crawl-dev 10:20:10 if felids can wear rings shouldn't they be able to hold wands 10:20:14 -!- syraine has quit [Read error: Connection reset by peer] 10:20:23 -!- Fa has quit [Read error: Operation timed out] 10:20:41 -!- syraine has joined ##crawl-dev 10:23:00 Unstable branch on crawl.akrasiac.org updated to: 0.11-a0-1303-g23ccb3b (32) 10:23:20 headzone: no, how would they point them at something? 10:25:45 holding them between their front paws :) 10:45:40 -!- Patashu has quit [Quit: MSN: Patashu@hotmail.com , Gmail: Patashu0@gmail.com , AIM: Patashu0 , YIM: patashu2 , Skype: patashu0 .] 10:46:07 -!- blabber has joined ##crawl-dev 10:50:13 -!- alefury has quit [Ping timeout: 248 seconds] 10:51:05 -!- bmh has joined ##crawl-dev 10:51:12 * bmh looks around 10:57:00 -!- bmh has quit [Quit: Page closed] 11:02:45 -!- absolutego has quit [Quit: Lost terminal] 11:15:22 imo, let stuff use wands by wielding them 11:15:33 felids, spiders 11:16:24 maybe pigs too, although pigs are weird 11:17:24 for felids it mainly just seems like a balance improvement 11:31:45 <|amethyst> FR: humanform spell allowing Felids, Octopodes, etc. to wear armour. It melds when they untransform 11:32:00 :) 11:32:13 that's got to be in the bad_ideas list multiple times already :P 11:33:53 -!- elliptic_ has joined ##crawl-dev 11:34:05 -!- elliptic has quit [Ping timeout: 260 seconds] 11:34:49 <|amethyst> headzone: yes, *if* you were able to make one step 11:34:51 <|amethyst> doh 11:49:38 -!- elliptic_ is now known as elliptic 12:02:35 just thought I'd mention that it seems rather harsh for a large pack of orange demons and rotting devils to ambush the player at least as early as D:6 12:02:45 ran into some vault with them spilling out of a couple rooms or something like that 12:06:12 Napkin: around? my CDO game stopped responding and won't resume, sounds like the same thing that happened to absolutego 12:06:45 hey 12:06:52 char name, elliptic? 12:06:57 hyperbolic 12:07:42 clouded (L12 HaTm) (Volcano) 12:07:45 Honor (L6 DEIE) (D:4) 12:07:51 jaccarmac (L2 OpTm) (D:1) 12:07:53 sose (L17 MfSk) (Orc:4) 12:08:00 omil (L4 DEAE) (D:2) 12:08:21 note: i killed XiN's 100% game, nothing else 12:08:45 your game ended on its own as well, elliptic 12:08:50 lock problem? 12:09:35 thanks! 12:09:53 np :) 12:10:28 -!- dtsund has joined ##crawl-dev 12:10:36 it seemed that games started to hang when a milestone was about to happen 12:10:52 <|amethyst> yeah, mine works again too 12:11:14 irgs, definitely sounds like a locking problem then 12:29:15 -!- Surr has left ##crawl-dev 12:53:23 Hey napkin, can you clear a phantom game on webtiles for me so people can watch me alistair's the vault? 12:54:09 for some reason my session has two games, one of which is the same game from earlier 12:55:09 and updates like "entered a treasure trove" are being applied to the 35 hour old one 12:58:50 "MinoMax"? 13:00:01 Yessir, Totes says he has a phantom game too 13:00:02 hmm, definitely there seems to be a problem with idle time.. 13:00:19 !tell edlothiol there is a problem with idle times 13:00:19 Napkin: OK, I'll let edlothiol know. 13:00:19 yeah, Araganzar also has a 500+ hour old game that's phantom 13:01:08 thanks, I'm not sure how to duplicate it but it's happening fairly frequently now 13:03:45 ok, i restarted the webtiles server, MinoMax 13:03:47 -!- shirish has joined ##crawl-dev 13:03:49 thanks for letting me know 13:04:12 sure - I filed a ticket on the problem a few days ago, let me link it in case you need that for anything 13:04:13 i'll try to catch edlothiol - he should know how to fix that 13:04:20 https://crawl.develz.org/mantis/view.php?id=5586 13:04:25 thanks man! 13:04:33 !tell edlothiol https://crawl.develz.org/mantis/view.php?id=5586 13:04:33 Napkin: OK, I'll let edlothiol know. 13:04:40 thank you, too :) 13:18:14 -!- clouded_ is now known as st_ 13:24:32 03elliptic * r09c327fa2af2 10/crawl-ref/source/ (evoke.cc spl-summoning.cc): Make earth elementals require more skill, water elementals require less. 13:41:36 -!- ToBeFree has quit [Read error: Connection reset by peer] 13:46:02 -!- dtsund has quit [Ping timeout: 244 seconds] 13:50:51 -!- MinoMax has quit [Ping timeout: 245 seconds] 14:05:26 Napkin: you say XiN's game was taking 100% CPU? Did you happen to notice if it got stuck very recently? 14:05:47 not stuck at 17:30 when i left work 14:06:01 because the auto-killer should have triggered. It doesn't act on locking issues that don't spin the CPU though. 14:13:53 -!- rast has joined ##crawl-dev 14:14:30 so im compiling on a box that where i dont have root 14:14:33 * Need to build contribs: lua/src 14:14:34 make: *** contrib: No such file or directory. Stop. 14:14:51 where do i get this contrib directory? 14:16:29 <|amethyst> rast: if this is a git clone, then from the top-level crawl directory run: git submodule update --init 14:16:45 its the 0.10.2 source 14:16:47 <|amethyst> alternatively, you can install your system' 14:16:48 rtfm ;) 14:16:56 <|amethyst> alternatively, you can install your system's lua-dev packages, whatever they're called 14:17:06 -!- dtsund has joined ##crawl-dev 14:17:16 im afraid to use git because i dont know how to make sure i get the right contrib for 0.10.2 instead of trunk 14:18:01 <|amethyst> INSTALL.txt is probably the best thing to follow 14:18:20 0.10.2 has written, what "version" of contribs are to be downloaded 14:18:31 <|amethyst> that too 14:18:33 ~ linked 14:19:12 <|amethyst> git checkout 0.10.2 (or better git checkout stone_soup-0.10 to get the latest fixes) then do the submodule init thing 14:29:04 so i did git clone git://gitorious.org/crawl/crawl.git 14:29:30 its downloading about 60Mb worth of stuff 14:30:54 wonderful, isn't it? ;) 14:31:27 i dunno, i only wanted 0.10.2 14:31:30 it has it's advantages though 14:31:34 *its 14:32:13 what were you doing before? 14:32:23 downloaded a src tarball without contribs? 14:32:27 right 14:33:01 this one? http://sourceforge.net/projects/crawl-ref/files/Stone%20Soup/0.10.2/stone_soup-0.10.2-nodeps.tar.xz/download 14:33:20 sec. but that looks like it 14:33:27 or was it the one with the dependencies, which failed? http://sourceforge.net/projects/crawl-ref/files/Stone%20Soup/0.10.2/stone_soup-0.10.2.tar.xz/download 14:33:48 i downloaded the without 14:33:57 http://sourceforge.net/projects/crawl-ref/files/Stone%20Soup/0.10.2/stone_soup-0.10.2-nodeps.tar.xz/download 14:34:05 was that my problem? 14:34:23 right, the one without the contrib deps doesn't have the contrib deps 14:34:46 i had no idea what"woth deps" meant 14:36:44 true, true, INSTALL.txt could use an update - and the downloads-page could do better by listing "tarball" and "tarball without deps" 14:37:15 "without" usually sounds like the wrong version 14:37:27 while "with" sounds like the extra think version 14:37:28 ;) 14:37:32 *thick 14:40:41 nice, with deps seems to be working fine 14:41:17 this after i finished getting it with git also 14:41:43 it'll be useful in future, if you ever want to upgrade 14:42:08 > and the downloads-page could do better by listing "tarball" and "tarball without deps" 14:42:11 done 14:43:00 -!- Textmode has joined ##crawl-dev 14:49:02 compile problems with libunix.cc 14:49:14 libunix.cc: In function ΓÇÿint getchk()ΓÇÖ: 14:49:15 libunix.cc:241: error: ΓÇÿget_wchΓÇÖ was not declared in this scope 14:49:44