00:03:43 <Henzell> Unstable branch on crawl.akrasiac.org updated to: 0.9.0-a0-298-gcc26816 (32)
00:06:09 -!- valrus_ has quit [Remote host closed the connection]
00:47:20 -!- Textmode has joined ##crawl-dev
01:07:29 -!- ortoslon has joined ##crawl-dev
01:16:44 <Henzell> LuckyNed the Grappler (L12 TrBe)  (Lair:5)
01:18:48 -!- valrus has quit [Remote host closed the connection]
01:24:47 -!- ophanim has quit [Quit: Leaving.]
01:51:13 -!- monqy has quit [Ping timeout: 260 seconds]
01:53:06 -!- monqy has joined ##crawl-dev
02:06:35 <Zaba> moin
02:23:44 -!- Jordan7hm has quit [Ping timeout: 248 seconds]
02:44:37 <Napkin> moin moin
02:53:28 -!- casmith789 has quit [Read error: Connection reset by peer]
02:59:34 <Napkin> ok, anything left to change on CDO?
02:59:51 <Napkin> kilobyte, wake up! work awaits you :>
03:13:44 <bhaak> Napkin: on CDO, pressing P gets you to 0.8.0. is this intentional? (i thought you always go to the topmost version in the menu list)
03:14:17 <Napkin> hey, i changed that yesterday evening.. since when is your dgl running?! ;)
03:14:47 <bhaak> just logged in
03:15:09 <Napkin> seriously? :-O
03:15:13 <Napkin> i just tested
03:15:16 <bhaak> yepp and just tried it again
03:15:23 <Napkin> do it again, please
03:15:39 <Napkin> keep the game open for a second
03:15:43 <Napkin> arg
03:15:46 <Napkin> you are too fast ;)
03:16:06 <bhaak> no, i didn't see your request. now I'm open
03:16:16 * due eats Napkin.
03:16:17 <bhaak> and entered by pressing P
03:16:53 <Napkin> and you didn't enter a submenu first?
03:17:04 <Napkin> 100% sure?
03:17:14 <bhaak> yes
03:17:23 <Napkin> seriously can't reproduce that
03:17:51 <bhaak> I can ttyrec it
03:18:01 <Napkin> when i press p from the main menu it starts svn/trunk
03:18:06 <Napkin> p & P
03:18:12 <Napkin> same for s & z?
03:18:25 <bhaak> what are s & z?
03:18:47 <Napkin> to start sprint or zotdefence
03:19:15 <Napkin> ahh, got it
03:19:30 <bhaak> yepp "This is Dungeon Crawl Stone Soup 0.8.0-1-g574be46"
03:19:40 <bhaak> although zotdefence doesn't want to start for me
03:19:45 <Napkin> the new feature to have different menus for admins and users caught me
03:19:51 <Napkin> fixed now, after you re-login
03:20:07 <bhaak> yes, works now
03:20:12 <Napkin> thanks :)
03:20:29 -!- galehar has joined ##crawl-dev
03:20:40 <galehar> hi
03:20:43 <Napkin> moin galehar
03:21:11 <Napkin> too many things to change on a release.. easy to miss something, bhaak
03:21:33 -!- ortoslon has quit [Read error: Connection reset by peer]
03:21:43 <bhaak> you need a release checklist :)
03:22:02 <bhaak> also a post mortem release conference meeting ;-)
03:22:20 <Napkin> oh yes, that's definitely required ;D
03:23:13 <Napkin> mantis changed too now
03:23:24 <Napkin> 2 "ancient branch"es already ;D
03:30:32 <kilobyte> still need greensnark for the Mac builds...
03:31:43 <kilobyte> bhaak: zotdef is disabled :(  Crawling with serious bugs like crashes or all monsters freezing and doing nothing while you kill them with impunity, with balance out of whack.
03:32:04 <bhaak> oh
03:32:42 <kilobyte> you can play it in trunk... trunk has to earn its reputation as "unstable" :)
03:34:09 <Napkin> did you work on the debian repository yesterday, kilobyte?
03:34:36 <Napkin> i see the lenny folder.. remember, that CDO is now on squeeze
03:34:46 <Napkin> for the biggest parts, at least
03:34:57 <Napkin> also, I saw the script called un437
03:35:07 <Napkin> am I supposed to let that work on the morgues?
03:35:23 <Napkin> is it ready for deployment?
03:35:46 <kilobyte> these packages are built against lenny's libraries -- they work on lenny, squeeze, to-be-wheezy and sid, and whatever old Ubuntu VM I had installed
03:36:33 <kilobyte> un437: yeah, should be fine
03:36:49 <Napkin> that doesn't sound too sure
03:36:52 <kilobyte> lemme backup all morgues in case I made a boo-boo, but I think I tested that script well enough
03:37:12 <Napkin> no, let me backup them
03:37:12 <kilobyte> I mean, I tested it on a number of morgues
03:37:20 <kilobyte> ok cool
03:38:10 <Napkin> that will take a long time though
03:38:12 <bhaak> Napkin has the right admin attitude. never trust a programmer :)
03:38:26 <kilobyte> it won't recover old games with cset_unicode since there's no way to tell what character was truncated, but that's no regression
03:39:23 <kilobyte> well, we slowly learn how to test stuff.  Remember the 0.7.0 release?  There was a 0.7.1 immediately, with several serious crashers.
03:39:56 <Napkin> which morgues do we want to convert, kilobyte?
03:39:57 <kilobyte> the script is idempotent -- you can run it over and over without negative effects
03:40:05 <Napkin> all of them? all >=0.8?
03:40:24 <kilobyte> all
03:40:29 <Napkin> yikes
03:40:56 <kilobyte> you'd have to set conditional SetCharset for different directories in apache otherwise, for no better result than this script can already give you
03:41:04 <Napkin> not those of 4.1.2a though, right?
03:41:18 <Napkin> or those too?
03:41:44 <kilobyte> those too, although I'm not aware of any player who played 4.1 with non-ASCII
03:46:52 <Napkin> time nice -n 19 ionice -c3 /bin/tar -cJf /home/services/crawl/backup/morgues-0.6+.tar.xz crawl-0.6/morgue/ crawl-0.7/morgue/ crawl-0.8/morgue/ crawl-svn/morgue/
03:46:56 <Napkin> let's see
03:47:09 <Napkin> care to elaborate more about the debian packages?
03:47:24 <Napkin> are they signed?
03:48:00 <Napkin> Suite stable and codename lenny, huh?
03:48:35 <kilobyte> it's the two aliases you can use in apt lines
03:49:20 <kilobyte> I am not sure what scheme to use: currently it's "deb http://crawl.develz.org debian lenny 0.8" or "deb http://crawl.develz.org debian stable 0.8"
03:49:59 <kilobyte> unlike most Debian packages which are targetted at particular archives, we care about Crawl versions instead
03:50:49 <kilobyte> hmm... good point, "stable trunk" would be confusing
03:51:01 <Napkin> i'd prefer deb http://crawl.develz.org unstable main
03:51:33 <Napkin> if you want them separated, then we could have "stable main" and "unstable main"
03:52:10 <Napkin> if - as you say - those packages work independent on distribution
03:52:39 <kilobyte> currently... I don't know if that is going to change
03:53:49 <kilobyte> actually, we could abuse the "component" field in such case: if, say, Ubuntu Lousy will need a separate build, it could be "stable lousy" while "stable main" attempts to be universal
03:53:49 <Napkin> I think it's cumbersome - at least for the user - to keep master-branch-based builds separated from the stable ones
03:54:07 <Napkin> sounds good
03:54:19 <kilobyte> lemme give it a try
03:54:24 <Napkin> wait
03:54:38 <Napkin> what about having stable and unstable builds in one?
03:55:53 <kilobyte> we can use the "distribution" field for that
03:56:14 <Napkin> let's not?
03:56:22 <kilobyte> it _is_ technically wrong but would probably be less surprising for the user
03:56:57 <Napkin> deb http://crawl.develz.org crawl main
03:57:03 <Napkin> something like that for all builds
03:57:19 <kilobyte> hmm... perhaps using some other word for the distribution would be better?  Instead of "lenny" as in my first attempt, it could be something neutral.
03:57:58 <kilobyte> having multiple builds in one distribution/component part would deny you the ability to use any version but the latest easily
03:58:09 -!- bmh has joined ##crawl-dev
03:58:14 <Napkin> sounds good to me, kilobyte
03:58:21 <Napkin> exactly what we want, right?
03:58:22 <bmh> we're on the front page of slashdot
03:58:54 <doy> heh
03:58:59 <doy> can we handle the traffic?
03:59:00 <Napkin> crawl, crawl-tiles & crawl-dev, crawl-tiles-dev
03:59:04 <bmh> http://games.slashdot.org/story/11/04/27/041234/Roguelikes-the-Misnamed-Genre
03:59:15 <Napkin> probably not.. but not much lag yet ;)
03:59:17 -!- bmh has quit [Client Quit]
03:59:20 <doy> just posted 10 minutes ago
03:59:26 <doy> with a link to the wordpress blog
04:00:52 <kilobyte> crap, we need the 0.8 post NOW
04:01:03 <kilobyte> or people will get the old version
04:01:04 <Napkin> do it, kilobyte!
04:01:10 <kilobyte> any idea what to write?
04:01:12 <Napkin> i'll modify the downloads page
04:01:21 <Napkin> proud to announce, blabla
04:01:31 * kilobyte has ALWAYS a writer's block when it comes to writing any news post.
04:03:50 <kilobyte> heh... crap.  Accidentally pushed the wrong codename yesterday when moving from work to home.
04:04:13 <Napkin> relax, kilobyte :)
04:04:22 <kilobyte> I temp-tagged it to let "make package-source" and Windows builds use the proper tag
04:04:26 <Napkin> we're not going to hurry, just because of a link to our website
04:04:38 <kilobyte> but the proposal to call it "Rage cat is raging" was only a mockery
04:05:40 <due> I can write something if desired
04:05:41 <kilobyte> I guess there are only two options: 1. use that, or 2. don't mention a code name to anyone who doesn't read git tags
04:05:46 <Napkin> please do, due :)
04:05:54 <kilobyte> whoo!
04:06:08 <due> can someone quickly summarise 5 "massive" changes for me?
04:06:19 <due> just in bullet point
04:06:47 <due> kilobyte: Also, I vote for Frantic Frenzied Felines.
04:06:52 <Napkin> * Improved heavy armour.
04:07:00 <kilobyte> raging rage cats, accursed rage cats who know the colour of your underwear thanks to Ashenzari, working armour to let you be safe against scratches, ...
04:07:01 <Napkin> * Split Enchantments into Hexes (harm foes) and Charms (self-buffs).
04:07:10 <Napkin> * New god: Ashenzari, god of divinations.
04:07:10 <Napkin> * New species: Felids, sentient cats.
04:07:15 <Napkin> btw, kilobyte?
04:07:18 <due> Ooh, right.
04:07:23 <due> Those will be perfect!
04:07:28 <due> Do we actually need a code name or...?
04:07:32 <kilobyte> we don't
04:07:33 <Napkin> could you remove the line "* New game mode: Zot Defence, a tower defence mini-game." from the changelog.txt in docs?
04:07:43 <kilobyte> I did, 0.8 doesn't have that
04:07:53 <Napkin> if you can think of something, sure, due - otherwise not required :)
04:07:59 <due> I vote for something alliterative using F and felines :)
04:08:00 <kilobyte> ... or does it?  I'm kind of sure I removed it, but lemme check
04:08:07 <Napkin> oh, i was in trunk
04:08:12 <kilobyte> ah cool
04:08:48 <Napkin> yes, you did :)
04:08:56 <due> Okay, blog post is titled "Faster, Pussycats! Kill! Kill!"
04:09:02 <kilobyte> where should we put the public key for the apt repository?  http://sprunge.us/hCJG
04:09:06 <kilobyte> <3
04:09:23 <Napkin> in a package inside the repository, kilobyte :)
04:09:23 <due> Because I *can't* resist referencing that movie.
04:09:43 <Napkin> sounds weird to me - i don't know that movie, due ;D
04:10:10 <kilobyte> Napkin: that's kind of a catch 22
04:10:51 <Napkin> ok, downloads page is fixed
04:11:11 <kilobyte> apt won't allow you to install the package unless you specify additional arguments, once it is in you don't need it anymore
04:11:12 <Napkin> due: please mention, that we are still awaiting delivery of shiny new MacOSX binaries
04:11:36 <due> OK.
04:11:43 <kilobyte> actually, should have mailed greensnark instead of waiting for him to come to IRC
04:11:46 <Napkin> it just asked "are you sure you want to install package even though you can't verify the key", kilobyte - did that change?
04:11:47 <kilobyte> will do
04:16:54 <Wensley> "Mennas to Society" was rejected as a codename? :(
04:17:22 <due> Wensley: Mennas was in 0.7
04:18:23 <Wensley> how about "Ashes to Ashes"
04:18:56 <due> Because "Frantic Frenzied Felines" is far morealliterative!
04:18:56 <Wensley> also, as a bullet point you should mention new stalker, since it goes so well with ash
04:19:07 <Napkin> oh, please also mention, that a debian/ubuntu repository is bein created right now, due
04:19:31 <due> Napkin: yes,goingtomentionthat andCAOrunning trunk
04:19:38 <due> whoops,sorry,had tomato sauce on my thumb
04:19:55 <Napkin> hehe, bon appetit! :D
04:21:05 <Wensley> also, do we want to mention the unofficial 0.8 celebration tournament happening soon in the newspost?
04:21:26 <Wensley> rearrange those words until they make sense
04:21:56 <due> there's an unofficial 0.8 celebrationtournament?
04:24:26 <Wensley> yes, elliptic is organizing it
04:24:33 <Wensley> ??tournament
04:24:33 <Henzell> tournament[1/4]: Preliminary rules for the 0.8 tourney (May, about 2 weeks long, dates to be determined): http://sites.google.com/site/hyperelliptical/tourney11a.html (download necessary because Google is annoying) Tell elliptic comments/suggestions!
04:24:50 -!- dpeg has joined ##crawl-dev
04:24:54 <due> dpeg: hi!
04:25:05 <due> dpeg: I'm working on th e0.8 release post for the blog, would appreciate your eye if you have time (when it isdone)
04:25:49 <Wensley> dpeg: http://slashdot.org/
04:28:11 <due> Oh crap we're on slashdot?
04:28:19 <due> I'll write faster.
04:28:37 <Wensley> only tangentially, but still
04:29:18 <dpeg> Hi!
04:29:29 <dpeg> due: yes, tell me when to read
04:29:44 <kilobyte> dpeg: hi!
04:29:53 <kilobyte> fortunately the downloads for 0.8 are on SF not CDO
04:30:10 <kilobyte> CDO can handle the homepage even for Slashdot
04:30:33 <dpeg> So is Slashdot important? (I wouldn't know :O )
04:31:04 <Wensley> back in the day it used to be notorious for bringing down sites that were linked, nowadays most sites are robust enough to handle it
04:31:16 <due> dpeg: "Frantic Frenzied Felines" is the code wword I'm running with
04:31:27 <Wensley> (see also http://en.wikipedia.org/wiki/Slashdot_effect)
04:32:10 <dpeg> interesting
04:32:24 <Wensley> due: "Tentacles & Bondage" is my new submission for codename
04:32:50 <due> Malign ateway was in 0.7, no?
04:34:19 <Wensley> gah, was it? I'm so behind on version knowledge
04:34:29 <Wensley> didn't *you* code it?!
04:34:49 <dpeg> I can never remember what appeared in which version. It's all distant past.
04:35:03 <due> Wensley: it was so long ago I don't recall
04:36:03 <due> Okay,
04:36:08 <due> Ash, felids, Ench split...
04:36:08 <Napkin> i think it was
04:36:11 <due> Need too more.
04:36:31 <Napkin> heavy armour changes are important
04:36:37 <due> How did we improve it?
04:36:53 <Napkin> GDR is back or what is it called?
04:37:01 <due> oh.
04:37:20 <Wensley> new stalkers are cool
04:37:21 <Napkin> just changed the formular to make it more efficient again after the big 0.7 nerf
04:37:29 <Wensley> got rid of paladins and makh knights
04:37:35 <Wensley> added abyss knights
04:37:44 <due> Ohh, yes.
04:37:52 <due> I can group them all together.
04:39:01 <Wensley> according to gitk, first references to malign gateway were in 0.8, in july of 2010
04:39:25 <Wensley> (if I'm reading this right)
04:40:03 <due> hm
04:40:08 <due> 0.7 was july 24th
04:40:14 <due> so it was *early* 0.8
04:40:19 <Wensley> due: new spells as well: liquefaction, tornado, the tloc one with the portal
04:40:34 <kilobyte> Malign, Leda, Tornado, Darkness, Sum Dragon, Sum Hydra
04:40:57 <kilobyte> the only of removed spells anyone ever used was Extension
04:41:04 <kilobyte> oh, Golubria
04:41:11 <Wensley> due: this is literally on july 25th, so very, very early 0.8
04:41:12 <kilobyte> the changelog missed it
04:41:59 <kilobyte> aren't we planning on possible minor 0.9 for the tournament?
04:42:52 <kilobyte> the only major features would be (so far) Ash changes and no victory dancing
04:43:03 <kilobyte> 0.8 is good enough though
04:43:10 <Wensley> butchering is a pretty big change
04:43:16 <kilobyte> oh, and that
04:43:30 <Napkin> don't give them too much, due
04:43:39 <due> I'm only oing five.
04:43:44 <Napkin> let them read this: http://crawl.develz.org/main/0.8.0.txt
04:43:46 <due> Ash, Felids, Ench split, backgrounds and heavy armour.
04:43:49 <kilobyte> I still have no idea what to do for gaining 1st rank in a skill
04:44:00 <kilobyte> plus spells maybe
04:44:51 <due> Okay, info on the Debian/Ubunttu repo?
04:45:16 <Wensley> Napkin: the highlights section is missing Passage of Golubria (not sure who updates that document)
04:45:56 <Napkin> being setup right now, due
04:46:50 <kilobyte> Napkin: seems to work, updating the codename was a trivial change
04:47:09 <kilobyte> I nuked the repository and re-imported the packages, but that was a few commands too
04:47:19 <Napkin> how to access it?
04:48:11 <Napkin> Wensley: document is from source docs/changelog.txt
04:48:45 <CIA-45> 03kilobyte 07stone_soup-0.8 * rbb40887794c5 10/crawl-ref/docs/changelog.txt: Add missing Golubria to the changelog.
04:48:47 <due> post is done
04:49:42 <Napkin> let me guess - you wrote it locally first again? ;)
04:50:20 <due> of course
04:50:36 <due> http://sprunge.us/bUeQ
04:52:01 <kilobyte> not visible on the front page
04:53:01 <Napkin> kilobyte: due always writes locally first
04:53:14 <Wensley> due, tell me you remembered to mention the most important change: "Sheep can catch fire and spread it to other sheep."
04:53:21 <Napkin> "post is done" != published ;)
04:53:29 <Napkin> haha, Wensley :D
04:53:44 <kilobyte> due: the instructions for Debian/Ubuntu are: Add the following line to /etc/apt/sources.list: deb http://crawl.develz.org/debian crawl 0.8
04:54:07 <due> kilobyte: Not add-apt-repository?
04:54:22 <dpeg> kilobyte: "gaining 1st rank in a skill" -- what do you mean here?
04:54:35 <kilobyte> due: Public key can be found at http://crawl.develz.org/debian/pubkey
04:55:02 <Napkin> due - just tell them instructions will be in the downloads page soon
04:55:03 <kilobyte> due: there's no such command... you might have some helper package installed, but it's not there on default installations
04:55:13 <due> kilobyte: it's an ubuntu thing  :)
04:55:50 <due> okay
04:56:38 <due> Napkin: preview's available
04:56:42 <due> anyone see any grammar/spelling issues?
04:56:52 <Napkin> checking
04:56:58 <Napkin> "Felines" is right, ja?
04:58:26 <Wensley> kilobyte: another changelog fix, the very first line under the "Levels" heading: New game mode: 'Zot Defence'.
04:58:31 <kilobyte> "Unlike their domesticated equivalents, they are carnivorous and have a slow metabolism." -- I'd axe that line
04:58:49 <kilobyte> especially since domesticated cats ARE carnivorous too, obligate carnivores in fact
05:00:01 <kilobyte> try to feed them vegetarian food for awhile and they'll get permanently blind and then die due to deficiency of several proteins, mostly taurine
05:00:41 <kilobyte> and slowmeta 1 comes from size, not from metabolism per se
05:01:04 <dpeg> kilobyte: yes, go ahead
05:01:45 <kilobyte> even after getting enlarged from "big rat" to "spriggan" size, it still would be weird if the metabolism would be same as for humans
05:02:22 <kilobyte> dpeg: I doubt wordpress handles concurrent edits well, I wouldn't touch the post when due is editing it
05:03:33 <due> yes, felines is correct
05:03:42 <Napkin> check query, due :)
05:04:16 <CIA-45> 03kilobyte 07stone_soup-0.8 * r99ea1d1f45b3 10/crawl-ref/docs/changelog.txt: Axe the changelog line about Zot Defence.
05:04:22 <due> kilobyte: you're right, i had more in that line and then added carnivorous
05:05:29 <Wensley> I think the new tutorial also deserves explicit mention
05:06:44 <Wensley> also maaaabye autofight
05:07:13 <due> Wensley: dude
05:07:16 <due> i'm writing a summary
05:07:18 <due> not an epic
05:07:30 <Wensley> people must know!
05:08:18 <Wensley> "Crawl 0.8: Ferocious Frenzied Furry Fleabitten Felines" is a pretty epic release, judging by the size of this changelog
05:08:36 <due> published
05:08:43 <due> poke myself of napkin for typos etc
05:08:50 <due> Napkin: i haven't added tags or categories, sorry
05:08:57 <due> http://crawl.develz.org/wordpress/faster-pussycats-kill-kill
05:09:07 <Napkin> update category to "News", due :)
05:09:25 <due> done
05:09:56 <kilobyte> due: are you messing with it, or can I amend it?
05:10:21 <Napkin> hehe
05:10:44 <due> kilobyte: i'm done
05:10:51 <due> kilobyte: feel free to modify as necessary
05:11:07 <due> now i have to work out why gnome refuses to show desktop icons until killall nautilus on my grandparents' pc :(
05:12:06 <Zaba> re
05:12:29 <due> hiya zaba
05:12:59 <Zaba> sup due
05:13:42 <Napkin> gconf setting of nautilus
05:13:49 <dpeg> due: good job!
05:14:01 <Napkin> indeed!
05:15:07 <dpeg> If the announcement is stable (kilobyte is editing it, right?), shall we use it for r.g.r.m?
05:18:06 <dpeg> I saw talk of 0.9 targets. While I won't do anything with agendas (my break), I'd suggest to try getting Vintermann code his food ideas. They'd help immediately.
05:18:19 <kilobyte> edited
05:18:23 <due> Napkin: i've been looking through them and none of them seem relevant
05:18:43 <kilobyte> I guess it'd be better to put the apt repositories on the download page rather than in the news post
05:18:45 <due> dpeg: did you see erisdiscordia is in ##crawl?
05:19:04 <dpeg> not on ##crawl, but I know he's arohd
05:19:08 <dpeg> around*
05:19:13 <dpeg> talked to him a bit :)
05:19:31 <kilobyte> due: great thanks for the post, too!
05:19:41 <due> kilobyte: that's okay! it seems to be the one thing i'm consistently okay at ;)
05:20:35 <dpeg> Should I write a post to newsgroup?
05:21:11 <due> dpeg: would be a good idea
05:24:55 <dpeg> typo in "focusses"?
05:25:28 -!- edlothiol has joined ##crawl-dev
05:34:37 <due> dpeg: Don't... think so
05:35:13 <due> Apparently both are accepted.
05:41:51 <due> oh
05:41:52 <due> kilobyte: argh, sorry, I just saved it -- didn't notice the "Warning: Kilobyte is editnig" until I hit update
05:41:56 -!- upsy has joined ##crawl-dev
05:41:56 <kilobyte> due: no worry, I'll try again
05:41:56 <kilobyte> actually, it's ok
05:41:58 <due> was just fixing a few things a proofreader friend pointed out :)
05:41:58 * kilobyte claims the Edit Pumpkin for the downloads page.
05:41:58 <Wensley> "For local builds, mention morgue location after hiscore or when doing a chardump." <-- would it be much more difficult to allow the user to open the file directly from the prompt bearing this message, launching in their default text viewer?
05:48:24 <due> Napkin: can you fix "the our" for me? I'm heading to bed shortly :)
05:48:39 <Napkin> sure thing
05:49:04 <Napkin> :)
05:53:44 * kilobyte releases the lock for downloads; I'll prepare trunk and armel builds first.
05:53:58 <Napkin> wordpress downloads page should be ok
05:54:09 <Napkin> i kept osx builds at 0.7.1 though
05:55:52 <Napkin> my turn at the relase post
06:12:04 <Ashenzari> Enchant armor description not up to date (https://crawl.develz.org/mantis/view.php?id=3883) by mursu
06:18:01 <Napkin> oh, done btw, sorry
06:18:15 <Napkin> nice reply @slashdot, kilobyte
06:20:02 -!- Textmode has quit [Ping timeout: 250 seconds]
06:20:58 <dpeg> just back
06:21:06 <dpeg> r.g.r.m and r.g.r.announce covered
06:21:08 <Napkin> wb, dpeg
06:21:12 <Napkin> awesome! :)
06:21:24 <kilobyte> well, I was more responding to the bashing of the article's poster, in order to not make DCSS look bad
06:21:37 <Napkin> yes, and you did great
06:22:22 <kilobyte> Google Groups don't have it yet
06:22:24 <bhaak> as if the story poster could have known about the release :-D
06:22:37 <kilobyte> "he laughs slowest who has the slowest newsfeed"
06:22:59 <bhaak> kilobyte: if dpeg crossposted, it has first to be approved by the rgra moderator
06:23:06 <kilobyte> bhaak: heh, good point
06:23:13 <kilobyte> bhaak: ah, ok, didn't know
06:24:20 <bhaak> kilobyte: but you're right. it's a weak story, just whoring for some incoming traffic
06:25:10 <bhaak> i tried myself to come up with a sensible comment, but couldn't find something I want to be immortalized on slashed :)
06:26:18 <kilobyte> "immortalized"?  Ok, Slashdot is not /b/, but it's a place full of flamewars even if it has some gems.
06:27:10 <Ashenzari> Lost wifi connection while playing on CDO, character no longer exists (https://crawl.develz.org/mantis/view.php?id=3884) by dsirus5
06:27:41 <bhaak> i didn't want to attach a quality tag to it. just that I don't want to be thought of as a dick  thousands years from now, when google and slashdot still exist :)
06:27:49 <dpeg> bhaak: yes, I did. As always :)
06:27:50 <Napkin> lol @ wifi
06:28:12 <kilobyte> bhaak: good point
06:28:32 <kilobyte> Slashdot is actually pretty good as discussion websites go.
06:28:36 -!- Cryp71c has joined ##crawl-dev
06:28:46 <kilobyte> heck, the best of general purpose ones
06:28:51 <Wensley> kilobyte: speaking of which, you have a (semi-facetious) response
06:29:05 <Cryp71c> Mornin
06:29:45 <Napkin> moin Cryp71c
06:30:19 <bhaak> i generally TRY to be more polite on things that expose itself to the net, so you get more of my real me in here. dpeg knows that i'm a complete asshole in real life ;)
06:30:44 <Napkin> kilobyte? files from the repository are left in /tmp - by design?
06:30:58 <Napkin> lol, bhaak
06:31:25 -!- Twilight-1 has quit [Read error: Connection reset by peer]
06:31:42 <Napkin> average cpu usage by apache processes on CDO, btw
06:31:47 -!- Twilight-1 has joined ##crawl-dev
06:32:05 <Napkin> so far.. but many americans are still sleeping
06:32:15 <kilobyte> Napkin: no, I just didn't delete them yet
06:32:17 <Cryp71c> I don't suppose anyone knows what specifically what value you get when you take a player->as_monster()->number (used to store the number of attacks for monsters) ?
06:32:27 <Napkin> should I?
06:32:32 <Cryp71c> Napkin, I'm not!!!
06:32:47 <Napkin> Cryp71c :D
06:33:03 <kilobyte> I'm afraid I might have to do another import, please let them rot for now
06:33:08 <Cryp71c> (I'm on a bus)
06:33:09 <Napkin> btw guys, we really, really need to redo the screenshot section
06:33:11 <Cryp71c> a web-bus.
06:33:27 <kilobyte> Napkin: good idea, what would you suggest?
06:33:29 <Napkin> it's terribly to modify and too many loading on at once
06:33:47 <Wensley> Napkin: I think that page needs some official Let's Plays
06:34:10 <Napkin> I think that wordpress creates the small representations on the fly - which is a big sin!
06:34:35 <Napkin> Wensley: like a pictured story you mean?
06:34:36 -!- eith has joined ##crawl-dev
06:35:05 <Wensley> that, or a narrated video of someone playing
06:35:20 -!- bmh has joined ##crawl-dev
06:35:21 <Napkin> kilobyte: not sure yet - definitely needs a to be splitted.. like obsolete don't need to be loaded automatically
06:35:36 <bmh> how's the slashdotting going?
06:35:44 <Wensley> at the very least, it would be nice if all the screenshots were consistently grouped
06:35:47 <Napkin> acceptable so far, bmh
06:37:33 -!- monqy has quit [Quit: hello]
06:37:47 <Cryp71c> Napkin, what's going on w/ Slashdot?
06:37:57 <Cryp71c> kilobyte, any advice you have for me for debugging a seg fault?
06:38:00 <Napkin> ok, i take that back - the smaller representations of pictures are created when uploaded and saved on disk
06:38:33 <Napkin> Cryp71c: someone did a spam post on slashdot and mentioned DCSS with a link: http://games.slashdot.org/story/11/04/27/041234/Roguelikes-the-Misnamed-Genre
06:38:34 <kilobyte> Russ Allberry (the moderator of r.g.r.announce) has just posted something on debian-devel so he's awake and probably should process dpeg's post soon
06:39:26 <Napkin> why a spy!
06:39:27 <Napkin> :D
06:39:37 <kilobyte> Cryp71c: not sure what you know about gdb.
06:39:55 <Cryp71c> My merge seems to seg fault whenever jellys try to calculate to-hit.
06:40:11 -!- Wensley has quit [Ping timeout: 240 seconds]
06:40:41 <Cryp71c> Which is odd, because I didn't change that code, I just moved the player_to_hit and mons_to_hit into one function seperating the code by checking atype() == ACT_PLAYER
06:40:46 <Cryp71c> kilobyte, not a lot.
06:42:07 <kilobyte> do you know how to use it without screen corruption?  I either start crawl on another console and attach to it by pid, or let it crash/break then use "shell reset"
06:42:17 <Cryp71c> But only because I havn't had to debug anything yet in this manner, so its better to start learning now that I'm clearly needing it.
06:42:20 <due> aaaaargh
06:42:28 <Napkin> due?
06:42:44 <due> killall nautilus fixes it -- but absolutely none of the configuration tweaks suggested resolve the issue.
06:42:57 <Napkin> damn
06:43:29 <Napkin> i had it the other way around - i had to turn it off
06:43:30 <due> i'm close to nuking .gnome*, .gconf, etc, and startnig the user config again but i'm too tired...
06:43:39 -!- ortoslon has joined ##crawl-dev
06:43:46 <due> heh -- i've added a launcher icon in the enu to run killall nautilus called "start desktop"
06:44:03 <kilobyte> after a crash or breakpoint, you do: "bt", "up" several times until you're in the function that is interesting, then "print some_variable" ("self" tends to be interesting for objects)
06:44:13 <kilobyte> although "bt" will include "self" already
06:44:15 <Napkin> oh, that's not such a bad idea - until have time for a real fix, due
06:44:26 <due> my grandparents absolutely love ubuntu so far! and it's so much easier for me to troubleshoot. i can at least ssh in rather than drive over
06:44:29 <due> or use freenx!
06:44:30 <due> or vnc.
06:44:43 <due> kilobyte: i cannot believe 'up' only just occurred to me now what it atually did; i feel stupid.
06:46:42 <kilobyte> don't worry, I'm a gdb noob too :(
06:47:26 <ortoslon> due: the 0.8 post made me cringe twice: first with the title and second with the abundance of exclamation marks
06:48:38 <due> ortoslon: awesome!
06:48:47 <Cryp71c> kilobyte, it seems when I attach gdb to my crawl process, the crawl process won't do anything
06:48:52 <Cryp71c> Its ignoring keystrokes.
06:48:56 <ortoslon> :)
06:51:04 <Cryp71c> kilobyte, yeah, as soon as i quit out of the gdb session, all the keystrokes I had typed in magically appear.
06:51:24 -!- bmh has quit [Ping timeout: 252 seconds]
06:51:29 <kilobyte> Cryp71c: "cont"
06:51:31 <bhaak> Cryp71c: type "cont" at the gdb prompt?
06:51:39 <bhaak> he
06:51:51 <Cryp71c> :)
06:51:52 <Cryp71c> Excellent
06:52:29 <Cryp71c> ah, its segmenting at if(weapon
06:52:58 <Cryp71c> hrm, seems odd.
06:53:28 <kilobyte> "if (pointer)"?
06:53:38 <kilobyte> or was there something after weapon?
06:54:07 <Cryp71c> the former, there is more stuff after weapon in the if statement, but not on that line
06:54:16 <kilobyte> it very strongly smells like the object itself being null
06:55:04 <Cryp71c> gdb gives me "$1 = (item_def *) 0x0" when I print weapon
06:56:15 <kilobyte> odd, that if statement should accept it fine if that variable is accessible
06:57:56 <Cryp71c> yeah, and I didn't restructure any of this code, when I collapsed mons_to_hit and player_calc_to_hit I literally created a new function, wrote "if(attacker->atype() == ACT_PLAYER) { paste_of_player method } else { paste_of_monster_method }
06:57:58 <Cryp71c> "
06:59:16 <Cryp71c> (Just to make sure I wouldn't change any functionality or anything, it was only a temporary measure)
07:00:25 <kilobyte> hard to help if we can't see the code.  Could you do the commit and push it to a test branch ("git push origin HEAD:wip-jelly-crash")?  You can then commit --amend when it will work well.
07:01:19 <Cryp71c> kilobyte, this is already in my combat_control branch, I just noticed it as I was running through things this morning.
07:02:05 <kilobyte> ah, cool
07:03:06 <Cryp71c> Its something else, it looks like any unarmed combat, even player.
07:03:25 <Cryp71c> There must be some missing initializer code or something else wierd.
07:07:33 <Cryp71c> yeah its a null weapon pointer when fighting unarmed, the address is empty 0x0
07:10:05 <Cryp71c> But the code is getting through the constructor and init_attack, which use the same if(weapon code...
07:10:09 <Cryp71c> hrm. *ponders*
07:11:06 <kilobyte> player::shield_tohit_penalty() seems to assume there always will be a weapon
07:13:21 <Cryp71c> Ah, I see that. I can add an if(weapon) in there
07:14:06 <kilobyte> the equivalent function seems to be melee_attack::calc_player_armour_shield_tohit_penalty()
07:14:07 <Cryp71c> well, wp
07:14:18 <kilobyte> which does check for weapon
07:19:20 <Cryp71c> yeah I see that. k I've added a check for weapon there
07:20:00 <Cryp71c> having to recompile to see if it fixes the player unarmed combat.
07:20:06 <Cryp71c> My netbook is oh so slow.
07:23:34 <kilobyte> don't worry, I'm doing armel builds in qemu.  It really makes me want to pull out the phone and let them race :p  (trying a cross build would be better, but it'd be a bitch to do that with old libraries)
07:25:25 <Cryp71c> armel?
07:26:07 <Cryp71c> nvm, google.
07:26:38 <Cryp71c> I didn't know the DCSS source would build on ARM?
07:26:59 <Cryp71c> you could technically run it natively (without ssh'ing into a server) on an ipad then, couldn't you?
07:27:36 <kilobyte> not sure about ipad, but on n900 it works just fine
07:27:58 <kilobyte> (no tiles, though, there's no OpenGL, just OpenGLES)
07:28:25 <Cryp71c> gotcha, so...you've built DCSS on a n900?
07:29:01 <Cryp71c> That's quite cool.
07:29:51 <kilobyte> hard to come up with decent keybindings, though :(
07:31:01 <kilobyte> using vi keys might work, but then, vi is an abomination unto Xom
07:32:19 <Cryp71c> hehe, I rather like the VI keys. I've not found anything faster once you get used to the layout (whether it be for editing or crawlling)
07:32:36 <Cryp71c> nooo, I'm <10% battery. Will the build finish before I die?
07:34:32 -!- syllogism has joined ##crawl-dev
07:34:47 <Cryp71c> So any other ideas why the monster code is crashing on the check for weapon? Isn't that precisely what that check exists for, to make sure weapon isn't null before using it as an object?
07:35:14 <Cryp71c> hrm, I'll be back
07:35:17 <Cryp71c> Moving to office.
07:35:21 <kilobyte> it crashes the line before
07:39:26 -!- Cryp71c_ has joined ##crawl-dev
07:40:02 -!- Cryp71c has quit [Ping timeout: 260 seconds]
07:40:54 -!- Cryp71c_ is now known as Cryp71c
07:40:57 <Cryp71c> I return.
07:57:33 <Cryp71c> kilobyte: yeah with the change it now will get to the same place the monster code gets to, but then segfault.
08:00:22 <CIA-45> 03Cryp71c 07unified_combat_control * rb1bddcdb8f63 10/crawl-ref/source/player.cc: Fix segfault in shield_tohit_penalty
08:01:43 <kilobyte> uhm, you use hands unitialized
08:01:58 <kilobyte> not crashy, but still not good
08:03:44 <Cryp71c> hrm. I didn't want to presume to give it a value, would HANDS_ONE be better than uninitialized?
08:04:02 <Henzell> erisdiscordia the Kobold Catapult (L27 KoBe)  (Zig:16)
08:06:55 <kilobyte> no need to assign it at all
08:08:17 <CIA-45> 03kilobyte 07unified_combat_control * rb45bb96131a2 10/crawl-ref/source/player.cc: Fix use of an uninitialized variable.
08:15:09 <Cryp71c> kilobyte: thanks, I see how you did that.
08:28:29 <kilobyte> two more crashes
08:29:13 <kilobyte> both "cerebov v test spawner" (delay:0 t:99) and "cerebov v 50 place:Slime:6" appear to work now
08:30:21 <CIA-45> 03kilobyte 07unified_combat_control * rc240e73d4324 10/crawl-ref/source/fight.cc: Don't assert on the arena if the player is not involved.
08:30:41 <CIA-45> 03kilobyte 07unified_combat_control * r0239fc72d134 10/crawl-ref/source/melee_attack.cc: Fix a crash when fighting unarmed.
08:43:43 <kilobyte> there's a crash when a deep dwarf unborn dies, but not sure if that's a fault of the rewrite, lemme check
08:44:24 <kilobyte> about Mac builds: greensnark said "I'll get to it tonight EDT"
08:47:54 <Napkin> awesome :)
08:52:26 <Napkin> hey, shouldn't travel_delay = 50 & explore_delay = -1 make auto-explore lightning fast and shift+direction have 50ms delay?
08:52:45 <Napkin> if set like that, both travel and explore have 50ms delay
08:54:16 -!- valrus has joined ##crawl-dev
08:54:59 <kilobyte> no idea, having either != -1 is a heresy
08:55:02 <kilobyte> lemme check
08:56:16 <kilobyte> bool run_dont_draw = you.running && Options.travel_delay < 0       && (!you.running.is_explore() || Options.explore_delay < 0);
08:58:22 <Napkin> uhm
09:00:01 <Napkin> if running and travel_delay is -1 and if either not auto-autoexplore or explore_delay = -1, then don't draw
09:00:33 <kilobyte> yeah... and auto-explore is a form of running as this code is concerned
09:00:41 <kilobyte> whether it should be, is another question
09:00:54 <Napkin> anyways.. if travel_delay > 0, it'll always draw
09:02:49 <Napkin> bool run_dont_draw = (you.running && !you.running.is_explore() && Options.travel_delay < 0) || (you.running && you.running.is_explore() &&  Options.explore_delay < 0);
09:05:09 <Napkin> so, currently these settings only allow an additional delay for auto-explore, using travel_delay as base
09:05:47 <kilobyte> good point, it's strictly less powerful than allowing both would be
09:05:53 <Napkin> but.. travel is shift+directions _and_ ctrl+g, right?
09:06:11 <kilobyte> yes
09:06:17 <Napkin> that's probably why it's like that
09:06:36 -!- casmith789 has joined ##crawl-dev
09:06:38 <Napkin> make ctrl+g lightning fast and allow auto-explore to be slower
09:06:59 <Napkin> the writer didn't consider or didn't care about shift+dirs
09:08:38 <Napkin> there 3 kinds of running and only 2 options
09:09:46 <Napkin> better not touch it, kilobyte ;)
09:10:42 <Cryp71c> kilobyte: sigh, parenthesis fail.
09:11:26 <Zaba> hm.
09:11:47 -!- ophanim has joined ##crawl-dev
09:14:27 <CIA-45> 03kilobyte * r53f8530f88ad 10/crawl-ref/source/fineff.cc: Fix a crash when a monster with Injury Mirror dies.
09:14:56 <Cryp71c> kilobyte: did you fix the parenthesis already for the crash in calc_to_hit?
09:15:10 -!- ortoslon has quit [Read error: Connection reset by peer]
09:16:26 -!- ortoslon has joined ##crawl-dev
09:16:33 <kilobyte> just what I pushed a hour ago
09:16:49 <kilobyte> got to go, bye!
09:16:56 <Cryp71c> I just had a break from work and went to fix it, then found I was out of date.
09:17:01 <Cryp71c> I must've missed your pushes fixing it.
09:17:03 <Cryp71c> thanks!
09:19:35 <Napkin> ahh, back to master branch \o/
09:21:30 -!- ortoslon1 has joined ##crawl-dev
09:21:30 -!- ortoslon has quit [Disconnected by services]
09:24:44 -!- ortoslon1 has quit [Client Quit]
09:58:53 -!- Zaba has quit [Ping timeout: 250 seconds]
10:03:11 -!- Zaba has joined ##crawl-dev
10:20:14 -!- DrPraetor has quit [Ping timeout: 276 seconds]
10:22:01 -!- ais523 has joined ##crawl-dev
10:24:26 -!- edlothiol has quit [Ping timeout: 250 seconds]
10:32:10 -!- Wensley has joined ##crawl-dev
10:32:31 <dpeg> !seen galehar
10:32:32 <Henzell> I last saw galehar at Wed Apr 27 08:20:40 2011 UTC (7h 11m 52s ago) saying hi on ##crawl-dev.
10:38:50 -!- edlothiol has joined ##crawl-dev
10:39:01 -!- valrus has quit [Remote host closed the connection]
10:51:23 -!- Zaba changed the topic of ##crawl-dev to: Stone Soup Development | Logs: http://tozt.net/crawl | 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.
10:57:07 -!- Khaki has joined ##crawl-dev
10:57:07 <Napkin> and again.. a fair spell fails a million times in a row
10:57:07 -!- Ashenzari has quit [Ping timeout: 258 seconds]
10:57:42 -!- Ashenzari has joined ##crawl-dev
10:57:43 -!- The topic of ##crawl-dev is: Stone Soup Development | Logs: http://tozt.net/crawl | 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.
11:02:43 -!- edlothiol has quit [Quit: edlothiol]
11:02:55 -!- edlothiol has joined ##crawl-dev
11:05:14 -!- Mu_ has joined ##crawl-dev
11:09:17 -!- edlothiol has quit [Ping timeout: 248 seconds]
11:09:25 -!- edlothiol has joined ##crawl-dev
11:21:46 -!- MarvinPA has joined ##crawl-dev
11:46:32 <CIA-45> 03MarvinPA * r0127c7c6ab67 10/crawl-ref/source/dat/descript/items.txt: Update Scroll of Enchant Armour description
12:01:06 -!- edlothiol has quit [Remote host closed the connection]
12:01:26 -!- edlothiol has joined ##crawl-dev
12:07:16 <dpeg> announcement is up on r.g.r.m
12:07:31 <dpeg> I replaced " and ' glyphs, but forgot some :(
12:09:33 <Wensley> ??fair
12:09:33 <Henzell> fair[1/3]: No, this game isn't fair (and neither is "fair" success). But it was probably your own fault, anyway. P.S. Casting spells at Fair success means 30-50% chance to fail, which will annoy you at best or get you killed at worst. Frequently. You really want Great or better. Well, you really want excellent or better.
12:10:25 <Wensley> I often feel like giving players the percentile range of spell success would be more useful than the misleading, context-free descriptor they get currently
12:12:13 <dpeg> But where does it end?
12:12:18 <Wensley> right there. :)
12:12:38 <Wensley> (although I feel like the experience-level descriptors of player ghosts are equally useless)
12:16:05 <Wensley> I'd like to submit that a percentile range is imminently understandable, with well-defined and well-understood minimum and maximum values, with obvious application to the quality being described. As loath as we may be to expose numbers to players (yes, I'm loath as well), numerical spell success still makes more sense than, say, numerical AC, a number whose calculation is labyrinthine in...
12:16:07 <Wensley> ...complexity and obscure in application, with no well-defined min or max values
12:16:22 <dpeg> I don't believe that there's not a good word indicating 30-50% chance of failure. "Fair" may be misleading, okay.
12:17:39 <Wensley> Note that I'm not saying we should be giving players exact estimates of their spell success rates, I enjoy the ambiguity
12:17:51 <dpeg> yes, I note :)
12:18:55 <Wensley> my sentiment is informed by observation of the usage of the learndb, which I notice being used for two things more than most others: looking up the meaning of spell success rates, and looking up the meaning of player ghost experience descriptors
12:19:02 <dpeg> silly idea: we can track how often the spell succeeded over the last 100 tries and print the result: "69/100" :)
12:19:43 <dpeg> Wensley: good point. I am not aware of this anymore, since I left ##crawl screaming.
12:19:52 * dpeg shuns the players.
12:20:03 <Wensley> I understand :)
12:20:37 <dpeg> As Paul wrote to me once: Developing computer games is wonderful, if it were not for the players :)
12:21:09 <dpeg> (But he talked about commercial games, and we're well off with our players. It was just a matter of me not having time to keep explaining things.)
12:21:16 <ais523> the thing is, a 30% fail chance adds up quickly; if you need the spell to succeed to avoid death, then you only have a 70% chance of surviving, and 50% if you needed to use it twice
12:21:23 <Wensley> haha, well, one of the things that I love about stone soup is that it has a thriving community, and as with any community there will be differences of opinion and clashes of ego
12:21:39 <ais523> in NetHack, throwing Mjollnir is considered incredibly risky because of the 1% chance it screws up
12:21:51 <ais523> because it can kill you, eventually, and probably will if you do it enough
12:22:03 <dpeg> ais523: yes, although NH standards are slightly twisted, at least from a Crawl point of view :)
12:22:12 <ais523> indeed
12:22:18 <dpeg> ais523: by the way, what do you think of Karadoc's fork? And why are there no replies?
12:22:35 <ais523> it should be noted that NetHack, by design, is easier to recover from mistakes in than Crawl (it's a philosophy difference between the two games)
12:22:41 <ais523> and I'm not aware of Karadoc's fork
12:22:41 <dpeg> yes
12:22:58 <dpeg> he announced it today (with source and explanation) on r.g.r.nethack
12:23:07 <dpeg> bhaak: same question to you!
12:23:20 <Eronarn> < Wensley> I'd like to submit that a percentile range is imminently understandable, with well-defined and well-understood minimum and maximum values, with obvious application to the quality being  described.
12:23:22 <elliptic> ais523: mistakes like moving around while wielding a cockatrice corpse and falling into a pit? not sure that's easy to recover from :)
12:23:22 <ais523> ah, I checked rgrn yesterday, but not since then
12:23:25 <Eronarn> it's actually not
12:23:30 <ais523> elliptic: even that is "oLSable
12:23:49 <elliptic> well, in crawl there are felids ;)
12:24:06 <ais523> dpeg: hmm, it seems to have the exact opposite goal to my fork
12:24:13 <ais523> not that that's a bad thing; different forks will appeal to different sorts of players
12:24:23 <Wensley> Eronarn: how so?
12:24:32 <Eronarn> Wensley: you can take the same number, and present it several different ways, and people will get different impressions out of it
12:24:33 <dpeg> ais523: yes, as long as it's not going the *band path, forks are fine
12:24:57 <Eronarn> 'you succeed 90% of the time' 'you fail 10% of the time' 'you fail 1 time in 10' 'you are 9 times more likely to succeed than to fail'
12:25:01 <elliptic> dpeg: what is wrong with the *band path?
12:25:09 <dpeg> I believe the logarithmic approach to chances is better conveyed with words (this might be an instance of what Eronarn means)
12:25:17 <ais523> elliptic: there are so many bands nowadays it's hard to keep track of them
12:25:31 <dpeg> elliptic: don't you think that bands suffer from overabundance of variants?
12:25:54 <elliptic> not really, back when I was a devoted angband player I loved the variety
12:26:07 <ais523> and even among NetHack forks, there's a lot of sharing of code and ideas
12:26:27 <ais523> e.g. Un has Spork's level compiler and Ace's autoexplore and tutorial
12:26:38 <elliptic> bands do that also
12:26:41 <Wensley> Eronarn: it is not unreasonable to assume that people will see "spell success rate" (as long as it is labeled as such) and think to themselves that this number describes the rate (percentage) of times that the spell will succeed.
12:26:42 <ais523> and Ace has Slash'EM's build system, and some of the controls
12:26:59 <ais523> so as a result, the forks get better cooperative
12:27:11 <Eronarn> Wensley: [citation needed]
12:27:36 <dpeg> Wensley: my point is that 95% might be closer to "will always work" in a player'
12:27:42 <dpeg> s mind than the current verbal description
12:27:43 <ais523> dpeg: I think the reason there are no replies yet is that the sort of people on rgrn would need to test it out before they commented; and most of them probably don't use Windows 7
12:27:58 <dpeg> ais523: yes, probably... he has some good ideas in there
12:28:00 <Eronarn> fwiw i think the current descriptors suck
12:28:10 <ais523> dpeg: some of them are already in most of the other forks
12:28:17 <Eronarn> but this is just because they're not tuned to the fact that even very low failure rates are risky in a roguelike
12:28:19 <ais523> and at least one is actually in (the unreleased) NetHack 3.5
12:28:22 <Eronarn> not because of anything inherent to using descriptors
12:28:24 <CIA-45> 03dolorous * r9bbe62bb7b1d 10/crawl-ref/source/beam.cc: Properly make Yred's Enslave Soul return MON_UNAFFECTED when appropriate.
12:28:33 <Napkin> they don't suck
12:28:50 <Wensley> dpeg: I'd argue that the word "excellent" gives the player little useful information, and at least if they are given a numerical range it can be said that it is the player's fault for misunderstanding the number, rather than the interface's fault for misrepresenting it
12:29:35 <ais523> there's a huge difference between random chances that can be spammed until they work (e.g. victory dancing), and random chances that can kill you if they fail (e.g. spellcasting)
12:29:44 <Eronarn> if the choice of display the interface has for a number leads to people misunderstanding that number, how could you possibly blame the player for that rather than the interface
12:30:12 <dpeg> greensnark kept pointing out that I and II (spells display) are way too complicated
12:30:22 <Wensley> Eronarn: no idea what you're saying
12:30:27 <dpeg> He'd probably mention it right now if he were here.
12:31:33 <Wensley> I feel like the I screen is way too empty nowadays, is that because we're using the full names of the spell schools in the description of each spell?
12:31:45 <Wensley> (or rather, summary of each spell)
12:32:10 <Eronarn> Wensley: i'm saying that the interface choices you make will affect the comprehension of the players; a lack of comprehension is NOT the fault of the players, but of the interface design
12:32:49 <dpeg> I am not sure how much players reading up spell chances via learndb actually affects their actions.
12:33:12 <Wensley> Eronarn: seeing a number such as 95% and assuming that it means something other than what it does is a logical fallacy on the player's behalf
12:33:51 <dpeg> As I see it, players want as much information (ideally, all the numbers) because they believe that this will allow them to make better decisions. As designers, we need to think what information they really need. I tend to sit on the conservative side of things here.
12:33:53 <Napkin> Oh, I solely blame the RNG!
12:33:59 <dpeg> !coffee Napkin
12:34:01 * Henzell hands Napkin a mug of irish coffee, brewed by Mnoleg.
12:34:04 -!- blackpenguin has quit [Ping timeout: 248 seconds]
12:34:14 <Napkin> dpeg :)
12:34:21 * dpeg hands Napkin the badge: Faithful follower of Xom.
12:34:33 <MarvinPA> i couldn't tell you what very good/great/excellent actually convert to in terms of percentages, but i know that generally i want all my spells at great as a minimum, especially if it's something i'll rely on in dangerous situations
12:34:52 <MarvinPA> and that very good might be acceptable for some spells, and that obviously i want everything at excellent eventually
12:35:02 <Napkin> same here
12:35:06 <MarvinPA> knowing the actual percentage numbers wouldn't be any extra use
12:35:08 <doy> yeah, i don't think we need to use numbers, i think the descriptors just need to be more logarithmically scaled
12:35:11 <Napkin> nooooooo, not Xom!
12:35:14 <Eronarn> Wensley: please do some reading about this topic or something, you're ignorant of some very basic principles of communication
12:35:26 <doy> what you really care about is 95% vs 98% vs 99% vs 99.9%
12:35:34 <doy> not 50% vs 70%, etc
12:35:36 <Wensley> Eronarn: give me some links, or a keyword
12:35:38 <Napkin> oh, that badge fits yourself quite well too, Eronarn
12:35:42 <Eronarn> Wensley: 'statistical communication'
12:36:18 <MarvinPA> right, and that's been improved a bit recently by rolling all the low success rates (5-30%) into a single description
12:36:20 <doy> because yeah, no spell can really be relied upon unless it's at least at great
12:36:35 <MarvinPA> but there's probably room for more improvement, and i'm pretty sure there's some discussion about exactly this on the wiki
12:36:36 <doy> so we just have great/excellent/perfect for things that are actually relevant
12:37:25 <dpeg> later!
12:37:25 <doy> and numbers would be a terrible way to display that anyway
12:37:32 <doy> "99.13% chance"
12:38:01 <Napkin> ciao dpeg o/
12:38:10 <MarvinPA> hm, or maybe it's on an old mantis issue
12:39:10 <MarvinPA> https://crawl.develz.org/wiki/doku.php?id=dcss:brainstorm:interface:interface_ideas&s[]=adjectives#casting_success_adjectives
12:39:11 <Eronarn> Wensley: http://www.decisionsciencenews.com/2010/12/03/some-ideas-on-communicating-risks-to-the-general-public/ here's a good specific article
12:39:48 <Eronarn> Wensley: here's another good one that you should read at least the first few pages of (the linked PDF, not the page) http://e-patients.net/archives/2008/11/lies-damn-lies-and-statistics-collective-statistical-illiteracy.html
12:40:06 <Wensley> awesome, wasn't looking forward to buying a $200 book
12:44:12 <Eronarn> note that in one of the examples listed in the first link, when switching from percentages to natural numbers ("you fail 1 time out of 10"):
12:44:16 <Eronarn> 'Without any training whatsoever, 16 out 24 physicians obtained the correct answer to this version. That is quite a jump from 1 in 24.'
12:46:24 <Cryp71c> I've missed this entire conversation, but I've read back a full page. We're discussing whether spell success descriptors should be numerical or remain alphabetical?
12:46:29 <doy> to be fair, that's specifically in the context of conditional probability, which is more complicated
12:46:30 <Cryp71c> s/We're/You're
12:48:34 <Eronarn> doy: yes, but it's only meaningfully more complicated because we're pretty bad at using numbers in several ways that compound in that case :) and if we accept that we're bad at using numbers in some circumstances, we should be on the lookout for other cicumstances where that holds true
12:48:47 <doy> yes, i agree(:
12:49:38 <Zaba> maybe spell success should be boolean, then the problem disappears in its entirety
12:49:41 <ais523> hmm, do birthday paradox situations come up in roguelikes?
12:50:08 <Henzell> AlphaQ the Slayer (L21 HOPr)  (Crypt:5)
12:50:10 <ais523> that's another case where people's intuition tends to be wrong
12:51:10 <Eronarn> ais523: all the time. 'omg, i found X, what're the odds??'
12:51:35 <ais523> that's not birthday paradox, that's the more general large-number-of-unlikely-coincidences situation
12:52:01 <doy> so, 'omg, i found 2 of X, what're the odds?' then?
12:52:02 <doy> d:
12:52:05 <ais523> I think someone once calculated that there are so many possible noticeable one-in-a-million coincidences that one will happen to someone you know well every couple of weeks on average
12:52:24 <Eronarn> the part that makes the birthday paradox the birthday paradox is that people misunderstand the statistics when relying on intuition
12:52:56 -!- valrus has joined ##crawl-dev
12:53:08 <Eronarn> it's not limited to that specific value
12:53:19 <Eronarn> in fact, you can formulate the birthday paradox different ways - within a week, wthin two days, the same day
12:54:05 -!- blackpenguin has joined ##crawl-dev
12:54:25 <ais523> Eronarn: indeed
12:55:19 <ais523> the birthday paradox mostly just says that if you take approximately sqrt(n) choices out of n possibilities, the chance that two are the same are not very small, but around (not exactly) 50%
12:56:50 -!- Khaki has quit [Quit: Page closed]
12:57:32 <Eronarn> ais523: right, but that happens a lot in crawl :) e.g. the odds of finding two katanas, necronomicons, etc. in the same game are a lot higher than people would expect
12:57:49 <ais523> not any more, with the removal of katanas
12:57:52 <ais523> but I know what you mean
13:00:18 <Eronarn> a good approach to designing this might be something like:
13:00:29 <Eronarn> -get a solid list of adjectives
13:00:44 <Eronarn> -ask people to order them
13:01:17 <Eronarn> -use the ones with the most unambiguous placement (means spaced evenly, and with a low std dev)
13:01:35 <Eronarn> -apply these to a log-ish scale
13:01:38 <Wensley> whoa, I didn't read anything in the changelog about katanas being removed
13:02:16 <Eronarn> i think good breakpoints would be: 50%, 75%, 90%, 95%, 97%, 99%, 99.5%, 99.9%, 100%
13:02:48 <ais523> Eronarn: there should also be one at, say, 1%
13:02:55 <ais523> for spells where you only care about them working once
13:03:09 <ais523> e.g. for the brand spells, if you want to use them together with a scroll of vorpalise rather than in combat
13:03:14 <doy> for spells where you only care about them working once, we should remove the spell
13:03:17 <doy> d:
13:03:35 <ais523> doy: well, with the brand spells, you might either care about one use, or about them working constantly
13:03:39 <ais523> depending on what you're using them for
13:04:32 <Eronarn> i don't think it's worth figuring out whether the caster can succeed at all and displaying that as a separate category
13:04:55 <Eronarn> it's so incredibly narrow in application and we've been consistently trying to remove places where it would be useful
13:05:02 <Napkin> hey kilobyte? is your auto apt-get update still accessing the previous "lenny" folder?
13:06:09 <ais523> <Eronarn> it's so incredibly narrow in application and we've been consistently trying to remove places where it would be useful <--- I'm beginning to wonder if this attitude is the defining difference between Crawl and other roguelikes
13:07:06 <ais523> in most roguelikes, having options that are normally useless but can occasionally be used for some tactical advantage, in maybe one game in ten, is seen as a good thing, even if it rather promotes spoilers
13:09:09 <Eronarn> ais523: scroll of immolation, scroll of noise... :)
13:09:12 <doy> i could see that being true
13:09:25 <ais523> Eronarn: scroll of immolation's a good example
13:09:38 <ais523> is the scroll of noise louder than simple tt? (tt is another good example, by the way)
13:10:09 <Wensley> I use tt all the time, on all characters :P too lazy to go hunt down things to kill
13:10:56 <Wensley> where does evaporate fall in this line of reasoning?
13:11:18 <Henzell> AlphaQ the Slayer (L21 HOPr)  (Crypt:5)
13:12:01 <Eronarn> scroll of noise is louder
13:12:05 <Eronarn> it also breaks mesmerize
13:12:18 <ais523> Wensley: Crawl players seem split as to whether evaporate is useless or broken
13:12:25 <ais523> and I personally am not sure how good it is
13:12:36 <Eronarn> people who think evaporate is useless are horrible players
13:12:37 <Wensley> broken in practice or theory?
13:12:45 <Wensley> I love evaporate
13:13:06 <Eronarn> people would still use evaporate if it were a L5 spell
13:13:10 <MadCoyote> evaporate is the whole point of playing a transmuter for me
13:13:24 <MadCoyote> I like being able to use  those useless potions
13:14:13 <Wensley> I also like the trade-off of decreased inventory space with multiple effects from a single, cheap spell
13:15:42 <MarvinPA> scroll of noise breaking mesmerise annoys me somewhat in that it's automatically marked as a useless item
13:15:49 -!- jpt9 has left ##crawl-dev
13:17:27 <Zao> MarvinPA: They do? Why didn't anyone tell me :(
13:17:45 <MarvinPA> it probably shouldn't be, i guess marking it as "dangerous" like immolation would work, even if it's not quite comparable to immolation/torment
13:17:48 <MarvinPA> yeah
13:18:23 <upsy> does projected noise too
13:18:28 <MarvinPA> probably
13:18:47 <Wensley> ??projected noise[2]
13:18:48 <Henzell> projected noise[2/5]: If you are under the mesmerised status, casting this spell is loud enough to break you out of the effect. When cast at a wall, you'll hear a dull thud. This can be useful for figuring out where unexplored areas remain.
13:19:02 <MarvinPA> oh yeah, haha
13:19:10 <MarvinPA> i want someone to write a bot using projected noise to map levels
13:19:23 <MarvinPA> because that'd be awesome
13:20:02 <MarvinPA> (except not really :P)
13:20:32 <Eronarn> remove projected noise already
13:20:39 <ais523> MarvinPA: that's as bad as kerio's stethoscopebot on NetHack, that made use of the fact that a) stethoscopes map one adjacent square, at the cost of half a turn, and b) saving and restoring causes the game to lose track of how many halfturns you've taken
13:21:00 <Eronarn> awesome
13:21:26 <Wensley> I think projected noise has the capability to be a very interesting spell, but I don't know how much work it would take to get it there
13:21:43 -!- ophanim1 has joined ##crawl-dev
13:22:38 <Eronarn> it really doesn't; we shouldn't have spells that affect out-of-los in such dramatic ways (if we should even have any at all)
13:22:58 -!- ophanim has quit [Ping timeout: 250 seconds]
13:23:06 <ais523> why is Crawl so dependent on only the LOS being relevant?
13:23:08 <Wensley> I mean more like spells to misdirect monsters
13:23:11 <ais523> because that makes it easier to balance?
13:23:31 <Eronarn> ais523: pretty much
13:23:41 <Eronarn> being able to fire out of LOS is really broken in a lot of games
13:24:02 <doy> ais523: because the only thing relevant to monsters is los
13:24:31 <doy> so having the player be able to affect monsters without the monsters being able to respond in any way tends to be broken
13:24:33 <Ashenzari> Asked about shopping list after game end. (https://crawl.develz.org/mantis/view.php?id=3885) by elliptic
13:24:39 <ais523> hmm, I think NetHack deals with the issue by having infinite LOS
13:24:43 <doy> right
13:24:54 <ais523> so something is only out-of-LOS if there's a wall or something in the way
13:24:55 <upsy> that bug is still around huh
13:26:03 <Wensley> Eronarn: why do lava orcs have -4 stealth, +2 stabbing?
13:27:33 <Eronarn> Wensley: they're made of stone, but orcs are good at stabbing
13:29:00 <Wensley> what situations could they stab something in, if they can't cast spells to confuse or sleep?
13:29:21 -!- effo has joined ##crawl-dev
13:30:22 <doy> wands, needles, other paralysis sources, distractions by allies, stunning with clubs
13:31:04 <Eronarn> invisibilityt
13:31:05 <Wensley> or maybe they could memorize all their spells while in ice form :P
13:31:16 <Eronarn> what?
13:31:35 <Wensley> the reason they can't read books is because they're too hot, yeah?
13:31:48 <Eronarn> they can't read books *when* they're hot
13:31:58 <Eronarn> they aren't always hot
13:32:01 <Wensley> ah, I read the mutation screen wrong then
13:32:09 <Wensley> I figured they could never cast spells, ever
13:32:10 <Eronarn> greyed out = inactive
13:32:46 <Eronarn> try looking at the second A screen, with !
13:33:06 <MarvinPA> i think they should work the same way as vampires and only show active mutations on A
13:33:15 <MarvinPA> rather than still have them there but greyed out
13:33:32 <Eronarn> yeah, it's something i've been considering
13:33:58 <effo> silly question but what would I enter at command line to have gcc compile all the crawl source?
13:34:00 <Wensley> whoa, hilarious temperature status screen
13:34:01 <Eronarn> i'm also thinking of dropping temperature changes from elemental damage
13:34:13 <Eronarn> effo: make, make wizard, make debug, etc.
13:34:15 -!- Zaba has quit [Ping timeout: 250 seconds]
13:34:25 <Eronarn> -jN to make it use more cores if you have them
13:34:30 <effo> and I just do that from the directory with all the source files?
13:34:40 <Eronarn> yeah (crawl-ref/source)
13:34:47 <Eronarn> you will probably get an error about submodules
13:34:51 <Eronarn> but it will tell you how to fix that
13:34:52 <effo> alright
13:35:51 <Wensley> is there any lava magic besides bolt of magma? I've always wanted a lava fists spell...
13:35:57 <Eronarn> not yet
13:36:16 <Eronarn> i'll probably drop the lava magic boost unless more get added
13:36:29 <effo> what does it mean when git "initializes empty git repositories" a hundred times
13:38:39 <Wensley> man, tension is the coolest mechanic
13:38:49 -!- Zaba has joined ##crawl-dev
13:39:03 <doy> "if only it worked"
13:40:49 <Wensley> why doesn't tension work?
13:41:23 <elliptic> lava orcs basically never get high temperature after early game currently unless you step in lava
13:43:27 <effo> I have gcc but got an error at the end of making saying 'i686-linux-gnu-g++' wasn't found
13:43:44 <doy> effo: your os probably packages gcc and g++ separately
13:43:56 <effo> more practice for me!
13:44:02 <doy> you should figure out what package g++ is in
13:44:08 <effo> my package manager somehow stopped working at the same time as my file explorer
13:44:12 <effo> so I get to term it up
13:44:38 <effo> I just did sudo apt-get install g++
13:44:57 <effo> funny how many programs you can just kinda guess with that command and it just works
13:49:01 <effo> package.h:12: fatal error: zlib.h: No such file or directory
13:49:01 <effo> compilation terminated.
13:49:25 <effo> this didn't show up when I tried without having g++, it only pops up now, immediately after I try to make
13:54:35 <Cryp71c> effo: you also need the zlib-dev lib
13:54:44 <Cryp71c> ??crawl source
13:54:46 <Henzell> I don't have a page labeled crawl_source in my learndb.
13:54:47 <Cryp71c> hrm, what's the learndb entry...
13:54:49 <Cryp71c> ??build
13:54:52 <Henzell> I don't have a page labeled build in my learndb.
13:55:05 <Cryp71c> ??make
13:55:07 <Henzell> I don't have a page labeled make in my learndb.
13:55:10 <Cryp71c> curses!
13:55:11 <effo> I have henzell ignored. can I grab that as a package like normal?
14:00:18 <effo> found the entry, got the stuff
14:00:20 <effo> thanks Cryp71c
14:01:20 <Cryp71c> np, I added it in there a couple weeks ago.
14:02:26 <effo> I basically have to make again any time I make a change to a source file, right? i.e. I just noticed that my first change had unbalanced parentheses and fixed/saved the file
14:05:56 <Cryp71c> effo: yes, anytime you make any changes, you need to rebuild the source. Once "crawl" exists, it can be moved around, deleted, emailed, distributed, etc. (it no longer even cares about the source), so by that logic the source also has nothing to do with the binary you're running.
14:06:10 <Cryp71c> So yes, you have to rebuild the executable to have changes you've made be reflected in the binary.
14:06:43 <effo> I grasp that I was just hoping that there was some sort of time saving measure for things I haven't modified since last compile
14:08:44 -!- valrus has quit [Remote host closed the connection]
14:09:10 <Cryp71c> yeah, unfortunately not.
14:10:41 <doy> effo: "make" itself is a time saving measure for things you haven't modified since last compile
14:10:51 <doy> it shouldn't recompile anything you haven't touched
14:12:08 <doy> the caveat there being that touching a header file counts as touching every file that includes that header, because of how c++ works
14:12:32 <MarvinPA> hmm, i feel like equipped randarts should be coloured the same as other equipped items by default
14:12:55 <MarvinPA> once you have it equipped, "this is an artefact" isn't particularly useful information compared to "this item is equipped and the item below it isn't"
14:13:17 <Cryp71c> MarvinPA: I regularly use the coloring to identify corrodable items and whatnot.
14:13:27 <Cryp71c> both in inventory and the % screen
14:17:38 <upsy> well it's not only artefacts that have that property
14:18:40 <st_> equipped stuff -> green, equipped artefacts -> lightgreen
14:19:24 <MarvinPA> well, green already means "uncursed" at the moment
14:19:54 <st_> I got rid of all that stuff, hate it
14:20:26 <MarvinPA> yeah, i'm increasingly removing more and more of the default "include = asdf.txt" lines from my .rc and reimplementing their effects myself :P
14:21:32 <MarvinPA> random tangent: does anyone know how bind function keys to a command with bindkey?
14:21:32 -!- monqy has joined ##crawl-dev
14:21:57 <MarvinPA> bindkey = [F1] didn't work, and nor did bindkey = [ \{27}[11~] which is the code that comes up when creating macros
14:29:49 <casmith789> \{32}
14:31:25 <casmith789> hey weird everything has been changed into negative numbers
14:31:55 <casmith789> \{-1013} for f3 for some reason
14:32:36 <MarvinPA> i think those codes depend on the system, anyway
14:32:45 <MarvinPA> and either way bindkey seems not to like them
14:44:35 <effo> I'm getting scope errors from libunix.cc now, does that sound familiar to anyone?
14:49:00 <Cryp71c> effo: that's another library, a newer one that was added in the last couple weeks or less, I think.
14:49:22 <Cryp71c> effo, what's one of the funcitons?
14:49:24 <effo> ah do I need to get a new version or something
14:49:36 <effo> let me look back one sec
14:50:19 <effo> i'll let you know when it finishes
14:50:30 <effo> I used that term to start a crawl session and can't find the scrollback
15:03:50 -!- syllogism has quit []
15:05:51 <elliptic> that sounds like the issues I was having with libncursesw
15:12:07 <effo> libunix.cc: In function ‘int getchk()’:
15:12:18 <effo> int putwch(ucs_t)
15:12:51 <effo> char_info character_at(int, int)
15:13:19 <doy> libncursesw-dev, probably
15:14:00 <effo> what to do about it?
15:14:38 <doy> apt-get install libncursesw-dev?
15:17:38 <effo> oh I see
15:17:48 <effo> I must've not gotten the right libncurses
15:18:18 <Cryp71c> No, you also need zlib, you need several libraries now.
15:18:26 <effo> yeah I got zlib
15:18:33 <Cryp71c> libncurses-dev, zlib-dev, and libncursesw-dev
15:18:43 <Cryp71c> The first and 3rd might just be diff packages of the same lib.
15:18:45 <Cryp71c> IDK
15:18:53 <effo> the libncursesw-div package isn't listed for my source I guess
15:19:30 <effo> and I got the first and second you listed
15:19:43 <doy> -dev
15:19:53 <effo> I typed it in right in the term :(
15:19:54 <effo> haha
15:20:04 <Cryp71c> try to build it now, but I'm pretty sure you need the libncursesw package.
15:20:22 <effo> me too because I had all the stuff but that this whole time
15:20:30 <effo> now I need to find it
15:20:51 <Cryp71c> -dev*
15:20:53 <Cryp71c> sorry :P
15:21:06 <effo> ah it's w5
15:21:27 <effo> libncursesw5-dev
15:21:45 <effo> ok makin'
15:23:42 <effo> alright, it seemed to make properly, now how can I run the resulting 'crawl' file?
15:23:52 <effo> ./crawl?
15:24:05 <Cryp71c> yes
15:24:12 <effo> dunno why I asked lol silly question
15:24:40 <effo> wow so much faster than online haha
15:29:23 -!- ibanix has joined ##crawl-dev
15:32:22 <CIA-45> 03Cryp71c 07unified_combat_control * ree3bd0d9b0e1 10/crawl-ref/source/ (fight.cc melee_attack.cc melee_attack.h): Externalize monster attack rounds to fight.cc
15:33:30 <Cryp71c> moving to bus. bbl
15:33:32 -!- Cryp71c has quit [Quit: Leaving]
15:34:14 <ibanix> new hacker Q: why are there two read_options functions, one with FileLineInput and one with StringLineInput?
15:37:30 -!- bmh has joined ##crawl-dev
15:37:36 -!- bmh has quit [Changing host]
15:37:36 -!- bmh has joined ##crawl-dev
15:37:45 <bmh> hello
15:38:07 <effo> so adding lua hooks for current and max burden was remarkably easy
15:39:02 <effo> only issue is that the values the game keeps are 10x your actual current/max burden
15:39:06 <effo> I suppose to avoid decimals?
15:39:10 <doy> yes
15:39:31 <effo> should I make it give that value and let people deal with it in their .rc or should I compensate for that
15:40:03 <effo> might as well test both ways
15:45:02 <effo> so to submit this for approval, I use git to make a patch, then upload that on mantis?
15:46:02 -!- Cryp71c has joined ##crawl-dev
15:46:09 <Cryp71c> Alrighty, on the bus.
15:46:11 * Cryp71c rides the bus.
15:46:34 <Cryp71c> !seen kilobyte
15:46:35 <Henzell> I last saw kilobyte at Wed Apr 27 14:16:48 2011 UTC (6h 29m 47s ago) saying got to go, bye! on ##crawl-dev.
15:46:44 <bmh> Cryp71c: where are you headed?
15:47:01 <doy> effo: yes
15:47:07 <doy> git format-patch -1
15:49:21 <Cryp71c> bmh, back home from work.
15:49:41 <Cryp71c> hrm, my changing around of attack_rounds seems to provide monsters the inability to miss.
15:50:32 <effo> next question, I modified the init.txt while I was testing, and don't want to include that in the patch. How can I make the patch without that change?
15:50:58 <effo> It was included in a commit I just made, I guess. I suppose I could pull a fresh copy and only change 1 file before I commit
15:51:04 <effo> local* commit
15:52:52 <Cryp71c> effo, you should be able to look in git log and see the most recent commit not done by you. Copy the first 8 or 9 chars of its hash (say its abcdefghijk), then do "git checkout abcdefghijk -- fileName.cc"
15:53:11 <Cryp71c> make sure you don't go too far back, or else you might reset someone's changes.
15:53:15 <effo> ah ok
15:53:18 <Cryp71c> Only go back to just before your commits
15:55:03 <effo> it's looking in crawl.ref for the file
15:55:12 <effo> the file I changed was init.txt in settings/
15:55:55 <effo> urf I just want to clear all my changes and delete the local commit so I can start over what's the fastest way for that
15:56:19 <Cryp71c> git reset hash --hard will reset all files to the hash.
15:56:40 <Cryp71c> git reset hash --hard -- file might work to reset just the file.
15:58:00 <effo> alright thanks
15:58:28 <effo> is hash the word 'hash' or an actual hash
15:58:58 <ais523> it's the hash itself, or the first few nybbles of it if that's unique
15:59:13 <ais523> Cryp71c: I think you can't hard reset an individual file in git, which is a little ridiculous
15:59:15 <effo> ok, and I get that from looking at the log of commits like he mentioned
15:59:51 <doy> ais523: that's just not true at all
16:00:04 <Eronarn> ais523: 'git checkout filename'
16:00:09 <doy> it's just called - yeah, what Eronarn said
16:00:10 <doy> (:
16:00:31 <ais523> doy: OK, you can, but it uses an entirely different command
16:00:50 <Eronarn> because conceptually it's not the same thing
16:01:36 <doy> thinking about git is a way that's not how git actually works does tend to be confusing, yes
16:01:39 <doy> d:
16:01:39 <doy> *in
16:01:50 <Eronarn> reset: make head point to a commit
16:02:02 <Eronarn> checkout: grab files from a commit, or file(s) from the current commit that head is on
16:04:25 <effo> so in order to properly make the patch, I should have only the file I'm patching listed when I do git commit -a
16:04:41 <effo> and then do git format-patch -1
16:04:46 <Eronarn> effo: 'git add filename', 'git commit' lets you commit specific files
16:04:49 <Eronarn> rather than having to do -a
16:05:12 <effo> what does the -a flag actually do?
16:05:15 <Eronarn> -all
16:05:18 <effo> heh
16:05:19 <effo> ok
16:05:33 <Eronarn> also, git format-patch -1 will grab the last commit (that's what the -1 means) and turn it into a patch
16:05:55 <Eronarn> so you have your commit to a file you don't want in your patch; if you do a new commit, then -1 it, you only get the changes made in the second commit
16:06:33 <Eronarn> so consider: a file with contents 'A\nB\nC'. you change B=>Q and commit it, then add \nD and commit that
16:06:45 <Eronarn> a patch from the first commit would be B=>Q
16:06:54 <Eronarn> a patch from the second commit would be +\nD, but *not* include B=>Q
16:07:07 <effo> interesting
16:07:33 <Eronarn> if you wanted B=>Q and +\nD, you'd have to do two patches (format-patch -2), or put them in the same commit, or squash commits
16:08:08 <Eronarn> the reason this works is that +\nD does not depend on B=>Q
16:08:16 <effo> when you make a patch with git will it always put it in your working directory
16:08:31 -!- Cryp71c has quit [Ping timeout: 250 seconds]
16:08:38 <doy> format-patch will, yes
16:08:41 <Eronarn> if the second commit were Q=>Z, you couldn't apply just the second patch without a conflict
16:09:07 <effo> alright, submitting my first patch
16:09:08 <effo> hooray
16:09:13 -!- Textmode has joined ##crawl-dev
16:09:49 -!- jpt9 has joined ##crawl-dev
16:10:01 <Eronarn> ah, i remember back in the day when i submitted my first patch...
16:10:05 <Eronarn> oh wait that was like 6 months ago.
16:10:42 <effo> I just pulled from master branch a few minutes ago, what should I list for Product Branch
16:10:48 <effo> probably the latest hm
16:11:00 <ais523> if it's from master, put 0.9
16:11:04 -!- Cryp71c has joined ##crawl-dev
16:11:22 <ais523> and it might help to say the exact version you pulled, in case it matters
16:11:31 <ais523> although for a patch, as opposed to a bug report, it probably doesn't
16:13:52 <effo> https://crawl.develz.org/mantis/view.php?id=3886
16:14:03 <effo> (I won't do that every time)
16:14:43 <doy> if you wait a minute or two, Ashenzari will do that for you, automatically
16:14:43 <doy> (:
16:14:56 <effo> haha ok
16:15:00 <effo> jumped the gun
16:15:35 <effo> so, since I can't just commit directly to the online repository, what should I do now to put things back in equilibrium
16:15:41 <effo> or anytime I make a patch to upload
16:16:01 <Ashenzari> Added Lua hooks for player's current/max burden (https://crawl.develz.org/mantis/view.php?id=3886) by effo
16:16:07 <effo> just checkout all the files I changed individually, do a rest like above, or what
16:16:10 <Cryp71c> effo, you can undo your local changes by "git reset origin/master --hard"
16:16:16 <effo> awesome
16:16:17 <jpt9> Hey.  Any idea when the 0.8 OS X binaries will be up?
16:16:21 <CIA-45> 03Cryp71c 07unified_combat_control * rdc95d1435f0f 10/crawl-ref/source/ (fight.cc melee_attack.cc melee_attack.h): Revert "Externalize monster attack rounds to fight.cc"
16:16:29 <doy> jpt9: just as soon as we figure out where greensnark is
16:16:30 <doy> (:
16:17:09 <effo> Cryp71c: I did that but my changes are still there
16:17:10 <Cryp71c> Gah, the handling of multiple monster attacks is a mess.
16:17:16 <Cryp71c> effo, did you add a file?
16:17:20 <Eronarn> effo: for the future, it's best to start a branch when you do stuff
16:17:25 <effo> no I just modified one
16:17:30 <effo> l_you.cc
16:17:30 <Cryp71c> Eronarn, I'm not sure his small changes warrant a branch.
16:17:40 <Cryp71c> effo, any output when you reset to origin/master ?
16:17:54 <jpt9> doy: Ah.  Okay.
16:17:56 <effo> oh is it exactly origin/master?
16:18:00 <effo> i used 'master'
16:18:01 <Eronarn> 'git checkout -b <name of your branch>', then do your stuff, and when you want to get back you do 'git checkout master' (optionally you can delete the branch with 'git branch -D <branchname>'
16:18:02 <effo> lemme try again
16:18:08 <doy> effo: yes, origin/master and master are different
16:18:21 <doy> effo: "master" is your copy, which includes your patch
16:18:23 <Eronarn> effo: you made your commits to master, because you didn't do them in a branch :)
16:18:28 <doy> "origin/master" is the copy on the remote server
16:18:42 <effo> oh ok that makes sense
16:18:48 <effo> so reset master is my version of master
16:18:52 <Cryp71c> effo, yes.
16:18:53 <effo> versus origin/master
16:18:56 -!- ophanim1 is now known as ophanim
16:19:02 <Eronarn> in particular, you're resetting your version of master to your version of master
16:19:08 <Eronarn> (which is why the changes were still there)
16:19:14 <effo> very useful!
16:19:41 <bhaak> dpeg: manche leute haben auch noch ein leben ausserhalb des netzes!
16:19:42 <effo> and then from here someone else looks at my patch and commits it if they like it
16:19:53 <Eronarn> Cryp71c: imo it's best to make a branch for anything that isn't a one-line text fix
16:19:56 <Eronarn> it's so easy to do
16:20:18 <effo> so I could just git checkout -b effo
16:20:21 <effo> and that makes a branch
16:20:25 <Eronarn> yep
16:20:26 <Cryp71c> yes
16:20:58 <Eronarn> think of a branch as a fork off of whatever branch you're currently on (master is itself a branch)
16:21:06 <Cryp71c> Eronarn, isn't there a -t for branch tracking + implied branch creation, vs just -b (for branches which are getting pushed up to origin)
16:21:34 <Eronarn> so if you checkout origin/master, then checkout effo, then make changes, then checkout effo2 - effo2 will have, in its history, the changes you made to effo
16:21:40 <effo> through all this I only have one copy of the files on my system, right?
16:21:59 <Eronarn> right; the differences are tracked in a .git directory
16:22:16 <Eronarn> by applying the differences in sequence, it can update your copy of the files to newer or older versions
16:22:26 <Cryp71c> effo, fear not, you don't have commit access, so the worst you can do is muddy your own git history.
16:22:33 <effo> yeah lol
16:22:40 <Eronarn> Cryp71c: i haven't used -t before, no idea what it does
16:22:50 <doy> effo: http://ftp.newartisans.com/pub/git.from.bottom.up.pdf this is not too long, and does a really good job of explaining what's actually going on
16:22:58 <effo> thanks doy
16:23:07 <bhaak> dpeg: it has some nice features, although it's mostly balance stuff. unfortunately, it seems there's no source repository, he started from (buggy) vanilla version and has a stupid name ;-). also unfortunately, no big interface improvements and that's not in his scope
16:26:14 <Cryp71c> Eronarn, the man page's description is a bit cryptic, but it seems useful.
16:27:35 <Cryp71c> It makes is so that when I'm in my branch, I can git pull without specifying anything and it will pull both my branch (and master, which always fails because of fast-forward)
16:31:18 <Eronarn> ah
16:31:32 <Eronarn> does it do that every time you check out the branch, too?
16:31:35 <Cryp71c> Though the man says something about upstream and whatnot.
16:31:57 <Cryp71c> Eronarn, not that I've noticed, but that may just be because there's no one else really committing to the branch.
16:31:59 <Eronarn> since i only really do git locally, so i'm not doing git pull <branch>... or maybe i'm misunderstanding how it works
16:33:16 <Cryp71c> Eronarn, well, afaik, when you git checkout -t branch_name it doesn't fork the current branch to a new local branch called "branch_name"
16:33:36 <Cryp71c> It actually looks to origin for a branch called "branch_name" and registers it as a remote branch, instead of a regular local branch.
16:33:59 <Cryp71c> Thus, when you're inside the branch, "git push" and "git pull" are sufficient since the git config already knows the affiliation with the remote server.
16:34:14 <Cryp71c> Though I'm sure there's more going on.
16:34:57 <effo> still don't quite get git but I'll get there
16:36:30 <Eronarn> effo: the best way to get used to it is to use it
16:36:36 <Cryp71c> gah the monster side of combat code is grinding me. Theres no clear exit paths, afaik the control flow of when attack(s) end is somewhere in monster energy, which makes sense, but is rather elborate.
16:36:45 <Eronarn> reading guides and such helps but you also need hands-on experience
16:37:16 <effo> yeah I'll try to get some practice in
16:37:22 <Eronarn> Cryp71c: huh? maybe i'm misunderstanding what you mean by attacks, but that's handled in fight.cc
16:37:28 <effo> so far though adding burden to the available lua hooks was all I could think of
16:37:57 <Eronarn> monster attacks -> monster gets multiple attacks -> monster's attacks are looped through -> loop can break if the monster has nothing left to attack and isn't able to change targets (hydra)
16:38:55 <Eronarn> i don't see how monster energy would play into that - monster energy would determine if a monster got to make multiple attack sequences, but not what actually happens inside of an attack sequence
16:39:33 <Eronarn> effo: implementing a new spell is a fun way to learn the code because you have to look all over the place
16:39:47 <Cryp71c> Eronarn, that's what I thought too, but in my rewrite I moved the looping of attacks outside of the melee_attack class (eg, each seperate attack gets its own instantiation of melee_attack) and found that even monsters that don't get multiple attacks (rats) were getting 4 attacks per turn because my rewrite had shifted some code around.
16:40:00 <Cryp71c> Eronarn, potentially including some energy-loss code, but its not quite that simple.
16:40:21 <Eronarn> Cryp71c: if you link to it on gitorious i can take a look and see if i can spot anything amiss
16:40:38 <Cryp71c> Eronarn, I've since revertted the commit, let me find it real fast.
16:40:49 <Cryp71c> Its kind of a big (slightly crude) change, 1 sec.
16:42:04 <Cryp71c> Eronarn, http://gitorious.org/crawl/crawl/commit/ee3bd0d9b0e1f18ccaff23ef32e213535de56bf6
16:42:21 <Cryp71c> I'm about to get off the bus, so if you find anything, please !tell
16:43:41 <Cryp71c> Eronarn, you will notice in the code which handles the attack rounds, its hard coded to have the number of attacks at "4" so there's something else which is ending the attack sequence before all 4 attacks are actually performed.
16:43:51 <Cryp71c> Whatever ends the attack sequence is what I've overlooked.
16:43:52 <Cryp71c> BBL
16:48:25 -!- Cryp71c has quit [Ping timeout: 252 seconds]
16:54:03 -!- galehar has quit [Remote host closed the connection]
17:02:51 -!- bmh has quit [Quit: bmh]
17:15:01 <due> morning
17:16:41 <ophanim> Question: I'm trying to grab the value of specific skills for a lua script using you.skill(), but I don't know how to point it at specific skills.  Anyone work with it before?
17:17:04 <due> ophanim: you.skill("throwing")?
17:17:14 <due> ophanim: if you have a wizard build, use the in-built interpreter to play around
17:18:08 <ophanim> due: Thanks!  I was missing the quation marks.
17:18:38 <due> er
17:18:43 <due> you don't need the question mark
17:18:46 <due> i was asking if you'd tried that
17:18:56 <due> otherwise chck the source code for str_to_skill :)
17:21:29 <ophanim> you.skill(throwing) returns 0 while you.skill("throwing") returns the correct value.
17:22:01 <ophanim> So apparently I do?   Not to say you're wrong.  I am a serious code noob.
17:22:29 <eith> theres a mix up between quotation marks and question marks i think there :p
17:22:56 <ophanim> maaaaaaybe
17:23:11 <due> erm
17:23:16 <due> you.skill("throwing") is correct
17:23:19 <due> ? = question mark
17:23:22 <due> " = quotation marks
17:23:28 <due> both where in the statement i made to you
17:23:50 <ophanim> quotation, not quation.  Typed too fast.
17:24:54 <due> ah
17:25:08 <due> my auto-correct spelling assumed the more common of the two, whoop
17:27:26 <ophanim> I accidently the word.
17:51:57 <effo> thanks for the tip about the built-in wizmode interpreter
17:58:18 <ophanim> How does the interpreter work?
18:00:58 <Napkin> geez.. fucking grinder
18:07:56 <due> Napkin: :DDDD
18:08:05 <due> ophanim: you type lua in, it gets executed, TADA
18:09:38 <ophanim> ah
18:12:35 -!- ais523 has quit [Remote host closed the connection]
18:18:47 <effo> due: can you use variables in the interpreter? ie x = foo; crawl.mpr(x)
18:19:15 <due> yes
18:19:25 <due> but you can just use print
18:19:28 <due> instead of mpr
18:21:58 <effo> if I change files and commit them locally, what happens if I pull from master online?
18:22:56 <due> you'll merge them in
18:23:01 <due> better to pull --rebase
18:23:07 <due> which sticks your commits on top of the server
18:23:11 <due> for pretty cleanness
18:23:21 <effo> ah ok
18:28:18 -!- Mu_ has quit [Quit: Defecator, may everything turn out okay so that you can leave this place.]
18:29:23 -!- ophanim1 has joined ##crawl-dev
18:30:39 -!- ophanim has quit [Ping timeout: 250 seconds]
18:31:24 -!- Twilight-1 has quit [Read error: Connection reset by peer]
18:31:52 -!- Twilight-1 has joined ##crawl-dev
18:34:17 -!- edlothiol has quit [Ping timeout: 240 seconds]
18:49:48 -!- edlothiol has joined ##crawl-dev
19:10:13 -!- ophanim1 is now known as oph
19:12:54 -!- edlothiol has quit [Quit: edlothiol]
19:13:13 -!- edlothiol has joined ##crawl-dev
19:22:48 -!- Wensley_ has joined ##crawl-dev
19:24:45 -!- Wensley has quit [Ping timeout: 248 seconds]
19:24:54 -!- Wensley_ is now known as Wensley
19:34:18 <ZorbaTHut> hey, is there a website dev around who can take an incoming link off of http://crawl.develz.org/wordpress/? I post my blog entries to two places and they both showed up on the "incoming links" list, I swear I'm not trying to spam your site :) I'd love for the second link, from "Zorba the Hutt", to be removed
19:38:36 <due> ZorbaTHut: hi
19:38:42 <due> ZorbaTHut: give me a moment to log in
19:39:19 <ZorbaTHut> cool, thanks :) I felt like kind of a jerk when I saw that
19:40:10 <due> it's fine!
19:42:14 <Ashenzari> White bars on tiles (https://crawl.develz.org/mantis/view.php?id=3887) by Mental Mouse
19:42:36 <due> I can't actually work out how to do so
19:43:10 <due> Napkin: can i delete incoming links or do i need to be an admin?
19:43:44 <Napkin> incoming link?
19:44:22 <due> the pingback/trackbacks
19:44:31 <due> i know i get the emails for approving/disapproving them
19:45:12 <Napkin> they are listed in the comments, i think
19:45:51 <Napkin> show me which one exactly
19:46:10 <Napkin> because, i disabled pingback/trackbacks, i think
19:46:42 <ZorbaTHut> napkin, it's the one from "Zorba the Hutt", pointed at http://zorbathut.livejournal.com/827037.html
19:47:10 <ZorbaTHut> (the other one, from "Zorba", I'd like to keep around)
19:47:30 <Napkin> that's neither pingback nor trackback
19:47:38 <Napkin> it's live-search result from google
19:48:28 <ZorbaTHut> hrm, no way to suppress that? I feel sort of bad for using two slots there :/
19:48:34 <Napkin> tip: change one posting to point to http://crawl.develz.org instead of http://crawl.develz.org/wordpress
19:48:47 <ZorbaTHut> aha, I'll try that
19:49:35 <ZorbaTHut> done, I assume it'll take some time to update on your site
19:50:44 <Napkin> google needs to update their result
19:51:21 <ZorbaTHut> yup. alright, thanks :)
19:51:39 <Napkin> np :)
19:53:20 <Napkin> well, how to skip an iteration in a php foreach construct?
19:54:42 <Napkin> "continue"
19:56:26 <Eronarn> a thing that php does that i really like: 'break' to break out of a loop, 'break 2' to break out of a loop and the loop above that, etc.
19:57:08 <ZorbaTHut> I've always liked the idea of named breaks - it seems less fragile than "break 2"
19:57:23 <Napkin> ok, I filtered it out for now
19:57:26 <Napkin> g'night o/
19:57:45 <ZorbaTHut> thanks, g'night :)
20:05:29 -!- elly has joined ##crawl-dev
20:07:26 <greensnark> Yay 0.8
20:07:27 <Henzell> greensnark: You have 5 messages. Use !messages to read them.
20:09:52 -!- eith has quit [Ping timeout: 258 seconds]
20:18:05 <due> hi snarky
20:18:32 <elly> mwahahahahaha
20:20:27 <due> felids have bad unarmed combat apts?
20:20:35 <due> oh, who cares
20:23:07 * due shiftily edits 0.8 announcement again
20:25:02 -!- edlothiol has quit [Ping timeout: 250 seconds]
20:31:50 -!- upsy has quit [Quit: Leaving]
20:32:43 <greensnark> Ok, Mac binaries are up on SF
20:32:53 <due> yay
20:32:59 <greensnark> Let me do the CAO honours
20:33:05 <due> plz do
20:36:11 * jpt9 gets out his iBook and upgrades...
20:39:08 <jpt9> Also, thanks for still supporting people like me who have 6-year-old PowerPC machines running OS X 10.4...
20:41:41 <due> jpt9: I think tha'ts cos snarky green is running an old Mac too :)
20:41:49 <jpt9> due: Ah.
20:42:01 <jpt9> due: I don't really use my iBook too often, but DCSS plays quite nicely on it.
20:42:19 <jpt9> due: Well, except for when I screwed up and managed to set the console version as the default shell in Terminal.app.
20:42:27 <jpt9> (Did manage to fix it.)
20:43:53 * jpt9 oohs... and the icon's much higher-res now.
20:46:57 <greensnark> due: I'm not on an old Mac :P
20:47:21 <greensnark> I build with the 10.4 SDK, but I can't actually test it on 10.4 or 10.5
20:47:35 <jpt9> Ah.
20:48:06 <jpt9> Well, I have a 10.4 machine (14" iBook, 1.33GHz G4, 512MB RAM, 10.4.10) if you want me to try stuff.
20:48:16 <jpt9> Also, I can't seem to Cmd-Tab from the tiles version...
20:48:28 <jpt9> (In fullscreen mode, that is.)
20:53:56 <greensnark> No no, I don't need help testing stuff, I just do builds :P
20:54:15 <greensnark> But by all means file bugs you find on Mantis
20:55:10 <jpt9> Ah.  Okay.
20:55:18 <jpt9> (Well, you said you couldn't test on 10.4 or 10.5...)
21:04:17 -!- Cryp71c has joined ##crawl-dev
21:04:38 <Cryp71c> Evening
21:04:47 <jpt9> Evening.
21:04:51 <Cryp71c> !messages
21:04:51 <Henzell> No messages for Cryp71c.
21:05:02 <Cryp71c> Eronarn, see anything in that commit?
21:05:56 <Eronarn> Cryp71c: my first thought upon reading it was something to do with either the change from attack_number to effective_attack_number you made at some point, or some of the break -> return(false) code
21:06:12 <Eronarn> but i couldn't seem to figure out if that would have the effect you described
21:06:38 <Eronarn> i'd have to compile it and muck around with it to get a better idea of what's going wrong
21:06:51 <Cryp71c> Yeah, those were the biggest shifts in congruency. The code is so highly segmented that I couldn't really identify what was causing what, so I revertted the whole thing so I can look at the original code and get a more "whole" picture before I move the code around.
21:07:26 <Cryp71c> Eronarn, a bit more study and understanding will help anyways, perhaps I'll find some routes to simplify the system, I've noticed a few things that certainly seem inefficient.
21:08:50 -!- casmith789 is now known as mikee
21:09:18 <Eronarn> actually, lemme give it a whirl
21:09:22 <Eronarn> don't have anything else to do this second anyways
21:09:27 -!- mikee is now known as casmith789
21:11:03 <Eronarn> Cryp71c: whcih commit should i be compiling?
21:11:20 <Cryp71c> http://gitorious.org/crawl/crawl/commit/ee3bd0d9b0e1f18ccaff23ef32e213535de56bf6
21:12:31 <Eronarn> hmm, it isn't showing up in my git log of the branch
21:12:48 <Eronarn> oh wait i didn't pull
21:13:13 <Cryp71c> its in my branch too
21:21:22 <Eronarn> Cryp71c: i'm on that commit and i don't notice any abnormal behavior
21:21:52 <Eronarn> try re-testing it with a clean build? perhaps you changed something beyond that commit and didn't notice
21:22:09 -!- Jordan7hm has joined ##crawl-dev
21:22:16 <Cryp71c> Eronarn, monsters never miss. Spawn a rat a let it attack you for 1 turn, you'll notice several things
21:22:29 <Cryp71c> 1) it will never miss, only do no damage or do damage.
21:22:50 <Cryp71c> 2) If it misses its first attack, it will attack again, up to a maximum of 4 times per actual attack.
21:22:53 <Cryp71c> Instead of just once.
21:23:26 <Cryp71c> (You'll notice the multiple attacks because the dark grey debug text that displays the rat's to-hit roll will have x2 or x3 to the side of it, indicating multiple attempts at attacks)
21:23:38 <Cryp71c> IIRC
21:27:54 <Cryp71c> k, gotta take a shower.
21:32:12 -!- Cryp71c has quit [Ping timeout: 250 seconds]
21:47:21 -!- Textmode has quit [Ping timeout: 250 seconds]
21:57:27 <Eronarn> !tell Cryp71c I figured out what the problem is, ping me when you're around so I don't have to try to cram it in a PM
21:57:28 <Henzell> Eronarn: OK, I'll let Cryp71c know.
22:26:08 -!- Jordan_7hm has joined ##crawl-dev
22:27:33 -!- Jordan7hm has quit [Read error: Connection reset by peer]
22:28:27 -!- ortoslon has joined ##crawl-dev
22:28:40 -!- Jordan_7hm is now known as Jordan7hm
23:00:13 -!- ortoslon has quit [Read error: Connection reset by peer]
23:05:21 -!- Textmode has joined ##crawl-dev
23:06:23 <CIA-45> 03kilobyte * r84417ff40d60 10/crawl-ref/source/util/gen_ver.pl: Don't show hashes in versions that fit an exact tag.
23:06:34 <CIA-45> 03kilobyte * r199e348f42ec 10/crawl-ref/source/ (dat/database/godspeak.txt main.cc): Let Lugonu explain to Abyssal Knights what they're supposed to do.
23:06:35 <CIA-45> 03kilobyte * r57d622063467 10/crawl-ref/source/dat/descript/branches.txt: Makeshift descriptions for unfinished branches.
23:06:35 <CIA-45> 03kilobyte * re146c5cc0a7b 10/crawl-ref/docs/ (034_changes.txt obsolete/034_changes.txt): Move 0.3.4 changes to obsolete/
23:07:24 -!- ortoslon has joined ##crawl-dev
23:14:50 -!- ortoslon has quit [Read error: Connection reset by peer]
23:27:24 -!- Twinge has quit [Read error: Connection reset by peer]
23:28:25 -!- Twinge has joined ##crawl-dev