00:29:58 <Eksell> Unstable branch on CRAWL.XTAHUA.COM updated to: 0.21-a0-225-g207ae17 (34)
02:48:59 -!- Amnesiac_ is now known as Amnesiac
02:51:06 <alexjurkiewicz> !tell hellmonk very weird issue on cpo -- tiles dont appear until i mouse over them?!
02:51:06 <Sequell> alexjurkiewicz: OK, I'll let hellmonk know.
02:51:09 <alexjurkiewicz> !tell hellmonk Error while drawing cell undefined at 28/-20: TypeError: Cannot read property 'pos' of undefined
02:51:09 <Sequell> alexjurkiewicz: OK, I'll let hellmonk know.
02:51:12 <alexjurkiewicz> render_cell @ cell_renderer.js:429
02:52:39 <Gretell> Monster database of master branch on crawl.develz.org updated to: 0.21-a0-225-g207ae17
02:55:14 <johnstein> maybe I should hold off on rebuilding hellcrawl then?
03:00:57 <alexjurkiewicz> might just be cpo
03:00:59 <alexjurkiewicz> i am too lazy to test webtiles locally
03:11:42 <amalloy> alexjurkiewicz: new realism update. character can't see things until player looks at them
03:11:49 <Rotatell> Unstable branch on crawl.beRotato.org updated to: 0.21-a0-225-g207ae17 (34)
03:11:58 <amalloy> IRL LOS
03:18:07 <alexjurkiewicz> :O
03:59:33 -!- amalloy is now known as amalloy_
09:47:12 <Cheibriados> Maxwells thermic engine wrong description 13https://crawl.develz.org/mantis/view.php?id=11201 by Le_Nerd
10:10:34 <espais> amethyst: are you online?
10:15:58 <espais> !tell amethyst I've been working with Napkin on fixing the captcha issue with Mantis.  My initial thought was to use Google's reCaptcha, but we were curious about your thoughts/plans for the fact that it would be verifying through a third party.
10:15:59 <Sequell> Sorry espais, I don't know who amethyst is.
10:16:11 <espais> !tell |amethyst I've been working with Napkin on fixing the captcha issue with Mantis.  My initial thought was to use Google's reCaptcha, but we were curious about your thoughts/plans for the fact that it would be verifying through a third party.
10:16:11 <Sequell> espais: OK, I'll let |amethyst know.
10:16:31 <Napkin> :)
10:17:00 <espais> !tell |amethyst interestingly mantis itself doesn't support recaptcha in terms of their dev plan, but i have a local instance of it working.  only debate now is if we're OK with google's privacy policy or not
10:17:00 <Sequell> espais: OK, I'll let |amethyst know.
10:17:15 <espais> hey Napkin
10:17:38 <Napkin> hey erik, what's up? :)
10:18:36 <espais> oh you know, trying to be productive at work and instead am looking over privacy policies and php code
10:18:42 <espais> and you?
10:19:41 <gammafunk> |amethyst: ping re above, if you're around
10:19:51 <Napkin> heh, same here - debian admin trying to make things work in centos/rhel... it's messy
10:20:05 <espais> one thing i've been curious about, i noticed you sign your emails in both english and deutsch...intrigues me because i lived in germany for two years :)
10:20:13 <espais> sorry gamma, missed the | when i did the initial tell
10:20:15 <gammafunk> It's easy to forget that you have to pipe to the amethyst command
10:20:33 <gammafunk> no worries, just wanted to make sure that he got an irc highlight if he was in fact around
10:20:36 <espais> oof, debian to rhel, that sounds like fun
10:20:59 <espais> i teach system administration so semester 1 we do cent and semester 2 we move to ubuntu
10:21:08 <espais> just to throw the students for a loop
10:21:33 <Napkin> send me your papers, could use it now ;)
10:21:58 <espais> heh, what's your issue?
10:23:49 <Napkin> oh, just trying to find my way around.. yum install ncurses-base to make TERM=xterm-256color work, yum install vte-profile to not get an error after every enter in shell, managed to write ifcfg-p5p2 and ifcfg-p7p2 to get ifcfg-bond0 running.. now, where do i set the default gateway?
10:24:14 <Napkin> nothing too exciting, just basics - which are just differently handled in debian
10:24:38 <espais> yea, i always feel like the deb CLI is more user-friendly in terms of commands and arguments
10:24:44 <espais> rhel just feels...messy
10:25:05 <espais> though the purists will say it's better that way :)
10:25:36 <espais> Napkin: this may be helpful for your gateway issues?  http://xmodulo.com/how-to-set-a-default-gateway-on-centos.html
10:25:45 <Napkin> can't tell yet... but definitely bash-completion is not very programmed yet - probably need to install...
10:26:27 <espais> did you also pull down the extras?
10:26:41 <espais> yum install bash-completion bash-completion-extras
10:27:13 <Napkin> wait, no gateway no fun ;)
10:27:23 <espais> ooh, yea
10:27:51 <Napkin> read that specifying gateway in /etc/sysconfig/network is deprecated
10:29:32 <espais> which version of cent are you using?
10:30:11 <Napkin> 1611
11:20:30 <|amethyst> !tell espais I don't have a problem with verifying through a third party.  If a user does have problems with that, perhaps we could ask them to email one of the admins
11:20:31 <Sequell> |amethyst: OK, I'll let espais know.
13:17:44 <gammafunk> wow, do two tiles of bushes block los?
13:18:01 <geekosaur> not that I've r seen
13:18:04 <geekosaur> *ever seen
13:18:08 <gammafunk> hrm
13:18:21 <gammafunk> !source los-type.h
13:18:21 <Sequell> https://github.com/crawl/crawl/blob/master/crawl-ref/source/los-type.h
13:19:04 <gammafunk> table implies that for LOS_SOLID_SEE, two tiles of features in that class "c" are opaque
13:19:09 <gammafunk> and "c" includes bushes
13:19:15 -!- amalloy_ is now known as amalloy
13:19:23 <amalloy> they totally do, yes
13:19:27 <gammafunk> deng
13:19:46 <gammafunk> well
13:20:25 <gammafunk> incoming gammafunk vaults feature dynamic, emergent gameplay featuring flavorful LOS manipulation through destruction of various bushes, shrubs, and hedges
13:21:11 <gammafunk> topiary portal, bush fires revealing horrible dragons, the works!
13:21:41 * geekosaur gets that bush maze fairly often and the outer walls don't block LOS when viewed longwise... *confused*
13:22:21 <geekosaur> I mean, not the way trees used to
13:22:35 <gammafunk> hrm, it's not something silly like them having to be contiguous, is it?
13:22:50 <gammafunk> or just adjacent, I guess
13:22:53 <geekosaur> I'm talking about longwise @..PPPPPPP
13:23:05 <amalloy> i just tried @PP and @P.P in wizmode
13:23:10 <amalloy> both block LOS to the other side
13:23:53 <geekosaur> hrm, then maybe we have a display bug because the area on the other side is not darkened
13:24:04 <gammafunk> ah, yeah, that could be the case
13:24:11 <gammafunk> you could try putting an item on the ground
13:24:15 <gammafunk> on the third bush
13:24:25 <espais> BURN EM ALL
13:24:25 <Sequell> espais: You have 1 message. Use !messages to read it.
13:24:27 <gammafunk> but I guess xv will just tell you
13:24:30 <espais> !messages
13:24:30 <Sequell> (1/1) |amethyst said (2h 3m 59s ago): I don't have a problem with verifying through a third party.  If a user does have problems with that, perhaps we could ask them to email one of the admins
13:24:31 <gammafunk> "can't see that place"
13:25:26 <espais> |amethyst: sounds good! i'm planning on pushing the updates in the next couple of days after i get a chance to clean up my comments
13:25:45 <gammafunk> the days of the captcha refresh minigame are coming to and end!
13:26:01 <amalloy> geekosaur: not darkened? it shows up as out-of-los blue for me
13:26:32 <espais> gammafunk: oh you never know.  maybe i'll incorporate a new captcha that requests a blood sample
13:26:38 <gammafunk> :o
13:26:50 <espais> somebody on reddit suggested something with sigmund i think
13:26:51 <amalloy> "fill this vial to prove you are a human"
13:26:54 <gammafunk> sorry but that's not going to make me play a vampire
13:27:02 <espais> ha
13:27:09 <espais> "click all the tiles with sigmund"
13:27:13 <espais> > sigmund is invisible
13:27:30 <gammafunk> even better, little dcss tactical puzzles
13:27:35 <espais> oof
13:27:39 <espais> i'd fail them all for sure then
13:27:57 <gammafunk> "enter the optimal command for the next move"
13:28:11 <amalloy> gammafunk: but on windows ^q closes your browser
13:28:30 <gammafunk> amalloy: you would fail this test since you'd be trying to enter "aa" for Slouch
13:28:36 <amalloy> wow
13:47:36 <espais> hmm
13:47:40 <espais> draconian's don't get hat sprites?
14:09:50 <espais> i kind of want to make a saucy dragon hat sprite now
14:09:58 <johnstein> !bug 11104
14:09:58 <Sequell> https://crawl.develz.org/mantis/view.php?id=11104
14:10:36 <johnstein> based on my research it seems like existing intent of crawlcode is to always block !invis when you have a self-inflicted halo, even when umbraed
14:11:32 <johnstein> for non-self-halo !umbraed you get the "this is useless, continue?" and umbraed you are able to go invis. I think the answer to that bug should be to just improve the messages
14:12:15 <johnstein> the code broke my brain a bit because there's "haloed", "halo_radius", backlit(), and umbraed(), and all the stupid intermediate combos
14:13:06 <johnstein> self-inflicted means TSO and Eos
14:15:40 -!- Fixer_ is now known as Fixer
15:05:10 <shmup> any old devs aware of the genesis of DCSS flavor text and its similarities to the earlier style of mtg flavor text?
15:05:20 <shmup> when it used to often use lines of poems, and things like lines from shakespeare
15:05:44 <shmup> re:
15:05:46 <shmup> https://github.com/crawl/crawl/blob/master/crawl-ref/source/dat/descript/quotes.txt
15:12:31 <amalloy> used to?
15:13:07 <amalloy> i guess it's been a little while since https://github.com/crawl/crawl/commit/16a58bf2d60ef6f9753e4688a368426b7be12a59 but it's not like we've decided to stop doing it
15:13:36 <amalloy> just nobody has been writing much i guess. i'm sure we would be happy to accept more
15:20:04 <shmup> used to regarding mtg, not dcss
15:20:18 <shmup> mtg has changed its direction for flavor text a few times
15:20:39 <shmup> my fault on wording :)
16:58:44 <|amethyst> shmup:  sometimes we quote Magic cards!
16:59:54 <|amethyst> (we should improve our citation format for those, though; at least list the card and edition!)
17:07:51 <|amethyst> shmup:  it was a Jyhad (later V:tES) card that introduced me to the poem "George Crabbe" by E.A. Robinson
17:08:21 <|amethyst> shmup:  (it was, however, Iron Maiden that introduced me to "Rime of the Ancient Mariner")
17:11:21 <|amethyst> shmup:  Back to your question, I'm not sure our use of quotations was influenced by M:TG, so much as Nethack
17:11:42 <|amethyst> shmup:  though unlike Nethack we're perfectly happy to quote contemporary popular music
17:12:12 <|amethyst> shmup:  I once made a list of all the post-1950s popular music quoted in Crawl, ordered by year.... but I seem to have lost it
17:14:48 <|amethyst> shmup:  we also quote webcomics reasonably often
17:21:32 <|amethyst> or maybe I went back to 1945, since we do quote at least two songs from the 40s
19:29:42 <johnstein> !quote should return a rando
19:29:43 <johnstein> random quote
19:34:01 <alexjurkiewicz> please no more 90s irc tech
19:36:03 <gammafunk> Yeah, implement !quote in Discord instead, so that it can be 2010s tech
19:41:11 <amalloy> yes, 90 seconds is too long to wait for a quote
19:48:15 <gammafunk> It took me 2010 seconds to get that joke
19:48:41 <gammafunk> !calc 2010.0 / 60
19:48:41 <Sequell> 33.5
19:48:49 <gammafunk> which means I still haven't gotten it
19:58:38 <gammafunk> hmmmm
19:59:12 <gammafunk> !source _is_near_stairs
19:59:12 <Sequell> 1/1. https://github.com/crawl/crawl/blob/master/crawl-ref/source/mon-place.cc#L736
20:00:29 <gammafunk> I believe this loop is not well designed....
20:02:06 <gammafunk> comments refer to a function called near_stairs(), which doesn't exist as such
20:04:27 <gammafunk> time to do some commit diving to see just how this got so messed up
20:05:14 <amalloy> %git 033b2f5904e3a87e5ce2605b2779be860bf46682
20:05:14 <Cheibriados> 07jpeg02 * 0.4-a0-2950-g033b2f5: Fix 1937489: Remove map markers overwritten by lakes or rivers, as was              already done for vaults. 10(9 years ago, 11 files, 238+ 129-) 13https://github.com/crawl/crawl/commit/033b2f5904e3
20:05:37 <amalloy> gammafunk: ^ is when that function and comment were added
20:06:28 <gammafunk> gah
20:06:39 <gammafunk> yeah I was about to git log -G
20:06:48 <amalloy> i usually prefer -S to -G
20:06:55 <gammafunk> but...this basically stopped working at that time?
20:07:35 <gammafunk> looks like jpeg just tried to make some kind of adjacency iterator
20:07:43 <gammafunk> but didn't implement it right
20:08:03 <gammafunk> so it only checks the location p 9 times
20:08:21 <amalloy> agreed
20:08:53 <gammafunk> and I'm not sure why this was changed in this commit, I guess it was an attempt at refactoring?
20:08:56 <gammafunk> I need to read the whole diff
20:10:46 <gammafunk> I see
20:10:53 <gammafunk> near_stair() looks like it had a radius argument
20:11:00 <gammafunk> which was 1 in the old calls to that
20:11:15 <gammafunk> so that's what it was going for, she just implemented the iteration incorrectly
20:15:09 <amalloy> so what is mg.proximity?
20:16:18 <gammafunk> it's the spawn location determinator
20:16:37 <gammafunk> as in should the monster be generated near the player, near a stair, away from the player
20:16:46 <gammafunk> I'll be adding a new on for "away from stairs"
20:16:50 <gammafunk> s/on/one/
20:17:10 <gammafunk> it also is used in a kind of secondary way for choosing a random monster type
20:17:32 <gammafunk> namely if the monster has proximity of near stairs and the stairs chosen are a branch
20:17:46 <gammafunk> the monster type is selected from the population of the destination branch
20:18:02 <gammafunk> which is how those spawns that come from stairs work
20:18:31 <gammafunk> but other than that it's for changing behaviour of how the monster's random creation location is chosen
20:21:41 <gammafunk> gah
20:21:53 <gammafunk> the code before this commit looked messed up as well
20:22:46 <gammafunk> oh, wait
20:22:52 <gammafunk> maybe everything was in one big, happy function
20:23:53 <gammafunk> hrm, nope
20:24:10 <amalloy> what was there before is hard to understand
20:24:39 <gammafunk> yes, but it looks like pval was never properly set in place_monster()
20:24:47 <gammafunk> from what I can tell it was simply always 0 before jpeg's commit
20:25:31 <gammafunk> so her commit did make that pval variable relevant, at least
20:25:39 <johnstein> is there an easy way to see how much crawl code is "original"? (henzell code)
20:26:33 <amalloy> johnstein: you'd probably want to start with https://github.com/crawl/crawl-ancient
20:26:49 <amalloy> the git history for crawl/crawl doesn't have enough information to answer the question
20:26:55 <gammafunk> alright, so what I should probably do is
20:27:04 <gammafunk> look at how the random spot near stairs is actually chosen
20:27:28 <gammafunk> and see if that even coincides with this "adjacent = near" idea that the function was trying to go for
20:28:49 <amalloy> and i guess you'd also want some idea of which versions of crawl-ancient correspond to linley. even if you knew that, i'm not sure how you could correlate crawl-ancient code with crawl code
20:29:59 <amalloy> good quickstart guide in the readme. "[...] Quitting ('Q') is a way of committing suicide if you can't even be bothered to help your character escape from the Dungeon alive. Well, that's it for the quick-start guide."
20:32:34 <|amethyst> that should be fixed, since Q is no longer canonically suicide
20:32:45 <amalloy> |amethyst: we're talking about crawl-ancient
20:33:07 <|amethyst> ah, good
20:33:57 <|amethyst> actually, nearly the same sentence is still in the current quickstart guide
20:34:01 <|amethyst> Quitting ('Ctrl-Q') lets
20:34:01 <|amethyst> you commit suicide if you can't even be bothered to help your character escape
20:34:04 <|amethyst> alive.
20:34:25 <amalloy> what is the current lore for ^q?
20:34:27 <|amethyst> Followed by a paragraph break and "Well, that's it for the quick-start guide."
20:34:43 <|amethyst> amalloy:  You "abandon" your character
20:34:53 <|amethyst> I guess they go native and probably turn into a shopkeeper or something
20:35:03 <|amethyst> %git :/moral issue
20:35:04 <Cheibriados> 07|amethyst02 * 0.11-a0-2832-g354d479: Make quitting less of a moral issue. 10(5 years ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/354d47989254
20:38:17 <amalloy> i'm sorta amazed that large chunks of the readme are old enough to vote
20:43:07 <johnstein> FR: when a character dies with randarts, create .loot files, similar to .bones, that can be used to populate randarts in shops (name and attributes)
20:43:21 <johnstein> not sure if that's really a good FR or not.
21:14:09 <Yermak> Distortion shouldn't exist.
21:15:41 <Yermak> Or at least the use of distortion should be always deliberate player's choice.
21:16:49 <Yermak> Treating each glowing weapon including god gift as possibly having distortion brand is just unplayable.
21:22:22 <gammafunk> https://github.com/crawl/crawl/blob/master/crawl-ref/source/mon-place.cc#L555
21:22:49 <gammafunk> mon-place.cc is turning into an exciting new journey into crawlcode for me
21:29:53 <hellmonk> agree w/ yermak btw
21:29:53 <Sequell> hellmonk: You have 3 messages. Use !messages to read them.
21:32:51 <hellmonk> hmm, time to figure out what got jacked up in webtiles
21:34:00 <hellmonk> probably the js change I made to piety display
21:35:56 <hellmonk> also I think I need to fix savecompat
21:40:37 <alexjurkiewicz> i was looking into merging latest vanilla crawl into hellcrawl btw. It is an exceptionally large amount of work
21:40:37 <Sequell> alexjurkiewicz: You have 1 message. Use !messages to read it.
21:40:48 <alexjurkiewicz> sadly
21:41:50 <hellmonk> gonna try to fix this display stuff rn
21:44:35 <amalloy> alexjurkiewicz: yeah, you kinda have to keep forks in sync with regular merges. if you let them drift the effort gets too large
21:46:01 <hellmonk> think I might have found it
21:49:37 <hellmonk> alexjurkiewicz: see if this fixes it
21:53:09 <amalloy> lol. good commit hellmonk
21:53:20 <hellmonk> yea ikr
21:53:44 <hellmonk> the disadvantage of doing a huge batch of commits and then forgetting to test the webtiles version
21:54:16 <hellmonk> its ok tho, I got the meme juice
21:55:58 <amalloy> hellmonk: imo rename C O M M A L E S S B O Y S in honor of your recent corpseless commit
21:56:21 <hellmonk> but I added a comma so it would have to be C O M M A B O Y S
21:56:34 <hellmonk> or P U N C T U A T I O N B O Y S
21:57:51 <advil> heh that's a good looking bug
21:58:31 <advil> wonder if there would be some way to add some kind of automated testing for webtiles, not really sure what I have in mind
21:58:50 <advil> but that seems like the sort of thing you'd hope travis would catch somehow
22:01:45 <alexjurkiewicz> hellmonk: testing
22:04:41 <|amethyst> advil:  would also nice to have unit tests that 'make test' runs
22:04:52 <geekosaur> automated testing is potentially possible, but you get to figure out how to drive either Xvfb or a remote control harness (e.g. vnc)
22:05:24 <|amethyst> geekosaur:  or something like Headless Chromium
22:06:01 <gammafunk> I have to wonder what class of webtiles errors it might find
22:06:27 <|amethyst> yeah, the difficult part there is to figure out what to test
22:09:48 <advil> some kind of js validator might be a good start
22:10:25 <|amethyst> you could do that without a browser even
22:10:30 <|amethyst> or half of that
22:10:52 <|amethyst> but how do you exercise all the messages crawl might send to the client?
22:11:43 <hellmonk> hmm, if that update went through it didn't fix it
22:12:12 <|amethyst> btw, interesting thing I've noticed watching GDC's classic game postmortems
22:12:17 <gammafunk> well, there are only so many kinds of json messages
22:12:48 <gammafunk> I wonder if just doing some form of the current make test and looking for javascript errors as the app runs would do it
22:13:01 <|amethyst> the console game designers all talk about implementation stuff, since implementing anything on an 8-bit console is interesting
22:13:12 <|amethyst> but the PC game designers stick to talking about design
22:13:51 <gammafunk> yeah, I could see that, since the audience of developers would mostly be unfamiliar with the rigors of working with 8-bit assembly and all that low-level stuff, I guess
22:14:54 <|amethyst> the interesting part IMO is mostly the graphics hardware
22:14:56 <|amethyst> or in the case of the Atari 2600, the lack thereof
22:15:22 <gammafunk> This is what we had to work with!
22:15:49 <alexjurkiewicz> hellmonk: it did fix it
22:15:50 <|amethyst> and now our graphics hardware is a transputer more powerful than your CPU
22:15:57 <hellmonk> ok, cool
22:16:41 <hellmonk> still seems broken while I'm testing it though
22:16:55 <alexjurkiewicz> might be a caching thing
22:17:01 <hellmonk> ah, maybe so
22:17:42 <alexjurkiewicz> I'm not sure why but tile images sometimes don't get noticed as changed by the httpd setup
22:17:51 <alexjurkiewicz> usually it fixed itself after a day or so
22:18:02 <alexjurkiewicz> I emptied my cache and tested successfully anyway
22:19:21 <|amethyst> there is an option you can set in webtiles to disable caching
22:20:07 <hellmonk> seems like I borked old savefiles though
22:20:07 <|amethyst> by default we cache aggressively, since in a typical dgl setup each trunk version gets its own path
22:24:23 <hellmonk> hmm, wonder what I did to jack up saves
22:24:29 <hellmonk> something with an enum probably
22:26:14 <hellmonk> I guess I can just purge old saves
22:26:24 <hellmonk> savecompat is for losers
22:35:45 <hellmonk> !tell johnstein also I screwed up savecompat again, so please nuke old saves when you update
22:35:45 <Sequell> hellmonk: OK, I'll let johnstein know.
22:35:57 <hellmonk> alexjurkiewicz: pls deleteroni old saves when you have the chance
22:36:56 <johnstein> hellmonk: I haven't run the latest version yet. does that have any impact on save compat?
22:36:57 <Sequell> johnstein: You have 1 message. Use !messages to read it.
22:37:17 <hellmonk> yeah, I think I screwed up an enum or something
22:37:19 <johnstein> ie is it worth me holding off for you to do save compat first? or has that boat already sailed and I should just nuke then rebuild?
22:37:27 <hellmonk> boat has sailed imo
22:37:38 <hellmonk> I don't want to dig through 3 weeks of commits looking for it
22:37:47 <johnstein> ok. later tonight I will nuke and rebuild fresh
22:37:52 <hellmonk> thanks