00:00:36 Unstable branch on CRAWL.XTAHUA.COM updated to: 0.22-a0-479-gc6e99f3 (34) 00:58:22 Unstable branch on crawl.beRotato.org updated to: 0.21-a0-643-g786e97d (34) 01:29:12 johnstein: you're probably aware of this, but it looks like /static/style.css got reverted to a previous version 01:34:26 I don't even know what that file is. did that reversion just happen after 786e97d installed? 01:34:29 %git e03441e 01:34:29 07gammafunk02 * 0.20-a0-1132-ge03441e: Two overflow transporter vaults 10(1 year, 1 month ago, 1 file, 95+ 0-) 13https://github.com/crawl/crawl/commit/e03441e5eb43 01:40:33 %git 0.20-a0 01:40:33 07Brannock02 * 0.20-a0: Finalize 0.19 changelog 10(1 year, 8 months ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/8ddec9435a29 01:40:36 could be that, or could be 786e97d: they both have the same file 01:40:51 is this a file I need to manaully install? 01:41:08 I'm not at all familiar with the static files for the webserver 01:42:10 you shouldn't need to manually install it; it was installed and working just fine before 01:42:37 yea, it's another fallout of me trying to do the version transfer stuff... :/ 01:42:39 ugh 01:43:05 my guess is that the update Rotatell just announced overwrote it 01:43:29 yea sounds like it. I modified the install script to stop setting the date of old versions to NOW 01:43:36 yeah, no worries, take your time fixing it; just letting you know 01:43:52 since that screwed things up before (I actually missed that this time too, but I think for sure I got it now) 01:44:02 but I didn't do any modification of the static stuff 01:46:54 %git 01:46:54 07gammafunk02 {GitHub} * 0.22-a0-479-gc6e99f3: Merge pull request #788 from NormalPerson7/fix-temple-vault 10(8 hours ago, 0 files, 0+ 0-) 13https://github.com/crawl/crawl/commit/c6e99f3da94b 01:46:58 &versions 01:47:20 CAO: 0.22-a0-474-g19b682e, CBRO: 0.22-a0-479-gc6e99f3, CDO: 0.22-a0-469-gf26638a, CJR: 0.22-a0-479-gc6e99f3da9, CPO: 0.22-a0-479-gc6e99f3, CUE: 0.22-a0-475-g101988f953, CWZ: 0.22-a0-422-g9628fc8, CXC: 0.22-a0-479-gc6e99f3, LLD: 0.22-a0-443-g17c3723931 01:50:07 should get fixed next time trunk is built 01:50:17 I can't rebuild it now since it's already up to date 01:50:32 it's not just that file, basically all the files in there.... 01:51:02 this is really annoying trying to install these old versions and not screwing something else up... 01:51:31 thanks 01:52:06 np. your stuff has been easy :) 01:52:06 I wouldn't worry about the other files much; everything in static/ hasn't been modified in a few years, with the exception of chat.js courtesy of advil 01:52:29 great :) glad to hear 01:52:44 oh. did I just blow away the mute fix? 01:54:24 yeah it got blasted back to the past as well 01:55:14 bleh 01:55:35 I hate to ask for this, but could you or another dev just push some new commit. maybe fix a typo or something 01:55:43 then I can rebuild and get everything back 01:55:55 but since client.js got blasted at the same time, I believe it will just work 01:55:59 and tomorrow I can try again to do my fancy updates after thinking about things more 01:56:11 ^status 01:56:11 21 Crawlers. CBRO disk usage=99% (135GB) | RAM usage=28% (4GB)| uptime/CPU= 01:56:10 up 158 days, 5:32, 5 users, load average: 0.91, 0.83, 0.98 (4 Cores) http://status.berotato.org 01:56:14 bah 01:56:22 yeah, sure, I noticed a FOUC with the ui-overhaul stuff that I can fix and push now 01:56:33 gotta be a master commit though 01:59:48 03Aidan Holm02 07* 0.22-a0-480-g991dd4b: Fix webtiles lobby flash-of-unstyled-content 10(78 seconds ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/991dd4b2b45e 02:03:00 johnstein: there ya go 02:06:38 thanks! 02:06:54 my dumb workaround 'install-oldtrunks' script really doesn't work well 02:07:17 in addition to it installing old crap in the webserver, it also seems to screw up the submodules 02:07:31 SDL2 in this case. had to go in and do some manual fixin' 02:09:15 shouldn't need to do anything with sdl2 for a webtiles build (you'd think) 02:09:23 !tell gammafunk I'm trying to install the right 0.20-a0 version for my transfers. I found this one: %git 0.20-b1~1 -> 0.20-a0-1132-ge03441e but when I try to compile, it gets to macro.cc and I ^status 02:09:24 johnstein: OK, I'll let gammafunk know. 02:09:26 oops 02:09:29 ^status 02:09:30 17 Crawlers. CBRO disk usage=97% (135GB) | RAM usage=12% (4GB)| uptime/CPU= 02:09:29 up 158 days, 5:45, 4 users, load average: 1.56, 1.37, 1.22 (4 Cores) http://status.berotato.org 02:09:54 Unstable branch on crawl.jorgrun.rocks updated to: 0.22-a0-480-g991dd4b2b4 (34) 02:10:29 !tell gammafunk it gets to macro.cc and I get a compile error that I forgot to copy. trying to figure out if A) that's the right version I should use for 0.20-a0, and B) how the heck to even get it to compile. 02:10:29 johnstein: OK, I'll let gammafunk know. 02:18:30 Unstable branch on crawl.beRotato.org updated to: 0.22-a0-480-g991dd4b (34) 02:20:28 wrt 'install-oldtrunks', probably just building the crawl binary is enough 02:21:06 ... unless it dies when it can't find maps etc I guess 02:36:44 it needs to be a legit install so players can choose to not transfer to newer versions 02:36:44 but this makes me wonder now 02:37:15 if I'm playing an old trunk version from like 0.18 or something, if there's a single webserver installed, shouldn't this issue happen a lot? 02:37:38 i.e. old trunk versions being out of sync with the latest trunk version's webserver installs? 02:37:54 as usual, I don't have a full understanding of what's really going on 02:39:02 ok new plan 02:39:08 there are two scripts: update-trunk and update-stable; my guess is that the latter does not clobber the stuff in static/ 02:41:14 update-stable only runs for experimentals and the stable releases. we run update-trunk every night, so I think the files in webserver/static get updated every single night (potentially) 02:42:02 and aren't you doing an automatic forced update? in this case at least you wouldn't need to ask players whether to update 02:42:27 also, my new plan is to not install the latest a0 for each version, I'm just going to point them all to MY latest version. so for 0.20, instead of 0.20-a0-1132-ge03441e, which I don't have and I can't even install (because it won't compile and because it breaks things), I can just use the existing: 0.20-a0-1126-g0f15420 which is close enough 02:42:52 heh. I'm confusing things by simultaneously discussing two things. 02:43:15 right, so as long as you only install newer revisions with update-trunk, and your trunk install is always the newest, you shouldn't have problems 02:43:27 general comments/questions on how we currently handle the normal updating of trunk every night and how we don't force old trunk games to update 02:44:21 since we don't force old games to update, but we only have a single static folder for the webserver (which as far as I can see should/could affect both trunk and stable), how do old trunk games that haven't been transferred deal with being 'out of sync' with the static files? 02:44:40 or is our saving grace that we usually don't change those files much, so it's not usually a big deal in practice? 02:44:50 oh, that's simple; we never break compatibility in that directory 02:45:17 oh ok. so it's a conscious decision to ensure the current static files play nice with old versions 02:45:31 that sounds fun when you are trying to add cool new stuff 02:45:52 yeah, 'fun' is one word for it... 02:45:54 heh 02:46:33 I'm using the Dwarf Fortress definition of FUN 02:47:14 pretty much 02:47:53 you also don't have access to the game version/hash from JS in that folder, so you can't do the same sort of compatibility stuff you can do in C 05:35:39 -!- Euph0ria is now known as TabMasher 05:38:06 -!- TabMasher is now known as Euph0ria 05:38:20 -!- Euph0ria is now known as EuphOria 09:59:30 johnstein: hm, do you mean to say that there is *just one* copy of static for every version? 09:59:50 I had sort of naively guessed something like one per each 10:00:18 though I guess since the lobby is part of static, it needs to be outside versioning to some degree 10:01:56 there's one copy of game_data/ per version, which I contains a static/, and then theres a top level static/ for the lobby 10:03:45 which one is it using when it is in a game? 10:04:21 i.e. does it actually use the per version static files? 10:09:20 games use game_data for most stuff, but the top level static/ provides an api for showing/hiding dialogs, used by inventory etc 10:30:04 oh, I misunderstood 10:30:16 webserver/static/ is specifically what I was trying to ask about 10:31:12 so that's what I was wondering if there's just a single copy of 10:34:25 and it seems the answer is yes 10:37:07 I wonder if it might be worth splitting that into a separate repo 10:39:05 or at least removing the code sharing between lobby and game 10:39:47 the added code duplication may or may not be better than the perpetual backcompat requirement 10:41:53 I made a commit in webtiles-changes 10:41:56 let me find it 10:43:08 %gti 912e21c75b080 10:43:13 %git 912e21c75b080 10:43:13 07gammafunk02 * 0.16-a0-3794-g912e21c: Add Makefile targets to build only the WebTiles client files 10(3 years, 5 months ago, 1 file, 18+ 8-) 13https://github.com/crawl/crawl/commit/912e21c75b08 10:43:24 splits building the client into its own make target 10:43:30 is there 10:47:50 is there an autumn tile for mons plant? 10:49:30 don't think so 10:55:12 advil: yea there’s only one webserver/static folder. not sure if any of the stuff in there ever gets copied to any of the version folders 10:55:34 also gammafunk you might be able to ignore my !tell from last night 10:55:57 I think I’m going to give up on installing old versions that match up to the latest ao 10:56:23 it’s too fraught with unintended perils 10:57:00 so I’m going to just update all games to whatever the highest version I actually have. which isn’t perfect but good enough. esp since so many of these games are quite likely abandoned anyway 11:02:47 johnstein: gotcha, I hadn't really looked into it further, but that seems pretty reasonable 11:06:59 webtiles chat FR: add a little green dot to the favicon when there are unread messages in a background webtiles tab 11:11:28 webtiles chat FR: allow the chat window to pop out to a resizable window 11:12:51 oh this was a sort of fun thing I added to webtiles changes, it's probably a bad idea, but it was on cpo for a while 11:14:09 %git 2df5926d53 11:14:09 07gammafunk02 {edlothiol} * 0.15-a0-2344-g2df5926: Player title colors and descriptions for WebTiles lobby and chat 10(3 years, 8 months ago, 10 files, 302+ 9-) 13https://github.com/crawl/crawl/commit/2df5926d53a2 11:15:27 def get_nerdtype(self, username): 11:17:46 that's very nifty; does it work for non-react webtiles? 11:18:21 no, since the trunk webtiles doesn't have that USR2 signal handler support 11:18:40 of course course it wouldn't be hard to merge; but I think you'd also want to think a bit harder about the general approach 11:18:56 one problem is that it's all done through sequell nicks that anyone can update 11:19:21 and I guess the other is that it's not really controlling for someone registering an account on a server where the relevant user hasn't already 11:19:29 and this could be seen as encouraging that 11:23:00 maybe if the query could be done through the script instead of using a nick it could be more secure, and for devs we could indicate which servers the account is on 11:23:16 there's a little txt file where we save the dev player accounts 11:28:20 it's the sort of thing I'd be tempted to punt on until SSO is finally implemented 11:38:28 -!- amalloy is now known as amalloy_ 11:44:56 yeah 11:48:43 -!- amalloy_ is now known as amalloy 12:03:42 gammafunk: https://pastebin.com/8LGddKDQ could I get your feedback on this vault? 12:04:45 -!- amalloy is now known as amalloy_ 12:25:53 aidanh, you told me to report bugs with the ui overhaul server? 12:26:29 if you set a skill to stop leveling at a certain point, and you reach that point with no other skills active, you normally get a prompt telling you to pick a skill to level. 12:27:19 when this happened to me (training axes), the menu to select a skill did not pop up. i hard-froze for a few seconds, then the game crashed. Error message, "Unfortunately, your game crashed" (1,25) into (80,24) 12:27:55 this was on webtiles, bero where you were watching me earlier. im going on lunch now but will return in a lil bit. 12:28:09 DarkwingDuck: thanks for letting me know 12:28:32 I've reproduced the other issues you mentioned, with the exception of being unable to scroll on menus 12:31:18 DarkwingDuck: and if I don't respond to any later messages, it's likely because I've gone to bed, but I will eventually see them 12:59:07 oh. to clarify - i can scroll on menus. pgup and pgdown work just fine. the menu issue was that the -more- at the bottom was still present even though there was no more content to scroll thru. 13:05:09 Unstable branch on crawl.akrasiac.org updated to: 0.22-a0-480-g991dd4b (34) 13:05:44 gammafunk: one more https://pastebin.com/hbeEquSb 13:15:09 -!- yesno_ is now known as yesno 14:23:01 ebering: thanks, will take a look and get back to you 14:53:43 ebering: btw if you want to go ahead and put those in a PR, you can, and I can review from there 14:54:03 if you feel they're too alpha for that though, I can just give you some feedback when I get to that 14:55:40 is the "Setting Up dgamelaunch and WebTiles" wiki page considered the most up-to-date reference? 14:56:21 for setting up a public facing server 14:56:33 heh, it is, yes, but you may run into issues not covered there 14:57:38 yeah i figured its not perfect but i just wanted to make sure there wasn't some better document 14:58:00 thanks 15:45:48 -!- amalloy_ is now known as amalloy 15:51:01 i wonder if darkwingduck's menu crash was http://crawl.berotato.org/crawl/morgue/DarkwingDuck/crash-DarkwingDuck-20180608-015817.txt - that's the only crashlog i see from that player, but it doesn't look related to UI code 15:56:09 amalloy: can you verify that the crash is the right file? 15:56:09 !crash DarkwingDuck 15:56:09 1. DarkwingDuck, XL9 DENe, T:12424 (milestone): http://crawl.berotato.org/crawl/morgue/DarkwingDuck/crash-DarkwingDuck-20180608-015817.txt 15:56:13 !lm DarkwingDuck crash 15:56:14 1. [2018-06-08 01:58:17] DarkwingDuck the Scholar of Death (L9 DENe of Kikubaaqudgha) Crash caused by signal #13: Broken pipe (D:7) 15:56:19 !lm DarkwingDuck crash x=file 15:56:20 1. [2018-06-08 01:58:17] [file=cbro/crawl/meta/git/milestones] DarkwingDuck the Scholar of Death (L9 DENe of Kikubaaqudgha) Crash caused by signal #13: Broken pipe (D:7) 15:56:24 so that's trunk 15:56:27 not the UI branch 15:56:34 you can probably do 15:56:45 !kw experimental 15:56:45 Keyword: experimental => explbr!=||file~~(smithgod|jump|combo_god) 15:56:58 !crash DarkwingDuck experimental 15:56:59 No milestones for DarkwingDuck (experimental crash). 15:57:09 actually I think that experimental simply hasn't been added to sequell 15:57:11 which we could do 15:57:19 assuming it has a milestone file set up etc 15:57:20 right, i wasn't sure whether something different needs to be done to see crashes in an explbr 15:58:00 it needs an entry in sequell's config I think 15:58:12 johnstein: hrm, are you aware that http://crawl.berotato.org/crawl/dev/ is public? 15:58:15 I'm not sure what that is 15:59:40 looks like it might be webtiles-changes 15:59:51 maybe thats dbro 16:10:31 that is dbro 16:11:04 but I see a few things I want to modify 16:12:22 I haven’t created the PR for sequell to add the stuff for ui-overhaul 16:12:34 I’ll do that maybe today 16:12:39 johnstein: I can add it to sequell, but I don't see the dir in meta 16:12:42 on cbro 16:12:51 I haven’t made the links either 16:13:12 gotcha, well you can make the full PR if you'd like, or just make the links work and let me know, either way 16:13:17 I can merge PR or make entries 16:16:04 ok. I’ll let you know. thanks 16:52:03 gammafunk: I'll just put them in a pr and force-push I guess but these feel pretty alpha, I still don't have a good sense of valut smithing 16:52:38 yeah, that's fine, and no worries, it just takes a bit of practice and learning all the syntax 16:52:51 I can review from the PR a bit more easilly 16:54:44 Does there strictly need to be elaborate loot distribution code or can vaults simply rely on *|$ to place items accordingly? 16:54:51 Say, if a vault only appears in Depths 16:55:34 muffindrake: certainly if it's only one branch 16:55:38 gammafunk: ok, pr sentoff 16:55:49 and it's not too bad if it only appears later on and has just one placement logic 16:59:11 New branch created: pull/789 (26 commits) 13https://github.com/crawl/crawl/pull/789 16:59:12 03advil02 07https://github.com/crawl/crawl/pull/789 * 0.22-a0-369-g864ee37: Allow ghosts built from bones files to be placed in vaults 10(6 months ago, 7 files, 119+ 5-) 13https://github.com/crawl/crawl/commit/864ee3737e19 16:59:12 03advil02 07https://github.com/crawl/crawl/pull/789 * 0.22-a0-370-gc5ebc23: Add some ghost mini-vaults that will place in Vaults 10(6 months ago, 1 file, 36+ 1-) 13https://github.com/crawl/crawl/commit/c5ebc234aa5c 16:59:12 03advil02 07https://github.com/crawl/crawl/pull/789 * 0.22-a0-371-g9494f11: Add a description line for mirrored player ghosts 10(6 months ago, 5 files, 9+ 3-) 13https://github.com/crawl/crawl/commit/9494f11c4b11 16:59:12 03advil02 07https://github.com/crawl/crawl/pull/789 * 0.22-a0-372-ge1cfe05: Add a permanent store for some fallback ghosts 10(6 months ago, 3 files, 197+ 99-) 13https://github.com/crawl/crawl/commit/e1cfe058fe32 16:59:12 03advil02 07https://github.com/crawl/crawl/pull/789 * 0.22-a0-373-gf8544e0: Don't use permastore when rewriting on level veto 10(6 months ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/f8544e074de5 16:59:12 03advil02 07https://github.com/crawl/crawl/pull/789 * 0.22-a0-374-g2cd10e7: Tweak ghost vault generation rates 10(6 months ago, 1 file, 6+ 6-) 13https://github.com/crawl/crawl/commit/2cd10e7afc61 16:59:12 03advil02 07https://github.com/crawl/crawl/pull/789 * 0.22-a0-375-gff08da7: Don't spawn randomly-placed ghosts on level generation 10(6 months ago, 1 file, 0+ 2-) 13https://github.com/crawl/crawl/commit/ff08da74b210 16:59:12 03advil02 07https://github.com/crawl/crawl/pull/789 * 0.22-a0-376-g958d761: Let ghosts use stairs 10(6 months ago, 1 file, 0+ 1-) 13https://github.com/crawl/crawl/commit/958d7619febf 16:59:12 03advil02 07https://github.com/crawl/crawl/pull/789 * 0.22-a0-377-g3395f5e: Use CHANCE rather than WEIGHT for non-Vaults ghost vaults 10(6 months ago, 1 file, 4+ 4-) 13https://github.com/crawl/crawl/commit/3395f5e703be 16:59:12 03advil02 07https://github.com/crawl/crawl/pull/789 * 0.22-a0-378-g30acc4a: Various vault adjustments based on feedback 10(4 months ago, 1 file, 35+ 26-) 13https://github.com/crawl/crawl/commit/30acc4a4f3b3 16:59:12 ... and 16 more commits 16:59:18 I have three vaults done, one of which only appears in D10- and Depths, the other two depths only 16:59:32 oh no what have i done 16:59:33 two small ones and one large 16:59:48 ebering: pls 16:59:59 it's ok, I think it's because he based it on a branch 17:00:04 so we got some spam, but nbd 17:00:09 saved by the god, the ellipsis 17:00:53 I think chei doesn't understand prs that target branches? 17:01:48 well, it just shows the commits relative to master 17:01:59 so when the pr is created, you'll see that 17:02:11 but all subsequent commits I think will just show up by themselves 17:02:14 Shouldn't you just PR to the fork instead, if you're building on top of that ghost vault pr? 17:02:56 what does the fork mean in this context? 17:02:58 Oh, wait 17:03:09 I assumed the changes were in a separate repo, not merely a branch 17:03:12 My mistake. 17:03:16 his fork branch should be relative to crawl/spooky_vaults, yeah 17:03:42 it is. chei just interprets everything relative to master I guess 17:04:10 the spam will only be for the initial PR at least 17:04:30 subsequent commits will just show up by themselves 17:07:22 Are the only people who can push commits to the ghost vault PR those that have write access to the main repo? 17:14:06 yes 17:14:16 if you make a PR, you'll make a new one 17:14:26 Aww 17:14:44 and presumably there isn't any way to give write access on a branch basis either? 17:15:01 we generally don't do that, no 17:15:03 not sure why it would really be needed in this case 17:15:27 a little chei spam is fine, you could however wait until branch gets merged 17:15:36 then submit your PR relative to trunk 17:15:38 It would somewhat reduce the amount of stray PRs, but fine 17:15:51 yeah, stray prs aren't a concern 18:01:33 ebering: Your first vault isn't calling ghost_vault_setup 18:20:47 What would be a near approximation of a bar counter? 18:20:49 Unstable branch on underhound.eu updated to: 0.22-a0-480-g991dd4b2b4 (34) 18:21:03 As in a tavern 18:21:23 as a tile in the game 18:35:51 muffindrake: oops thanks 18:55:16 sadly there's no very good bar counter 18:55:26 a bush is the closest thing that sort behaves like a bar 18:56:04 Oh, good idea 19:07:13 where is the source for Cheibriados? i was interested in seeing how it produces PR notifications 19:08:14 http://s-z.org/neil/git/cheibriados.git 19:16:43 yes, apparently each new branch created (including PR) gets commits announced based on master. it doesn't know there's another branch that already had many of these commits announced 19:17:39 my @revs = split /\n/, `git rev-list $head $exclude_old $exclude_master`; ## where $exclude_old is empty for new PRs, and $exclude_master is "^master" 20:16:19 -!- amalloy is now known as amalloy_ 20:39:42 -!- amalloy_ is now known as amalloy 20:50:47 -!- eb__ is now known as eb 23:59:35 Unstable branch on CRAWL.XTAHUA.COM updated to: 0.22-a0-480-g991dd4b (34)