01:38:09 <Gretell> Unstable branch on crawl.develz.org updated to: 0.25-a0-942-g87a53962fc (34)
02:10:40 <Gretell> Windows builds of master branch on crawl.develz.org updated to: 0.25-a0-942-g87a53962fc
03:30:53 <Kelbell> Fork (bcrawl) on crawl.kelbi.org updated to: 0.23-a0-3054-g1a545eca1e
03:32:25 <Kelbell> Fork (bcadrencrawl) on crawl.kelbi.org updated to: 0.22.1-2675-gec35108dac
05:04:31 <aidanh> gammafunk: iirc thats a function in tags.cc that ideally should be static, but isn't because that would prevent it being accessed from tests
05:05:57 <aidanh> currently I just copy pasted the declarations into the test file, instead of moving them into tags.h
05:06:02 <aidanh> hacky, I know
05:14:39 <gammafunk> aidanh: well, that's sort of why I asked; it does seem a bit weird to have the unit tests influence the organisation of crawl functions, but it seems like better maintenance to make them non-static
05:16:27 <gammafunk> er
05:16:44 <gammafunk> I got to stop saying "make them non-static", since they are non-static already!
05:17:40 <gammafunk> hrm, with your current approach we still have to have a second declaration in tags.cc itself in order to avoid compiler warnings like the one I pasted
06:39:11 <aidanh> well it's tricky
06:39:44 <aidanh> if we only write them once, they must be in tags.h, but that makes them public
06:42:33 <aidanh> re unit tests influencing organisation, I'm pretty firmly convinced that that's probably the best thing that could happen
07:20:42 <aidanh> I think moving static function declarations to a clearly marked "private" section of their header file is perfectly workable
10:09:42 <advil> I don't know why I read SA sometimes but lol: "Meanwhile, the Crawl subreddit now explicitly bans criticism, or any discussion about disliking changes."
10:09:53 <advil> is there anything except that on the subreddit?
10:28:12 <ebering> later down the thread that poster got torn into by other goons
10:30:35 <Rotatell> Pestilence (L27 NaBe) ASSERT(mgrd(mons->pos()) == s) in 'show.cc' at line 275 failed. (Abyss:4)
10:33:21 <advil> yeah I saw
10:33:21 <advil> hadn't yet gotten there
10:33:21 <advil> it's just such a weird example of people's perceptions being shaped by their preexisting filters
10:34:40 <|amethyst> More like "The Crawl subreddit, while not explicitly, frowns upon praise, or any discussion about liking changes.  Bootlicking shill."
10:35:39 <advil> haha
10:35:56 <advil> is amethyst there you or is that someone else?
10:36:22 <|amethyst> someone else, I didn't realize there was an "amethyst" there
10:36:39 <advil> they are missing the punctuation
10:36:46 <advil> it did seem unlikely to me that it was you
10:38:00 <|amethyst> oh, I should get on the discord
10:38:04 <|amethyst> now that I have discord
10:38:47 <|amethyst> also, I need to remember the @crawlcode password (I've just stayed logged in for the past 3 years), so I can give access to other people
10:38:52 <advil> someone just posted on it the other day
10:38:58 <|amethyst> that was me :)
10:39:00 <advil> haha
10:39:09 <advil> I seem to remember that someone else had access at least?
10:39:10 <advil> not me
10:39:23 <|amethyst> wheals
10:39:30 <advil> with tweetdeck it's not actually to hard to have teams on a twitter account these days
10:39:34 <advil> *too hard
14:17:09 <Kelbell> shummie (L27 DgVM) ASSERT(attacker) in 'attack.cc' at line 351 failed. (Pan)
16:06:39 -!- aidanh_ is now known as aidanh
16:58:40 <ebering> advil: read your removals draft
16:58:44 <ebering> when do you plan to post it?
17:01:09 <advil> not sure
17:01:15 <advil> still needs some work
17:01:26 <ebering> I enjoyed reading it
17:01:43 <advil> thanks!
18:39:35 <Henzell> ArbitraryName (L5 DEWz) Crash caused by signal #15: Terminated (D:4)
22:18:22 <aidanh> advil: that stats script is neat; it'd be nice if we had a metrics server that can show how that changes over time
22:18:40 <aidanh> and things like code coverage, etc
22:40:26 <twelwe> need any great ideas?
22:44:13 <gammafunk> aidanh: sorry, I sort of missed the tail end of our conversation about the declarations. Do you recommend that I simply declare the relevant helpers in tags.h (allowing the unit tests to influence organisation for that better, as you say)
22:44:44 <gammafunk> aidanh: or do you recommend I just add the declarations to tags.cc as well? right now we're getting compilation warnings in gcc, so I have to do one or the other
22:47:20 <aidanh> gammafunk: I reckon having a single declaration at the bottom of tags.h is fine
22:47:39 <gammafunk> sounds good, will do
22:47:48 <aidanh> thanks!
23:35:16 <alexjurkiewicz> you could write a github action to run the script and add the output to a google sheet
23:35:45 <alexjurkiewicz> google sheets can have charts in them, and you "publish" charts as a static image URL that updates whenever the spreadsheet data does
23:36:13 <alexjurkiewicz> embed that image into a markdown doc explaining how the whole setup works either in the repo or the wiki
23:39:54 <aidanh> hm, that could work
23:40:52 <aidanh> won't work for historical data though
23:44:21 <aidanh> i wonder what to even google for... 'software development metrics tools' doesn't give much
23:48:57 <aidanh> I guess the modern open source solution is probably prometheus + grafana
23:55:52 <alexjurkiewicz> it could work for historical data if you just insert it manually 🤠
23:56:40 <alexjurkiewicz> i don't think you want a tool that is designed for metrics. Those things are designed for ingesting lots of data very frequently (every minute at the slowest)
23:56:47 <alexjurkiewicz> you want a spreadsheet