| --- Log opened Fri Jan 19 00:00:32 2018 |
| --- Log closed Fri Jan 19 00:05:44 2018 |
| --- Log opened Fri Jan 19 00:15:18 2018 |
| 00:15 | | TheWatcher [chris@GlobalOperator.Nightstar.Net] has joined #code |
| 00:15 | | Irssi: #code: Total of 33 nicks [23 ops, 0 halfops, 0 voices, 10 normal] |
| 00:15 | | mode/#code [+o TheWatcher] by ChanServ |
| 00:16 | < simon_> | sub options ($self) { |
| 00:16 | < simon_> | my @categories = grep { $_->is_online } $self->select_collection; |
| 00:16 | < simon_> | @categories = sort_by { $_->name } @categories; |
| 00:16 | < simon_> | return [ map { [ $_->name => $_->id ] } @categories ]; |
| 00:16 | < simon_> | } |
| 00:16 | | Irssi: Join to #code was synced in 67 secs |
| 00:16 | < simon_> | I wonder if there's a neater way to express this code. |
| 00:16 | | Reiver [quassel@Nightstar-ksqup0.co.uk] has joined #code |
| 00:16 | | mode/#code [+ao Reiver Reiver] by ChanServ |
| 00:17 | <&[R]> | By using something other than perl clearly |
| 00:17 | < simon_> | ;-) |
| 00:17 | | Kindamoody is now known as Kindamoody[zZz] |
| 00:18 | | Orthia [quassel@Nightstar-ksqup0.co.uk] has joined #code |
| 00:18 | | mode/#code [+o Orthia] by ChanServ |
| 00:41 | < simon_> | there's this optional parameter to a function called 'remove_advanced'. it means: if on a site with an ID that matches one particular regex, don't include the front page. if on a site with another specific ID, don't include pages for which the 'show_order' (an index value that indicates the way the pages are sorted in a list) >= 1000. |
| 00:42 | < simon_> | also, the code is self-documenting. |
| 00:43 | < simon_> | I'm pretty amazed by my boss, who wrote this, to get the idea to overload what 'remove_advanced' means for two entirely different, mutually exclusive purposes (you can't be on both sites). why not have 'remove_frontpage' and 'remove_... wait, that second definition is OBSCURE. |
| 00:46 | < simon_> | this way, the database table schema can be re-used for multiple sites, and special meaning can be encoded into one of them that is only addressed in the code. if the database schema were to be extended with another column called 'advanced', it'd default to 0 for all rows on one site. |
| 00:46 | < simon_> | I'm sure that was the rationale. least effort. |
| 00:56 | | Jessikat [Jessikat@Nightstar-6s0.159.132.82.IP] has joined #code |
| 01:23 | | celticminstrel [celticminst@Nightstar-m9434e.dsl.bell.ca] has joined #code |
| 01:23 | | mode/#code [+o celticminstrel] by ChanServ |
| 01:37 | <&[R]> | https://imgur.com/gallery/5sYsb |
| 02:26 | < McMartin> | Woot. It's taken me about a week and a half of near-random experimentation to get this API to work, but I just got a successful result from it. |
| 02:26 | < McMartin> | Time to send it off to be tested on devices that aren't sitting here on my desk. |
| 02:30 | < Jessikat> | I've already read more coding books in 2018 than the whole of 2017 |
| 02:47 | <&[R]> | Good job |
| 02:55 | <&ToxicFrog> | I've just started The Zen of Graphics Programming and Rolling in the Deep |
| 04:18 | | Vornotron [Vorn@Nightstar-1l3nul.res.rr.com] has joined #code |
| 04:18 | | mode/#code [+qo Vornotron Vornotron] by ChanServ |
| 04:30 | | bowlich [bowlich@Nightstar-su4ao3.kynda.net] has joined #code |
| 04:30 | | bowlich [bowlich@Nightstar-su4ao3.kynda.net] has quit [[NS] Quit: leaving] |
| 05:21 | | celticminstrel [celticminst@Nightstar-m9434e.dsl.bell.ca] has quit [[NS] Quit: KABOOM! It seems that I have exploded. Please wait while I reinstall the universe.] |
| 05:30 | | Vash [Vash@Nightstar-1l3nul.res.rr.com] has joined #code |
| 05:41 | | Jessikat` [Jessikat@Nightstar-ro5.klt.132.82.IP] has joined #code |
| 05:43 | | Jessikat [Jessikat@Nightstar-6s0.159.132.82.IP] has quit [Ping timeout: 121 seconds] |
| 05:46 | | * McMartin juggles registers |
| 06:11 | | * simon_ registers juggles |
| 06:19 | | himi [sjjf@Nightstar-1drtbs.anu.edu.au] has quit [Ping timeout: 121 seconds] |
| 06:27 | < McMartin> | 741 bytes so far |
| 06:34 | < McMartin> | And the core logic and interaction is done; everything left is display cleanup. |
| 06:40 | < McMartin> | I've probably implemented the Game of Life a dozen times, but it's still unusually gratifying to see it spinning away inside a Game Boy emulator |
| 06:40 | < McMartin> | This is the first one where I've added a button to shoot gliders at boring parts of the system though |
| 06:45 | | Vornlicious [Vorn@Nightstar-a9pfkr.sub-174-211-10.myvzw.com] has joined #code |
| 06:47 | | Vorntastic [Vorn@Nightstar-1l3nul.res.rr.com] has quit [Ping timeout: 121 seconds] |
| 07:09 | < simon_> | the last time I implemented Game of Life was for a compiler course test suite. |
| 07:23 | < Vornlicious> | I don't think I've ever implemented gol |
| 07:31 | | Vash [Vash@Nightstar-1l3nul.res.rr.com] has quit [Connection closed] |
| 07:39 | | Vornotron [Vorn@Nightstar-1l3nul.res.rr.com] has quit [Ping timeout: 121 seconds] |
| 08:36 | | Jessikat` is now known as Jessikat |
| 08:56 | | * macdjord did some work on an ultra-efficient cache-based high-speed GoL evolver in C, but never finished it. |
| 08:58 | < McMartin> | Yeah, I'm not super-interested in things like Golly |
| 08:58 | < McMartin> | This is just an excuse to write a complete program that can live as a Game Boy ROM |
| 09:04 | | macdjord is now known as macdjord|slep |
| 10:34 | | Degi [Degi@Nightstar-othrpl.dyn.telefonica.de] has joined #code |
| 10:47 | < simon_> | sigh. the thing I spend most time on at work is figuring out how to create test objects. |
| 10:50 | <&[R]> | Could be worse |
| 10:50 | | Kindamoody[zZz] [Kindamoody@Nightstar-eubaqc.tbcn.telia.com] has quit [Ping timeout: 121 seconds] |
| 10:56 | | * [R] wonders what possesses people who are having problems, and getting errors to /not/ read the errors? |
| 11:00 | <@TheWatcher> | "Liliputian snotwasel foxtrot omegaforce" |
| 11:00 | <@TheWatcher> | *snotweasel |
| 11:02 | <&[R]> | wat |
| 11:07 | < Vornlicious> | https://m.imgur.com/gallery/H0uVqFe |
| 11:09 | <&[R]> | Ah |
| 11:09 | <@TheWatcher> | It's originally from http://blog.johnath.com/2007/10/11/todo-break-internet/ but appears in a better form in a talk given at USENIX Enigma in 2016: https://www.youtube.com/watch?v=wxeIrkxRDMA |
| 11:10 | <&[R]> | I was specifically thinking about this guy who is having trouble connecting to a DB via PHP and isn't reading the error messages that get spewed out. |
| 11:10 | <@TheWatcher> | Oh, that's just being a freaking dumbass, then~ |
| 11:18 | <&[R]> | Pretty much yeah |
| 11:35 | < Jessikat> | Today: concurrency is hard |
| 11:35 | < Jessikat> | I think I'm just going to mandate that no one gets more than one processor core ever |
| 11:40 | < Vornlicious> | Even just one processor core tends to be a problem these days |
| 12:04 | <@TheWatcher> | Y'know, I think that growing up in the 80s in the UK has uniquely equipped me to approach concurrency in the correct frame of mind. |
| 12:06 | <@TheWatcher> | Because the mere mention of threads causes anxiousness bordering on terror, and a deep premonition of impending apocalypse |
| 12:11 | | Jessikat` [Jessikat@Nightstar-8gd.equ.132.82.IP] has joined #code |
| 12:15 | | Jessikat [Jessikat@Nightstar-ro5.klt.132.82.IP] has quit [Ping timeout: 121 seconds] |
| 13:27 | | Emmy [Emmy@Nightstar-9p7hb1.direct-adsl.nl] has joined #code |
| 14:16 | <&ToxicFrog> | TheWatcher: may I quote you elsewhere? |
| 14:17 | | * ToxicFrog concurrently snuggles Jessikat` |
| 14:19 | <@TheWatcher> | TF: feel free :) |
| 14:20 | | Jessikat` is now known as Jessikat |
| 14:21 | < Jessikat> | ToxicFrog, how would that actually work? :o |
| 14:27 | <&ToxicFrog> | Jessikat: comfortably, I hope! |
| 14:30 | < Vornlicious> | Well it is different io objects. As long as task switching happens fast enough to keep both buffers full you're fine |
| 14:55 | | Netsplit Krikkit.Nightstar.Net <-> Deepthought.Nightstar.Net quits: @macdjord|slep, @Alek |
| 14:55 | | mode/#code [+o macdjord|slep] by ChanServ |
| 14:55 | | Netsplit over, joins: macdjord|slep |
| 15:00 | | Alek [Alek@Nightstar-7or629.il.comcast.net] has joined #code |
| 15:00 | | mode/#code [+o Alek] by ChanServ |
| 15:27 | | Vornicus [Vorn@Nightstar-1l3nul.res.rr.com] has joined #code |
| 15:27 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
| 15:27 | | Degi [Degi@Nightstar-othrpl.dyn.telefonica.de] has quit [Connection closed] |
| 16:37 | | Kindamoody[zZz] [Kindamoody@Nightstar-eubaqc.tbcn.telia.com] has joined #code |
| 16:37 | | mode/#code [+o Kindamoody[zZz]] by ChanServ |
| 16:50 | | Kindamoody[zZz] is now known as Kindamoody |
| 17:31 | | Kindamoody [Kindamoody@Nightstar-eubaqc.tbcn.telia.com] has quit [Client exited] |
| 17:36 | | Kindamoody|autojoin [Kindamoody@Nightstar-eubaqc.tbcn.telia.com] has joined #code |
| 17:36 | | mode/#code [+o Kindamoody|autojoin] by ChanServ |
| 17:36 | | Kindamoody|autojoin is now known as Kindamoody |
| 17:50 | | Jessikat [Jessikat@Nightstar-8gd.equ.132.82.IP] has quit [[NS] Quit: Bye] |
| 17:56 | | Vornlicious [Vorn@Nightstar-a9pfkr.sub-174-211-10.myvzw.com] has quit [[NS] Quit: Bye] |
| 17:56 | | Vorntastic [Vorn@Nightstar-1l3nul.res.rr.com] has joined #code |
| 18:13 | | * McMartin quietly leaves the ZX81 display mechanism on TheWatcher's desk, flees cackling into the depths of the night |
| 18:14 | < McMartin> | (In which its nonstandard characterset has 0x76 as newline because it is, in fact, crucially important that its newline character is the same as the opcode of the HALT instruction) |
| 18:24 | < McMartin> | Also, unfortunately for me, I should probably be making that Game Boy code more concurrent. |
| 18:24 | < McMartin> | Because the alternative is flinging continuations around in places I really don't want to |
| 18:39 | <&[R]> | <Xaquseg> is it just me or have we somehow gotten into a situation where everything is so smart it's impossible to actually use? |
| 18:39 | <&[R]> | Anyone else feel that way? |
| 18:39 | | * [R] does sometimes |
| 18:40 | < workeegs> | ^ |
| 18:43 | <@ErikMesoy> | ^^ |
| 18:44 | <&[R]> | Like gradle, had an issue where it wouldn't build (or rather it would /hang/ on a build). Had to strace it to find out that if all the things it tries to download 404 it just hangs. No error messages, nothing. |
| 18:44 | <&[R]> | I shouldn't have to strace things to debug an issue D: |
| 18:52 | <@Attilla> | everything has too sleek a UI that it just covers up any problems? |
| 18:57 | <&[R]> | No, more like they do very complicated things and don't log anything |
| 18:59 | <&ToxicFrog> | McMartin: ...why does \n have to be the same as HALT? |
| 19:18 | < McMartin> | ToxicFrog: Because the character lookup code uses the CPU as a fetch mechanism, jumping directly to the display file and "executing" it, and replacing each "instruction" (read: character) loaded with NOP before it actually hits the CPU |
| 19:19 | < McMartin> | HALT is passed through, which means "wait until next interrupt" which is of course tuned to arrive at the start of the next line. |
| 19:20 | < McMartin> | Its successors were marginally less ridiculous about this, though I remain Not A Fan of their bitmap layouts |
| 19:21 | < McMartin> | (IIRC some other quirk of the ZX81's interrupt registers is used to actually get the correct character data to appear by having it asserting on the data lines as a side effect or something; interrupt processing on the Z80 remains something of a mystery to me because it's usually 100% snaffled by the BIOS-equivalent.) |
| 19:22 | < McMartin> | (I spent the morning waiting for rush hour traffic to dissipate reverse-engineering a Sega Genesis cartridge's startup code, though, and the portion of that related to the Z80 coprocessor has put me on notice that I don't get to stay ignorant of it for too long.) |
| 19:23 | < McMartin> | The Game Boy's Z80 variant has a much simpler interrupt system |
| 19:23 | < McMartin> | (and, to be clear, HALT is not actually a shutdown operation; it just sleeps the CPU until the next interrupt arrives. So that's real handy for avoiding busyloops if you're mostly responding to VBLANK, etc.) |
| 19:25 | < McMartin> | (But it's certainly always exciting to be disassembling some code out of a data block and discovering that it's using instructions that reference registers that never appeared in any of the books you read) |
| 19:31 | <&[R]> | http://blog.nobl.ca/fuse.log.txt <-- I'm not the one being silly here am I? |
| 19:36 | < McMartin> | I'm not 100% positive but I think by "absolute link" he means "link to an absolute path on the remote system" |
| 19:36 | | * abudhabi mutters darkly about sites too aggressive in logging the users out. |
| 19:36 | < McMartin> | I suspect that means the part he's afraid of is "what if that absolute path *does* exist on the local system" |
| 19:36 | <&[R]> | abudhabi: Patreon is /horrible/ for that |
| 19:37 | <&[R]> | I think as soon as you log in from one host, it invalidates all other sessions you might have. |
| 19:37 | < McMartin> | That doesn't look to me like anyone's being silly, just frustrated and possibly confused, and I don't know enough about using fuse to do networked filesystems to really have an opinion |
| 19:39 | <@abudhabi> | [R]: I can understand money-sites doing that, to an extent. I wouldn't want my bank to be lenient with sessions. |
| 19:39 | <@abudhabi> | I can understand shops doing that, even if re-authing for purchases is the better solution IMO. |
| 19:40 | <@abudhabi> | I don't understand sites like fuckin' Roll20. |
| 19:40 | <&[R]> | Fair enough, but I'd like to not have to log in everytime, since they do something to make FF not autofill the UN/PW, so it's a hassle. |
| 19:41 | <@abudhabi> | My bank is so strict that its solution is strictly per-tab basis. |
| 19:41 | <@abudhabi> | If you close the tab, and reopen it, session is lost. |
| 19:41 | <@abudhabi> | And you'll get logged out after like 10 mins of inactivity. |
| 19:42 | <@abudhabi> | (In that tab, that is. Activity in other tabs doesn't count for session keeping.) |
| 19:42 | <&[R]> | I think mine's similar... except I can't have a PW longer than 8 charactes. |
| 19:42 | <@abudhabi> | Mine requires a long password, and demands a random selection of characters from said password at login. |
| 19:43 | <&[R]> | That's... weird |
| 19:43 | <@abudhabi> | Like, "type in the 1st, 5th, 6th, 10th and 18th character". |
| 19:43 | <&[R]> | I think that's to beat keyloggers |
| 19:43 | <&[R]> | But it's so weird |
| 19:50 | | Jessikat [Jessikat@Nightstar-bt5k4h.81.in-addr.arpa] has joined #code |
| 20:25 | <&[R]> | Updated the log |
| 20:25 | <&[R]> | He's a douche |
| 20:27 | <&ToxicFrog> | [R]: ThirdChoiceNick is being a jackass, but my best guess here is that either (a) they are concerned about what happens if the file does exist on the local system, or (b) they think the symlinks should be resolved server-side. |
| 20:27 | < McMartin> | Oh well |
| 20:27 | <&ToxicFrog> | But since they refuse to actually say, who knows? |
| 20:27 | < McMartin> | (a) seems like a legitimate attack vector especially if the symlink's permissions are checked instead of the targets |
| 20:27 | < McMartin> | But the "US" in "FUSE" should mean that isn't true |
| 20:28 | | macdjord|slep [macdjord@Nightstar-a1fj2k.mc.videotron.ca] has quit [[NS] Quit: Live Long and Prosper.] |
| 20:28 | <&ToxicFrog> | symlinks always have permissions 777 and in practice the target's permissions are what's actually used. |
| 20:28 | <&ToxicFrog> | Oh, and for (b) sshfs has -o follow_symlinks and -o transform_symlinks |
| 20:29 | <&ToxicFrog> | So as best I can tell the answer is "they think (a) is a problem in some poorly defined way that they refuse to communicate" |
| 20:31 | | macdjord [macdjord@Nightstar-a1fj2k.mc.videotron.ca] has joined #code |
| 20:31 | | mode/#code [+o macdjord] by ChanServ |
| 20:32 | <&[R]> | And more |
| 20:32 | <&ToxicFrog> | I'm honestly having trouble coming up with legitimate attack vectors for (a) that couldn't also be exploited in the absense of absolute symlinks |
| 20:33 | <&ToxicFrog> | If they have a legitimate security issue here, the FUSE and/or SSHFS developers (depending on which layer it's in) should address it, but they haven't actually demonstrated that to be the case; and whatever their perceived issue is, they can mitigate it by mounting with -o follow_symlinks |
| 20:34 | <&[R]> | Yeah |
| 20:36 | < McMartin> | ToxicFrog: General path traversal attack, yeah |
| 20:36 | <&ToxicFrog> | In the meantime, my canned response to "I found a security issue, but I won't tell you what it is and I'm being super condescending about it" is "good for you, now fuck off while the adults get some work done" |
| 20:37 | <&[R]> | ... and the finale |
| 20:37 | <&ToxicFrog> | ...and now they're PMing me on freenode, wtf |
| 20:37 | <&[R]> | Sorry |
| 20:37 | <&[R]> | :/ |
| 20:39 | <&[R]> | <][_R_][> Have fun reporting a bug and giving no details. We'll see how that plays out. |
| 20:39 | < workeegs> | shrug emoticon |
| 21:43 | <&[R]> | Now that I think about it, (b) could be an attack vector possibly... hmm. Maybe not. (thought about making a symlink while the system is setup to present symlinks as the files/directories they link to) |
| 21:44 | <&[R]> | That wouldn't really give you access to something you didn't already have access to |
| 22:27 | <&[R]> | MS really needs to stop this "for bussiness" branding crap. If you're going to release two products, name them differently like /everyone/ else does. |
| 22:30 | | Degi [Degi@Nightstar-othrpl.dyn.telefonica.de] has joined #code |
| 23:06 | | McMartin [mcmartin@Nightstar-rpcdbf.sntcca.sbcglobal.net] has quit [[NS] Quit: brb] |
| 23:15 | | McMartin [mcmartin@Nightstar-rpcdbf.sntcca.sbcglobal.net] has joined #code |
| 23:15 | | mode/#code [+ao McMartin McMartin] by ChanServ |
| --- Log closed Sat Jan 20 00:00:40 2018 |