| --- Log opened Sat Jan 26 00:00:00 2008 |
| --- Log closed Sat Jan 26 00:25:16 2008 |
| --- Log opened Sat Jan 26 00:25:50 2008 |
| 00:25 | | TheWatcher[zZzZ] [~chris@Nightstar-29731.dsl.in-addr.zen.co.uk] has joined #code |
| 00:25 | | Irssi: #code: Total of 15 nicks [10 ops, 0 halfops, 1 voices, 4 normal] |
| 00:25 | | mode/#code [+o TheWatcher[zZzZ]] by ChanServ |
| 00:26 | | Irssi: Join to #code was synced in 37 secs |
| 00:49 | | Thaqui [~Thaqui@Nightstar-14364.jetstream.xtra.co.nz] has joined #code |
| 00:49 | | mode/#code [+o Thaqui] by ChanServ |
| 01:48 | | Thaqui [~Thaqui@Nightstar-14364.jetstream.xtra.co.nz] has left #code [Leaving] |
| 01:59 | | Serah [~Z@87.72.35.ns-26506] has joined #Code |
| 02:04 | | cs70wac [~cs70wac@Nightstar-25074.ucsd.edu] has joined #code |
| 02:07 | | gnolam [lenin@85.8.5.ns-20483] has quit [Quit: z?] |
| 02:09 | | cs70wac is now known as JPL |
| 02:17 | | JPL [~cs70wac@Nightstar-25074.ucsd.edu] has quit [Quit: Audi 5000] |
| 03:30 | | C_tiger [~c_wyz@96.232.26.ns-4600] has joined #code |
| 05:11 | | C_tiger [~c_wyz@96.232.26.ns-4600] has quit [Ping Timeout] |
| 05:26 | | C_tiger [~c_wyz@96.232.26.ns-4600] has joined #code |
| 05:42 | | C_tiger [~c_wyz@96.232.26.ns-4600] has quit [Quit: And away she goes!] |
| 05:44 | | C_tiger [~c_wyz@96.232.26.ns-4600] has joined #code |
| 06:00 | | C_tiger [~c_wyz@96.232.26.ns-4600] has quit [Quit: And away she goes!] |
| 06:10 | | C_tiger [~c_wyz@96.232.26.ns-4600] has joined #code |
| 07:01 | | * Vornicus does battle with unit testing. |
| 07:09 | <@Vornicus> | The worst bit, really, is the massive amount of data that is necessary for testing of complex manipulations to be viable. |
| 07:12 | < jerith> | Yeah. |
| 07:12 | | * jerith spends quite a lot longer (programmer and CPU cycles) setting up data than running tests. |
| 07:13 | < jerith> | Eventually I crafted something that looked similar to a stripped-down prod dataset and just used that for everything with minor mods where necessary. |
| 07:24 | <@Vornicus> | Well, really, the trick is, "construct this data, then /do the algorithm by hand/ to make the answer" |
| 07:26 | < jerith> | Ah. |
| 07:40 | | C_tiger [~c_wyz@96.232.26.ns-4600] has quit [Ping Timeout] |
| 07:46 | | C_tiger [~c_wyz@96.232.26.ns-4600] has joined #code |
| 08:18 | | * Vornicus hunts around, finds a Thing He Needs in the python's itertools. |
| 08:20 | | C_tiger [~c_wyz@96.232.26.ns-4600] has quit [Ping Timeout] |
| 08:21 | <@McMartin> | ? |
| 08:22 | <@Vornicus> | I'm trying to decrypt the uplink data files, which appears to require XORing with the string "REDSHIRT" repeatedly |
| 08:23 | <@Vornicus> | itertools.repeat(), combined with zip(), will make the loop trivial. |
| 08:24 | <@Vornicus> | Indeed, it makes the entire decrypter a one-liner |
| 08:28 | <@McMartin> | Heh |
| 08:33 | <@Vornicus> | ...hrm. that file doesn't get recognized by file, either. Dunno. |
| 08:35 | <@Vornicus> | And, actually, it's cycle that I wanted, not repeat, and I should learn to read. |
| 08:39 | | You're now known as TheWatcher |
| 08:40 | <@Vornicus> | file('de-data.dat','w').write(''.join(chr(ord(a)^ord(b)) for a,b in zip(file('data.dat','r').read(),itertools.cycle('REDSHIRT')))) # for the record |
| 09:02 | | Vornicus is now known as Vornicus-Latens |
| 11:00 | | You're now known as TheWatcher[afk] |
| 12:46 | | gnolam [lenin@Nightstar-10613.8.5.253.static.se.wasadata.net] has joined #Code |
| 12:47 | | mode/#code [+o gnolam] by ChanServ |
| --- Log closed Sat Jan 26 13:37:10 2008 |
| --- Log opened Sat Jan 26 14:42:00 2008 |
| 16:41 | <@ToxicFrog> | Hmm. It strikes me that hibernation is much, much harder than it sounds at first glance. |
| 17:01 | | gnolam [lenin@Nightstar-10613.8.5.253.static.se.wasadata.net] has quit [Ping Timeout] |
| 17:03 | | gnolam [lenin@Nightstar-10613.8.5.253.static.se.wasadata.net] has joined #Code |
| 17:03 | | mode/#code [+o gnolam] by ChanServ |
| 17:05 | | Doctor_Nick [~nick@68.42.57.ns-4111] has joined #code |
| 17:20 | | You're now known as TheWatcher[afk] |
| 17:24 | | C_tiger [~c_wyz@96.232.26.ns-4600] has joined #code |
| 17:42 | | C_tiger [~c_wyz@96.232.26.ns-4600] has quit [Connection reset by peer] |
| 17:50 | | C_tiger [~c_wyz@96.232.26.ns-4600] has joined #code |
| 17:56 | | Doctor_Nick [~nick@68.42.57.ns-4111] has quit [Quit: Konversation terminated!] |
| 18:12 | <@MyCatVerbs> | ToxicFrog: it sounds, to me, quite hard even at first glance. |
| 18:12 | <@MyCatVerbs> | ToxicFrog: like, how do you deal with processes that have mlock(2)'d certain pages of their address space? |
| 18:14 | <@MyCatVerbs> | ToxicFrog: if you don't honour mlock()'s guarantee that mlock()'d pages won't ever be swapped to disk, you create godawful security vulnerabilities. ssh-agent *needs* that guarantee for safety. ?_? |
| 18:14 | <@ToxicFrog> | The assumption is that hibernate does not respect mlock, I believe. |
| 18:15 | <@MyCatVerbs> | In which case hibernation fucks users. |
| 18:15 | <@ToxicFrog> | Either that or mlock'd pages are written as zeroes. |
| 18:15 | <@ToxicFrog> | Hibernation fucks users in either case. |
| 18:15 | <@ToxicFrog> | There is no safe way to reconcile hibernation with mlock. |
| 18:15 | <@MyCatVerbs> | This is true, but losing control of your ssh keys is a permanent issue. |
| 18:15 | <@ToxicFrog> | You have to choose either a security hit or a stability hit. |
| 18:15 | <@MyCatVerbs> | Losing the running copy of ssh-agent is a temporary annoyance. |
| 18:16 | <@ToxicFrog> | True. So you write them as zeroes. |
| 18:16 | <@ToxicFrog> | (or, if you want clearly defined behaviour, you kill all processes with mlocked pages) |
| 18:17 | <@ToxicFrog> | Anyways, what I was actually thinking of was the hardware. |
| 18:17 | | * MyCatVerbs would tend to prefer the latter. |
| 18:17 | <@ToxicFrog> | All your volatile storage is gone except memory. |
| 18:17 | <@ToxicFrog> | Your wireless card and crypto accelerator are looking around and wondering why the driver is telling them to use firmware that hasn't been loaded yet. |
| 18:17 | <@ToxicFrog> | Your video card is being told to use shaders it doesn't have on chip anymore. |
| 18:19 | <@MyCatVerbs> | Jah. So you have to make the kernel send gotosleep() to all your drivers, then audit all your drivers to make sure they respond correctly to gotosleep(). :/ |
| 18:20 | <@ToxicFrog> | Yeah. |
| 18:22 | | C_tiger [~c_wyz@96.232.26.ns-4600] has quit [Quit: And away she goes!] |
| 18:56 | | You're now known as TheWatcher |
| 19:19 | | C_tiger [~c_wyz@96.232.26.ns-4600] has joined #code |
| 19:40 | | C_tiger [~c_wyz@96.232.26.ns-4600] has quit [Quit: And away she goes!] |
| 19:45 | | C_tiger [~c_wyz@96.232.26.ns-4600] has joined #code |
| 19:52 | | C_tiger [~c_wyz@96.232.26.ns-4600] has quit [Connection reset by peer] |
| 20:09 | | C_tiger [~c_wyz@96.232.26.ns-4600] has joined #code |
| 20:25 | | * jerith prods Vorn with a codestick. |
| 20:30 | | You're now known as TheWatcher[afk] |
| 20:51 | <@Vornicus-Latens> | ow |
| 21:23 | | Vornicus-Latens is now known as Vornicus |
| 21:27 | | Vornicus is now known as Vorn-Testing |
| 21:27 | | Vorn-Testing is now known as Vornicus |
| 21:28 | | Vornicus is now known as Vorn-Testing |
| 21:29 | | Vorn-Testing is now known as Vornicus |
| 22:00 | | AnnoDomini [AnnoDomini@83.21.15.ns-25782] has quit [Ping Timeout] |
| 22:08 | <@Vornicus> | Who was it that was trying to make a 2-player settlers board? |
| 22:08 | <@Vornicus> | Was it Der? |
| 22:08 | | AnnoDomini [AnnoDomini@83.21.46.ns-3566] has joined #Code |
| 22:08 | | mode/#code [+o AnnoDomini] by ChanServ |
| 23:15 | <@Vornicus> | it was Der. |
| 23:29 | | * McMartin crackles with SRMD |
| 23:29 | | * McMartin has just built an OS X app bundle for UQM that is made of nothing but symlinks. |
| 23:30 | <@McMartin> | This horrific construct cannot move or breathe, but it automatically updates itself to reflect new recompiles or SVN updates to the content without rebundling! |
| 23:39 | <@ToxicFrog> | |
| 23:44 | | * EvilDarkLord is learning Perl for a programming course consisting of two Perl assignments, might have the wrong mentality when actually trying to structure his code. |
| 23:46 | <@EvilDarkLord> | Also, augh at context-dependent methods of referring to a variable. |
| 23:51 | <@McMartin> | That reminds me. |
| 23:51 | <@McMartin> | Vorn! |
| 23:51 | <@McMartin> | Is there a way to start a non-app-bundle as if it were one? |
| 23:51 | | * McMartin is having problems where SDL programs run from the Terminal have the Terminal eat all key input, not the application |
| 23:52 | <@McMartin> | Faking an .app as above and running "open Whatever.app" works, but is sleazy. |
| 23:59 | <@ToxicFrog> | EvilDarkLord: "context-dependent methods"? |
| --- Log closed Sun Jan 27 00:00:02 2008 |