| --- Log opened Wed Jun 13 00:00:47 2012 |
| 00:05 | <&McMartin> | AHAHAHAHAHAHA |
| 00:05 | <&McMartin> | AHAHAHA |
| 00:05 | <&McMartin> | hAHAHA |
| 00:05 | <&McMartin> | *gasp* |
| 00:05 | <&McMartin> | HAHAHAHAHHA |
| 00:05 | <&McMartin> | So good |
| 00:05 | <&McMartin> | They made the bash shell prompt do the READY. thing |
| 00:05 | < Rhamphoryncus> | yes, these are good candies |
| 00:05 | < Rhamphoryncus> | oh, not what you meant |
| 00:05 | < Rhamphoryncus> | hmm? |
| 00:06 | <&McMartin> | This Commodore-themed Mint distro |
| 00:06 | <&McMartin> | It is *hilarious* |
| 00:06 | < Rhamphoryncus> | ah |
| 00:11 | <&McMartin> | I will not be satisfied until I have Ophis running on it |
| 00:11 | <&McMartin> | Not that I expect this to be difficult; It's got Python 2.6 on it already |
| 00:16 | <@ToxicFrog> | McMartin: yes. And yeah, if I were doing this on An Actual Computer I would not even have considered C. |
| 00:16 | <@ToxicFrog> | However, I've already written that version as one of my undergrad projects~ |
| 00:16 | <&McMartin> | Heh |
| 00:16 | <&McMartin> | C emitter~ |
| 00:17 | <&McMartin> | Is Mint an Ubuntu variant or just another Debian descendant? |
| 00:17 | <@Tamber> | Since they have a Mint Debian, I'd presume the normal one is an Ubuntu variant. |
| 00:18 | | * McMartin nods |
| 00:18 | <@ToxicFrog> | Both; mainline Mint is Ubuntu derived, the RR version is Debian |
| 00:18 | <&McMartin> | I guess I'll find out if this Mint variant is more Ubuntu-y or more Debian-y. |
| 00:19 | <&McMartin> | Once I put in the VMware tools. |
| 00:20 | <&McMartin> | The installer looks Ubuntu-y |
| 00:20 | <&McMartin> | Except for being blue and blocky-fonted, because lol |
| 00:26 | | You're now known as TheWatcher[T-2] |
| 00:27 | <@ToxicFrog> | I may need to try this out at some point just for the hell of it~ |
| 00:27 | <@ToxicFrog> | How big is it? |
| 00:28 | | You're now known as TheWatcher[zZzZ] |
| 00:29 | | gnolam [lenin@Nightstar-202a5047.priv.bahnhof.se] has quit [[NS] Quit: Reboot] |
| 00:32 | | gnolam [lenin@Nightstar-202a5047.priv.bahnhof.se] has joined #code |
| 00:36 | <&McMartin> | Two DVD torrents, but Just The First Disc is a working LiveDVD. |
| 00:36 | <&McMartin> | And kind of looks like just the first disc also gets you bootstrapped to the point where you no longer need DVDs to update stuff, because synaptic |
| 00:42 | <@ToxicFrog> | Aah. I was kind of pondering throwing it onto my bootstick for laughs~ |
| 00:43 | | Derakon[AFK] is now known as Derakon |
| 00:44 | < Rhamphoryncus> | Maybe some sort of weighting scheme... have each face store the slope which implies the desired vertices, then pick the median of those vertices |
| 00:47 | < Rhamphoryncus> | A slope implies that face is a triangle with 3 vertices, rather than just a central point |
| 00:49 | < Rhamphoryncus> | maybe RGB9_E5 XD |
| 00:50 | < Rhamphoryncus> | Which is 3 14-bit floats with a 9 bit mantissa and 5 bit exponent, but the exponent is the same for all so it adds up to only 32 bits |
| 00:50 | <&McMartin> | TF: Well, if it's a big bootstick I think it will work~ |
| 00:52 | < Rhamphoryncus> | Since I don't actually want an exponential ulp I'd do a fixed point version: 17 bits for the barycenter, 5 bits of difference for each of those vertices |
| 00:53 | < Rhamphoryncus> | Maybe I want the vertices to be exponential.. |
| 00:53 | <@ToxicFrog> | McMartin: 4GB |
| 00:53 | <@ToxicFrog> | Of which currently 40MB is taken up with bootable stuff~ |
| 00:54 | <&McMartin> | Yeah, OK, no, it's too big for that. |
| 01:14 | < Rhamphoryncus> | Maybe just store the normal using a spherical latitude/longitude system |
| 01:15 | | * McMartin exports the VM. Let's see how big it really is. |
| 01:20 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 01:21 | < Rhamphoryncus> | I think that'd work. My half-assed random map generator will do Bad Things (just white noise right now), but it should retain the right shape |
| 01:27 | | io|driving is now known as iospace |
| 01:32 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has joined #code |
| 01:32 | | mode/#code [+o himi] by ChanServ |
| 01:38 | <&McMartin> | Answer: 5GB |
| 01:46 | | rms [rstamer@Nightstar-e681a855.org] has joined #code |
| 01:46 | | mode/#code [+o rms] by ChanServ |
| 02:06 | | Kindamoody[zZz] is now known as Kindamoody |
| 02:16 | | Attilla [Obsolete@Nightstar-d5f09953.as43234.net] has quit [Ping timeout: 121 seconds] |
| 02:18 | <@ToxicFrog> | Welp |
| 02:26 | | celticminstrel [celticminst@Nightstar-5d22ab1d.cable.rogers.com] has joined #code |
| 02:30 | | Kindamoody is now known as Kindamoody|out |
| 02:38 | <&McMartin> | Oh man, this is so hilariously wrong |
| 02:40 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 02:42 | <&McMartin> | https://hkn.eecs.berkeley.edu/~mcmartin/vision.png |
| 02:47 | <~Vornicus> | ...silliness |
| 02:47 | <&McMartin> | so good |
| 02:48 | <&Derakon> | Oh, so -- a name I heard on NPR yesterday, one of their producers, I think. Monty Carlos. |
| 02:53 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has joined #code |
| 02:53 | | mode/#code [+o himi] by ChanServ |
| 03:06 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 03:19 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has joined #code |
| 03:19 | | mode/#code [+o himi] by ChanServ |
| 03:38 | | Noah is now known as NoahTheVeganZombie |
| 04:20 | < celticminstrel> | So if I have classes A, B, C, D, where B and C both inherit from A and D inherits from virtual B and virtual C, does D have two instances of A or one? |
| 04:20 | <&Derakon> | Agh multiple inheritance~ |
| 04:24 | < celticminstrel> | XD |
| 04:25 | | * celticminstrel applies Google. |
| 04:26 | < celticminstrel> | That said, I'm not entirely sure multiple inheritance is the right thing to do here. |
| 04:33 | <@ToxicFrog> | celticminstrel: this is known as "diamond inheritance" and is the main argument against allowing multiple inheritance, especially in C++~ |
| 04:33 | | NoahTheVeganZombie is now known as Noah |
| 04:34 | < celticminstrel> | Well, that was kind my question; would that make a diamond, or do you need A to be a virtual base in order to do that? |
| 04:34 | < celticminstrel> | ^kinda |
| 04:37 | < celticminstrel> | Hm, after reading some stuff I think multiple inheritance doesn't really add anything useful to this situation. |
| 04:40 | < celticminstrel> | And now I've found the answer to the actual question, which is that no, it's B and C that need to inherit virtually. |
| 05:08 | <&McMartin> | MI almost never adds anything useful. The closest thing to utility it gives is mixins, and mixins are an argument for duck typing and protocols as much as anything else -_- |
| 05:38 | < celticminstrel> | "Almost never" is almost certainly an exaggeration. |
| 05:39 | <&McMartin> | I'm an OO skeptic, more or less |
| 05:39 | <&McMartin> | Not really a fan of inheritance, prefer to do encapsulation in other ways when I want to do it at all. |
| 05:41 | <&McMartin> | Anyway, TF was slightly inaccurate above: If there is only one instance of A in D, then you have diamond inheritance; if there are two instances of A (one in B and one in C) then you have more traditional multiple inheritance. |
| 05:41 | <&McMartin> | Both options are provably incorrect in common cases. |
| 05:41 | <&Derakon> | Straight-up inheritance comes in handy regularly. |
| 05:41 | <&Derakon> | Fancier stuff is usually unnecessary though. |
| 05:41 | <&McMartin> | My rule of thumb is that if your inheritance graph is more than three deep something has gone horribly wrong |
| 05:42 | < celticminstrel> | I should probably deal with integer overflow here... is it guaranteed that the result will always be less than the operands? |
| 05:42 | <&McMartin> | Also, as noted, I tend to prefer protocols to true inheritance. |
| 05:42 | < celticminstrel> | It's unsigned, as a note. |
| 05:42 | < celticminstrel> | And it's multiplication. |
| 05:42 | <&Derakon> | Protocols can sometimes be handled by inheritance though. |
| 05:42 | <&McMartin> | Derakon: Yes, but given the choice I'll pick protocols, views, and composition over inheritance about 95% of the time in my designs. |
| 05:43 | < celticminstrel> | (My inheritance graph is, incidentally, three deep in most places, though a couple are on a fourth level. :P ) |
| 05:43 | <&Derakon> | One example -- for Pyrel, the base Thing class handles basically everything related to sticking game entities into the various query-able containers. |
| 05:43 | <&Derakon> | Everything that inherits from Thing, which includes some fairly disparate types, gets all that for free. |
| 05:43 | <&McMartin> | Yeah, implementation sharing is fine in my book |
| 05:44 | <&McMartin> | But you can also do that with modules and functions with a halfway decent type system. |
| 05:44 | <&McMartin> | I use inheritance in Ophis, because I need the notion of passes having a queryable namespace full of functions. |
| 05:44 | <&Derakon> | ...so, what, the "put thing into container" and "take thing out of all containers it's in" and so on would be imposed from outside? |
| 05:45 | <&McMartin> | ... yes? You've just named cons and filter. |
| 05:45 | <&McMartin> | ... sort of. |
| 05:46 | < celticminstrel> | So is "if( (a * b) / a != b)" a sufficient test for overflow? |
| 05:46 | <&Derakon> | McM: yeah, but I find it useful to have that stuff be alongside the object instead of handled exterior to it, at least where Pyrel is concerned. |
| 05:47 | <&McMartin> | Yeah, I'm aware that OO is a discipline |
| 05:47 | <&McMartin> | My design instincts are just noticably more aligned with the data-directed school, which is admittedly OO Turned Sideways. |
| 05:48 | <&Derakon> | Pyrel has plenty of data-driven OO~ |
| 05:48 | <&McMartin> | Most of the other annoying things OO does boil down to "I prefer functional styles when I can get away with it" |
| 05:48 | <&McMartin> | Right, what I mean is, I'd rather hang nouns to verbs than hang verbs onto nouns. |
| 05:48 | < celticminstrel> | Well, I shall try it, anyway. |
| 05:49 | <&McMartin> | This means, for me, a lot of Strategy pattern, a lot of Visitor pattern, and a lot of architectures that do things like publish/subscribe and chain-of-responsibility |
| 05:50 | <&Derakon> | Pubsub is handy. |
| 05:51 | <&McMartin> | Swing is pubsub-based - this is a large part of why it's the second-best GUI lib I've ever used. |
| 05:59 | | cpux|2 [cpux@Nightstar-c5874a39.dyn.optonline.net] has joined #code |
| 06:01 | | cpux [cpux@Nightstar-c5874a39.dyn.optonline.net] has quit [Ping timeout: 121 seconds] |
| 06:03 | | Vash [Vash@Nightstar-241cb5d4.wlfrct.sbcglobal.net] has quit [[NS] Quit: I lovecraft Vorn!] |
| 06:20 | | Noah [nbarr@490720.C448F4.12FDC5.2C3558] has quit [Client closed the connection] |
| 06:20 | | Noah [nbarr@490720.C448F4.12FDC5.2C3558] has joined #code |
| 06:31 | | iospace is now known as iospacingout |
| 06:36 | | celticminstrel [celticminst@Nightstar-5d22ab1d.cable.rogers.com] has quit [[NS] Quit: And lo! The computer falls into a deep sleep, to awake again some other day!] |
| 06:44 | | iospacingout is now known as iospacedout |
| 06:46 | | Derakon is now known as Derakon[AFK] |
| 07:14 | <&McMartin> | OK, time to see if I can make Github let me have a website. |
| 08:22 | < Rhamphoryncus> | owwww. http://stackoverflow.com/questions/7464347/interpreting-a-uint16-t-as-a-int16-t and the answers vs http://stackoverflow.com/questions/50605/signed-to-unsigned-conversion-in-c-is-i t-always-safe |
| 08:22 | < Rhamphoryncus> | So, unless C++ abandoned that behaviour, you really should just use a cast |
| 08:23 | < Rhamphoryncus> | Taking the address, casting the pointer, and dereferencing is just *wrong* |
| 08:27 | <&McMartin> | There are better ways to typepun, dmmit |
| 08:27 | <&McMartin> | SUCCESS. |
| 08:27 | <&McMartin> | https://github.com/michaelcmartin/Ophis/downloads |
| 08:28 | <&McMartin> | I still need to make the other downloadables, and make the actual home page have information on it. |
| 08:30 | | * McMartin does a roundtrip test. |
| 08:35 | <&McMartin> | Looks decent |
| 08:37 | <&McMartin> | Tomorrow I can do the distutils and standalone-python-script versions and maybe make a working web page. |
| 08:37 | < Rhamphoryncus> | Just posted an answer of my own. Care to critique? |
| 09:12 | <&McMartin> | I don't see it |
| 09:21 | < Rhamphoryncus> | http://stackoverflow.com/a/11010190/84034 |
| 10:01 | | McMartin [mcmartin@Nightstar-adac7b01.pltn13.sbcglobal.net] has quit [[NS] Quit: 3.4] |
| 10:04 | | Reiver [reiverta@5B433A.F67240.92B5E0.991656] has joined #code |
| 10:04 | | Derakon[AFK] [Derakon@Nightstar-a3b183ae.ca.comcast.net] has quit [Ping timeout: 121 seconds] |
| 10:08 | | McMartin [mcmartin@Nightstar-adac7b01.pltn13.sbcglobal.net] has joined #code |
| 10:08 | | mode/#code [+ao McMartin McMartin] by ChanServ |
| 10:15 | | RichyB [MyCatVerbs@Nightstar-3b2c2db2.bethere.co.uk] has joined #code |
| 10:15 | | You're now known as TheWatcher |
| 10:46 | | Attilla [Obsolete@Nightstar-dac31284.as43234.net] has joined #code |
| 11:09 | | Tsubaki [Phox@Nightstar-81851eab.asahi-net.or.jp] has joined #code |
| 11:15 | | Rhamphoryncus [rhamph@Nightstar-5697f7e2.abhsia.telus.net] has quit [Client exited] |
| 11:21 | | Derakon [Derakon@Nightstar-a3b183ae.ca.comcast.net] has joined #code |
| 11:21 | | mode/#code [+ao Derakon Derakon] by ChanServ |
| 11:25 | | Tsubaki [Phox@Nightstar-81851eab.asahi-net.or.jp] has quit [Client closed the connection] |
| 11:25 | | Noah [nbarr@490720.C448F4.12FDC5.2C3558] has quit [Client closed the connection] |
| 11:57 | | Tsubaki [Phox@Nightstar-0f1c6776.asahi-net.or.jp] has joined #code |
| 12:52 | | iospacedout is now known as io|driving |
| 13:14 | | Reiver [reiverta@5B433A.F67240.92B5E0.991656] has quit [Client closed the connection] |
| 13:18 | | Derakon [Derakon@Nightstar-a3b183ae.ca.comcast.net] has quit [Ping timeout: 121 seconds] |
| 13:21 | | Derakon [Derakon@Nightstar-a3b183ae.ca.comcast.net] has joined #code |
| 13:21 | | mode/#code [+ao Derakon Derakon] by ChanServ |
| 14:32 | | io|driving is now known as iofficespace |
| 15:31 | | celticminstrel [celticminst@Nightstar-5d22ab1d.cable.rogers.com] has joined #code |
| 15:47 | | Attilla [Obsolete@Nightstar-dac31284.as43234.net] has quit [Ping timeout: 121 seconds] |
| 15:51 | | Kindamoody|out is now known as Kindamoody |
| 15:52 | | Attilla [Obsolete@Nightstar-29ee7733.as43234.net] has joined #code |
| 16:35 | | Vash [Vash@Nightstar-241cb5d4.wlfrct.sbcglobal.net] has joined #code |
| 16:35 | | mode/#code [+o Vash] by ChanServ |
| 18:08 | | Noah [nbarr@490720.C448F4.12FDC5.2C3558] has joined #code |
| 18:12 | < celticminstrel> | Apparently definitions are allowed to be extern? |
| 18:33 | < celticminstrel> | Ooh, it seems that non-type template parameter packs can be handled without specialization by turning the parameter pack into an array. |
| 18:59 | < celticminstrel> | Bah, no explicit template arguments for constructors. |
| 19:02 | | Kindamoody is now known as Kindamoody[zZz] |
| 19:07 | <@rms> | Why would you want that? |
| 19:14 | < celticminstrel> | Variadic constructor. Though... I just remembered initializer lists... that might be a little simpler... |
| 19:19 | < celticminstrel> | Except it complains it doesn't exist even though the header is included... o.O |
| 19:20 | < celticminstrel> | Oh wait, I see why. It's guarded. Probably not supported yet. |
| 19:26 | | * iofficespace almost is finished with her latest batch of testing... yayyy~ |
| 19:59 | | Attilla [Obsolete@Nightstar-29ee7733.as43234.net] has quit [Client closed the connection] |
| 19:59 | | Attilla [Obsolete@Nightstar-def845e9.as43234.net] has joined #code |
| 20:14 | < iofficespace> | finally, done with testing the interrupt routing on this board... 22 different sets of tests... |
| 20:36 | | RichyB [MyCatVerbs@Nightstar-3b2c2db2.bethere.co.uk] has quit [[NS] Quit: Leaving] |
| 21:52 | | Rhamphoryncus [rhamph@Nightstar-5697f7e2.abhsia.telus.net] has joined #code |
| 22:00 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 22:16 | | * ToxicFrog find the problem! |
| 22:16 | | * ToxicFrog gets to implement TCP! |
| 22:19 | <~Vornicus> | |
| 22:22 | <@ToxicFrog> | Well ok not quite |
| 22:22 | <@ToxicFrog> | The problem is that ix_send_msg blocks until the message is in the C-ring, not until it has been received |
| 22:23 | <@ToxicFrog> | This means that it's possible for a sender to send multiple messages before the receiver receives any of them |
| 22:23 | <@ToxicFrog> | Which in turn means the order they are received in depends on when the receiver calls ix_recv_msg() relative to the location of the messages in the C-ring |
| 22:23 | <@ToxicFrog> | Which means the receiver needs some way of reassembling them in order |
| 22:33 | <~Vornicus> | ix? |
| 22:33 | <@ToxicFrog> | I have no idea. It's the prefix the system functions on this chip use. |
| 22:34 | <@ToxicFrog> | The actual chip is called - depending on which part you're looking at - IMAP5, IMAPCAR2, or XC-CORE, so I have no idea where IX comes from. |
| 22:34 | | Derakon_ [chriswei@Nightstar-a3b183ae.ca.comcast.net] has joined #code |
| 22:35 | | * Derakon_ discovers that the currently-available Cocoa build of wxPython doesn't do mouse events properly. |
| 22:35 | < Derakon_> | Nominally it was fixed ~3 months ago, but the build was made last year. |
| 22:35 | < Derakon_> | So I get to download the 47MB source tarball at ~25KB/s and try to build it myself. Hooray. :( |
| 22:52 | <&jerith> | 23:42 < keturn> http://bpaste.net/show/c6FCNF9PnO5EoxsMEti6/ |
| 22:52 | <&jerith> | 23:43 <@dash> keturn: you're a weird dude |
| 22:52 | <&jerith> | 23:49 <@dash> http://www.bpaste.net/show/jW37aEjI71lY3LBOFP16/ |
| 22:52 | <&jerith> | 23:49 < mwhudson> http://paste.ubuntu.com/1039794/ |
| 22:52 | <&jerith> | 23:49 <@dash> ^-- lvalues! |
| 22:52 | <&jerith> | 23:49 <@dash> mwhudson: wow even worse |
| 22:52 | <&jerith> | 23:49 <@dash> <3 |
| 22:54 | <@ToxicFrog> | what |
| 22:55 | <&jerith> | Yay Python. \o/ |
| 22:55 | <&jerith> | Also /o\ |
| 22:56 | < Rhamphoryncus> | heh, I can trace through exactly what the generator is doing |
| 22:56 | <&jerith> | Yup. |
| 22:56 | <&jerith> | I'm particularly fond of the sys.modules one. |
| 22:56 | <@Tamber> | I haven't the slightest clue what's going on :D |
| 22:56 | < Rhamphoryncus> | I don't get the point of it |
| 22:56 | <&jerith> | Tamber: It's an abuse of exception handlers and generator syntax. |
| 22:57 | <@Tamber> | Ah. |
| 22:57 | <&jerith> | Rhamphoryncus: The point is to make your head asplode. |
| 22:58 | < Rhamphoryncus> | Tamber: in particular "except ...:" takes an expression, which is evaluated right then to find the type of exception to match, and of course (yield) is a valid expression.. |
| 22:58 | < Rhamphoryncus> | jerith: I guess I'm too used to thinking of sys.modules that way |
| 22:58 | <@Tamber> | My brain is refusing to parse that sentence, which I'm going to take as a very strong hint that I should go sleep. <_> |
| 22:59 | | * Derakon_ finally finishes downloading the wxPython source, runs ./configure, it errors out. Argh. |
| 22:59 | <&jerith> | 23:50 < fzzzy> mwhudson: that one is crazy |
| 22:59 | <&jerith> | 23:50 < mwhudson> fzzzy: not really, it's just combining modestly crazy things in unexpected ways :) |
| 22:59 | < Rhamphoryncus> | put another way: python is a dynamic language, so "except EXPR:" takes an expression, not a type |
| 23:01 | < Rhamphoryncus> | Same reason python doesn't have switch/case. You'd want it to take expressions, but you'd need to evaluate *every* option each time |
| 23:03 | <@rms> | Bah, switches are just giant goto messes anyways |
| 23:03 | < Derakon_> | They're also sometimes handy syntactic sugar. |
| 23:03 | < Rhamphoryncus> | exceptions are just giant goto messes >.> |
| 23:23 | | Tsubaki_ [Phox@Nightstar-90b7ea4e.asahi-net.or.jp] has joined #code |
| 23:23 | | Derakon_ [chriswei@Nightstar-a3b183ae.ca.comcast.net] has quit [[NS] Quit: leaving] |
| 23:24 | | Tsubaki [Phox@Nightstar-0f1c6776.asahi-net.or.jp] has quit [Ping timeout: 121 seconds] |
| 23:41 | | himi [fow035@D741F1.243F35.CADC30.81D435] has joined #code |
| 23:41 | | mode/#code [+o himi] by ChanServ |
| --- Log closed Thu Jun 14 00:00:01 2012 |