| --- Log opened Tue May 22 00:00:53 2012 |
| 00:03 | <&McMartin> | That still wouldn't work, because you'd have to pick the overload based on the contents of a string that isn't even guaranteed to be constant |
| 00:03 | <&McMartin> | But you can still do printf-like things - I give you QString::arg() |
| 00:08 | < Rhamphoryncus> | It looks like you can template the overload |
| 00:10 | < Rhamphoryncus> | qstring::arg is ugly |
| 00:11 | <&McMartin> | It's clean in practice, though |
| 00:11 | <&McMartin> | QString("formatstring %0 %2 $1").arg("zeroth").arg(1).arg(2.5); |
| 00:12 | < Rhamphoryncus> | err.. it has both % and $? |
| 00:12 | | cpux [cpux@Nightstar-c5874a39.dyn.optonline.net] has quit [Client closed the connection] |
| 00:12 | <&McMartin> | I may have confused % and $. |
| 00:13 | < Rhamphoryncus> | ah okay |
| 00:17 | <&McMartin> | But it's dynamically-typed in the arg function, there's just one arg function, and it does positional replacement so that you don't have the standard printf-internationalization trap. |
| 00:18 | < Rhamphoryncus> | That's why I want one like python3's .format() |
| 00:18 | < Rhamphoryncus> | Which is actually extensible |
| 00:22 | | maoranma [nbarr@Nightstar-9e797a1c.pools.spcsdns.net] has joined #code |
| 00:24 | | cpux [cpux@Nightstar-c5874a39.dyn.optonline.net] has joined #code |
| 00:25 | | ShellNinja [abudhabi@Nightstar-300ad1ff.adsl.inetia.pl] has joined #code |
| 00:25 | | Noah [nbarr@Nightstar-9e797a1c.pools.spcsdns.net] has quit [Ping timeout: 121 seconds] |
| 00:32 | | Attilla [Obsolete@Nightstar-3f363270.as43234.net] has quit [Ping timeout: 121 seconds] |
| 00:32 | | cpux [cpux@Nightstar-c5874a39.dyn.optonline.net] has quit [[NS] Quit: Well, most things get better when I kick them!] |
| 00:35 | | iospacedout is now known as iospace |
| 00:36 | | cpux [cpux@Nightstar-c5874a39.dyn.optonline.net] has joined #code |
| 00:54 | | Derakon[AFK] is now known as Derakon |
| 00:56 | | maoranma is now known as Noah |
| 00:59 | | eckse_ [eckse@Nightstar-0a8171a4.dsl.sentex.ca] has joined #code |
| 01:02 | | eckse [eckse@Nightstar-44362701.dsl.sentex.ca] has quit [Ping timeout: 121 seconds] |
| 02:01 | <&McMartin> | Oh god this code is terrible |
| 02:01 | | * McMartin is filled with shame |
| 02:01 | | * McMartin excises it with fire and sword |
| 02:13 | | Kindamoody[zZz] is now known as Kindamoody |
| 03:02 | | Kindamoody is now known as Kindamoody|coding |
| 03:11 | | Rhamphoryncus [rhamph@Nightstar-5697f7e2.abhsia.telus.net] has quit [Client exited] |
| 04:27 | | eckse [eckse@Nightstar-5f2dc230.dsl.sentex.ca] has joined #code |
| 04:27 | | mode/#code [+o eckse] by ChanServ |
| 04:30 | | eckse_ [eckse@Nightstar-0a8171a4.dsl.sentex.ca] has quit [Ping timeout: 121 seconds] |
| 04:30 | | Kindamoody|coding is now known as Kindamoody |
| 04:32 | < celticminstrel> | Ooh, I can declare a variable as "auto&" and it works just as I expect. |
| 04:50 | < celticminstrel> | Okay, boost::multi_array doesn't seem to like being assigned... |
| 05:01 | | Kindamoody is now known as Kindamoody|out |
| 05:04 | | * celticminstrel decides to try destructor+placement new. |
| 05:07 | < celticminstrel> | ...now I need to deal with making a deep copy of a vector of polymorphic pointers. :| |
| 05:10 | < celticminstrel> | ...oh hey, I don't need to save the multi_array after all because it never ever changes once initialized. |
| 05:22 | < celticminstrel> | Okay, so I just need to define copy constructors and have clone() defer to them, it seems. |
| 05:31 | <&McMartin> | God, auto |
| 05:31 | <&McMartin> | C++ has needed that since 1998. |
| 05:32 | < celticminstrel> | Yeah, that and enhanced-for are the main new things I'm using. |
| 05:33 | < celticminstrel> | Actually probably more enhanced-for than auto, but that's just because there's not much point updating old code to use auto. |
| 05:34 | <~Vornicus> | what's auto do? |
| 05:37 | < celticminstrel> | Wait a second, I don't need to define a copy constructor if all it does is a straight-forward "defer to super and then copy over all extra fields", because that's the default behaviour, right? |
| 05:38 | <&McMartin> | Vornicus: a variable declared auto is declared based on type inference on its initializing expression. |
| 05:38 | <&McMartin> | Not so meaty for double |
| 05:39 | <&McMartin> | Pretty nice for, say, std::map<std::string, std::vector<string> >::const_iterator& |
| 05:39 | < celticminstrel> | Yes indeed. |
| 05:39 | < celticminstrel> | I don't need to define these copy constructors, right? |
| 05:40 | <&McMartin> | From what you describe, nah |
| 05:40 | <&McMartin> | But you said you wanted to deep-copy *pointers* |
| 05:40 | <&McMartin> | pointers will be shallowcopied |
| 05:40 | <&McMartin> | Or rather the pointer will be deepcopied. All 64 bits of it. Not the thing pointed at. |
| 05:41 | < celticminstrel> | I have a vector of pointers. |
| 05:41 | < celticminstrel> | The things pointed to don't have any further pointers (I think). |
| 05:42 | <&McMartin> | What's being copied here, though, the vector? |
| 05:42 | <&McMartin> | Because if so the only copy constructor that's gonna get called is the vector's. |
| 05:44 | < celticminstrel> | Yeah, the vector is what's being copied. |
| 05:45 | < celticminstrel> | So instead of just naively assigning I need to iterate through and clone every element. |
| 06:29 | | iospace is now known as iospacedout |
| 06:30 | | Derakon is now known as Derakon[AFK] |
| 06:32 | | 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!] |
| 06:59 | | eckse [eckse@Nightstar-5f2dc230.dsl.sentex.ca] has quit [Ping timeout: 121 seconds] |
| 07:10 | | Vash [Vash@Nightstar-241cb5d4.wlfrct.sbcglobal.net] has quit [[NS] Quit: I lovecraft Vorn!] |
| 07:12 | | Attilla [Obsolete@Nightstar-3f363270.as43234.net] has joined #code |
| 07:36 | | himi [fow035@D741F1.243F35.CADC30.81D435] has quit [Connection closed] |
| 07:50 | | gnolam [lenin@Nightstar-202a5047.priv.bahnhof.se] has quit [[NS] Quit: Thunderbolts and lightning, very very frightening] |
| 08:26 | | Reiver [reiverta@5B433A.F67240.A412A8.CE8081] has joined #code |
| 10:24 | | RichyB [MyCatVerbs@Nightstar-3b2c2db2.bethere.co.uk] has joined #code |
| 11:08 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has joined #code |
| 11:08 | | mode/#code [+o himi] by ChanServ |
| 11:56 | | celticminstrel [celticminst@Nightstar-5d22ab1d.cable.rogers.com] has joined #code |
| 12:01 | | Rhamphoryncus [rhamph@Nightstar-5697f7e2.abhsia.telus.net] has joined #code |
| 12:30 | | gnolam [lenin@Nightstar-202a5047.priv.bahnhof.se] has joined #code |
| 12:40 | | McMartin [mcmartin@Nightstar-bc7067be.pltn13.sbcglobal.net] has quit [Operation timed out] |
| 12:52 | | iospacedout is now known as IOfficespace |
| 12:52 | | celticminstrel is now known as celmin|away |
| 12:52 | | IOfficespace is now known as iOfficespace |
| 12:59 | | McMartin [mcmartin@Nightstar-e4a532c9.pltn13.sbcglobal.net] has joined #code |
| 12:59 | | mode/#code [+ao McMartin McMartin] by ChanServ |
| 14:17 | | Kindamoody|out is now known as Kindamoody |
| 14:45 | | Reiver [reiverta@5B433A.F67240.A412A8.CE8081] has quit [Client closed the connection] |
| 15:22 | | Kindamoody is now known as Kindamoody|afk |
| 15:51 | | celmin|away is now known as celticminstrel |
| 16:04 | | Noah [nbarr@Nightstar-9e797a1c.pools.spcsdns.net] has quit [Client closed the connection] |
| 16:04 | | Noah [nbarr@Nightstar-9e797a1c.pools.spcsdns.net] has joined #code |
| 16:53 | | gruber [lenin@Nightstar-202a5047.priv.bahnhof.se] has joined #code |
| 16:53 | | gnolam is now known as NSGuest52 |
| 16:53 | | gruber is now known as gnolam |
| 16:53 | | NSGuest52 [lenin@Nightstar-202a5047.priv.bahnhof.se] has quit [Ping timeout: 121 seconds] |
| 16:57 | | eckse [eckse@Nightstar-0523d800.dsl.sentex.ca] has joined #code |
| 16:57 | | mode/#code [+o eckse] by ChanServ |
| 17:16 | | EvilDarkLord is now known as Maze |
| 17:21 | | eckse [eckse@Nightstar-0523d800.dsl.sentex.ca] has quit [Ping timeout: 121 seconds] |
| 17:40 | | eckse [eckse@Nightstar-c383bbe3.dsl.sentex.ca] has joined #code |
| 17:40 | | mode/#code [+o eckse] by ChanServ |
| 18:03 | | Vash [Vash@Nightstar-241cb5d4.wlfrct.sbcglobal.net] has joined #code |
| 18:03 | | mode/#code [+o Vash] by ChanServ |
| 18:16 | < gnolam> | https://www.youtube.com/watch?v=zmO06_MpnMk |
| 18:24 | <~Vornicus> | that is crazy awesome |
| 19:00 | | eckse_ [eckse@Nightstar-6604bb2a.dsl.sentex.ca] has joined #code |
| 19:01 | < gnolam> | That it is. |
| 19:02 | < gnolam> | I've seen many C64 instrument hacks, but that one's the coolest I've seen yet. |
| 19:03 | | eckse [eckse@Nightstar-c383bbe3.dsl.sentex.ca] has quit [Ping timeout: 121 seconds] |
| 19:04 | | ToxicFrog [ToxicFrog@ServerAdministrator.Nightstar.Net] has quit [Operation timed out] |
| 19:12 | | eckse [eckse@Nightstar-837b03ce.dsl.sentex.ca] has joined #code |
| 19:12 | | mode/#code [+o eckse] by ChanServ |
| 19:12 | | Kindamoody|afk is now known as Kindamoody |
| 19:14 | | eckse_ [eckse@Nightstar-6604bb2a.dsl.sentex.ca] has quit [Ping timeout: 121 seconds] |
| 19:20 | | ToxicFrog [ToxicFrog@ServerAdministrator.Nightstar.Net] has joined #code |
| 19:20 | | mode/#code [+o ToxicFrog] by ChanServ |
| 19:21 | | Kindamoody is now known as Kindamoody[zZz] |
| 19:28 | | Vash [Vash@Nightstar-241cb5d4.wlfrct.sbcglobal.net] has quit [[NS] Quit: I lovecraft Vorn!] |
| 19:35 | < celticminstrel> | I should either make a better level format, or make a level editor... |
| 19:36 | <@Tamber> | Or both? |
| 19:36 | < celticminstrel> | True. |
| 19:37 | < celticminstrel> | The current one was supposed to be human-readable, but then stuff got out of hand. |
| 19:58 | < sshine_> | I'm trying to prove a lower bound on the number of elements in a set. |
| 20:00 | < sshine_> | I've got a dynamic programming algorithm that has the subproblems characterised by {(i,j) in n^2 | i < j}. it is easy to reason that its size is O(n^2) because it is a subset of {(i,j) in n^2}. |
| 20:01 | < sshine_> | but how I say that it has Omega(n^2), I don't know. |
| 20:02 | < RichyB> | If it's a dynamic programming algorithm, doesn't it *always* hit *all* of its subproblems? |
| 20:02 | < sshine_> | RichyB, yes, in the worst case. |
| 20:02 | < RichyB> | and that's (mostly) the definition of DP as opposed to divide and conquer with memoization. |
| 20:02 | < sshine_> | I didn't know there was a difference between DP and divide-and-conquer with memoization. |
| 20:03 | < sshine_> | oh |
| 20:03 | < sshine_> | I suppose you may say that DP always calculates the whole table... |
| 20:03 | < RichyB> | DP works upwards from the base cases. |
| 20:03 | < sshine_> | hm |
| 20:03 | < RichyB> | D-C works downwards from the complete problem. |
| 20:03 | < sshine_> | the TAs do refer to top-down DP algorithms, but I don't care much when it's one or the other... it's all DP to me. =) |
| 20:03 | < sshine_> | ah |
| 20:03 | < sshine_> | good point |
| 20:04 | | Maze is now known as EvilDarkLord |
| 20:04 | < sshine_> | so what I want to do is determine how many subproblems there are |
| 20:04 | < sshine_> | I want to come to Theta(n^2) but so far I only have O(n^2) and Omega(n). |
| 20:05 | < sshine_> | basically, not all subproblems of the form (i,j) in NxN are subproblems that my algorithm will address because it only recurses while i < j. |
| 20:05 | < RichyB> | oh |
| 20:06 | < RichyB> | { (i,j) for i <- [1..n], j <- 1..n, i < j } |
| 20:06 | < sshine_> | I could say that (i,j) where i >= j are symmetric to subproblems that I am looking at, but that doesn't say that my algorithm doesn't have a lower bound than n^2. |
| 20:06 | < RichyB> | draw that, it's a triangle |
| 20:06 | < sshine_> | yeah, it is a triangle. do I say "proof by triangle"? :D |
| 20:06 | < RichyB> | It has n*n/2 elements. |
| 20:06 | < sshine_> | ah |
| 20:06 | < sshine_> | yeah, ok |
| 20:06 | < sshine_> | I suppose I could consider that informal induction. |
| 20:06 | < RichyB> | Well, you can break it into levels. |
| 20:07 | < RichyB> | It's sum(sum([1..j]) for j in [1..n]) |
| 20:07 | < RichyB> | er |
| 20:07 | < RichyB> | worry |
| 20:07 | < RichyB> | sorry |
| 20:07 | < RichyB> | It's sum(length([1..j]) for j in [1..n]) |
| 20:08 | < RichyB> | which is of course sum(j for j in [1..n]) |
| 20:08 | < RichyB> | which is (j) + (j-1) + (j-2) + (j-3)... + 1. |
| 20:08 | < sshine_> | it's a lower triangle, too, isn't it? |
| 20:08 | < sshine_> | yeah. |
| 20:08 | < sshine_> | thanks! |
| 20:09 | < RichyB> | Yeah. There are j terms in that sum and their mean value is j/2. :) |
| 20:09 | < RichyB> | No worries. |
| 20:13 | <@ToxicFrog> | RichyB: see, I think of DP as being top-down as well. Yeah, the base cases are the results that actually get calculated first, but the same is true of D+C |
| 20:13 | <@ToxicFrog> | When you get down to it, DP is just memoized overlapping D+C. |
| 20:13 | < sshine_> | divide-and-conquer kind of has to be top-down, otherwise it would be conquer-and-assembly. ;-P |
| 20:14 | < RichyB> | FWIW, the wikipedia article basically disagrees with my distinction that DP is bottom-up. |
| 20:15 | < RichyB> | I happen to think of the distinction as being that DP is bottom-up filling out of arrays with subproblem answers because that was how DP was demonstrated to me in lectures. |
| 20:16 | < RichyB> | D&C-memoised isn't really different enough in runtime or structure to be worth caring much about the distinction. |
| 20:16 | < RichyB> | and the general consensus (as measured by wikipedia) is that D&C-memoised *is* DP. |
| 20:38 | <&McMartin> | Yeah |
| 20:38 | <&McMartin> | My experience and discoveries matched yours |
| 20:38 | <&McMartin> | D&C-memoized is *better* >_> |
| 20:39 | < RichyB> | It's usually shockingly easier to write. I don't know about better. |
| 20:40 | < RichyB> | Filling out dense arrays in a sequential order tends to have smaller constants than filling out entries of a hash table keyed on a complicated tuple. |
| 20:43 | <&McMartin> | Absent evidence that it's constant-factor and not programmer time and maintenance cost you need to optimize... |
| 20:45 | < RichyB> | Point taken. |
| 20:45 | < RichyB> | Swim, swim, hungry! |
| 20:45 | | RichyB [MyCatVerbs@Nightstar-3b2c2db2.bethere.co.uk] has quit [[NS] Quit: Leaving] |
| 20:47 | <~Vornicus> | oh, derp |
| 20:47 | <~Vornicus> | I was wondering what happened to MyCatVerbs and there indeed he is |
| 20:48 | | Derakon [Derakon@Nightstar-a3b183ae.ca.comcast.net] has joined #code |
| 20:48 | | mode/#code [+ao Derakon Derakon] by ChanServ |
| 20:51 | | Derakon[AFK] [Derakon@Nightstar-a3b183ae.ca.comcast.net] has quit [Operation timed out] |
| 20:57 | | Vash [Vash@Nightstar-241cb5d4.wlfrct.sbcglobal.net] has joined #code |
| 20:57 | | mode/#code [+o Vash] by ChanServ |
| 21:05 | | Noah [nbarr@Nightstar-9e797a1c.pools.spcsdns.net] has quit [Connection reset by peer] |
| 21:05 | | maoranma [nbarr@Nightstar-9e797a1c.pools.spcsdns.net] has joined #code |
| 21:06 | | maoranma is now known as Noah |
| 21:19 | | Pandemic [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has joined #code |
| 21:21 | | Pandemic [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has quit [[NS] Quit: shutting down!] |
| 21:37 | | Derakon [Derakon@Nightstar-a3b183ae.ca.comcast.net] has quit [Ping timeout: 121 seconds] |
| 21:38 | | Derakon [Derakon@Nightstar-a3b183ae.ca.comcast.net] has joined #code |
| 21:38 | | mode/#code [+ao Derakon Derakon] by ChanServ |
| 21:59 | | Derakon [Derakon@Nightstar-a3b183ae.ca.comcast.net] has quit [Ping timeout: 121 seconds] |
| 21:59 | | Derakon [Derakon@Nightstar-a3b183ae.ca.comcast.net] has joined #code |
| 21:59 | | mode/#code [+ao Derakon Derakon] by ChanServ |
| 22:36 | < celticminstrel> | Is it just me or does SDL_ttf not understand kerning? |
| 22:37 | < Rhamphoryncus> | That would suck but I haven't touched it yet |
| 22:37 | < Rhamphoryncus> | At a minimum I need to learn opengl textures first, heh |
| 22:39 | < celticminstrel> | Well, it has to be either SDL_ttf or Arial Unicode, right? |
| 22:39 | < celticminstrel> | And I imagine the latter is unlikely. |
| 22:42 | < Rhamphoryncus> | yeah |
| 22:42 | < Rhamphoryncus> | Or you're misusing it somehow |
| 22:42 | < gnolam> | ISTR SDL_ttf being rather primitive. |
| 22:43 | | * gnolam gave up on it and rolled his own FreeType layer. |
| 22:43 | < gnolam> | And on the topic of FreeType I have only one thing to say: disguised pointers. >:( |
| 22:44 | | Attilla [Obsolete@Nightstar-3f363270.as43234.net] has quit [[NS] Quit: ] |
| 22:45 | < gnolam> | Yes. They really do that nasty shit of misusing typedef to try to hide the pointer status of types. |
| 22:48 | <&McMartin> | Well |
| 22:48 | <&McMartin> | Like int_ptr_t, or typedef void *foo;? |
| 22:55 | < gnolam> | typedef foo* bar; |
| 22:55 | < gnolam> | With no clue that bar is actually a pointer. |
| 22:56 | <&McMartin> | ... it's right there in the *? |
| 22:56 | <&McMartin> | I mean, yes, this is one of the few places where I'm completely OK with Hungarian |
| 22:57 | < gnolam> | Right. Which you only discover when you go snooping around in the headers. |
| 22:57 | | * McMartin nods |
| 22:57 | < gnolam> | Which you should not have to do. |
| 22:58 | <&McMartin> | It's OK if everything's a pointer, I guess. I'd have to check in UQM to see how we did that. |
| 22:58 | <&McMartin> | But that used total encapsulation |
| 22:58 | <&McMartin> | I don't recall if we just did "struct foo;" or "struct foo_impl; typedef foo_impl *FOO;" |
| 23:09 | | Attilla [Obsolete@Nightstar-3f363270.as43234.net] has joined #code |
| --- Log closed Wed May 23 00:00:08 2012 |