| --- Log opened Thu Aug 12 00:00:32 2010 |
| 00:11 | | Alek [omegaboot@Nightstar-c5f3565b.il.comcast.net] has quit [[NS] Quit: ] |
| 00:17 | | Alek [omegaboot@Nightstar-c5f3565b.il.comcast.net] has joined #code |
| 00:33 | | Attilla [Obsolete@Nightstar-0a7d7712.threembb.co.uk] has quit [Ping timeout: 121 seconds] |
| 00:34 | | Attilla [Obsolete@Nightstar-30ecd985.threembb.co.uk] has joined #code |
| 00:34 | | mode/#code [+o Attilla] by Reiver |
| 00:46 | | * McMartin rages at Fedora, which seems to no longer include the C API man pages with gcc. |
| 00:52 | <@McMartin> | Oy. |
| 00:52 | <@McMartin> | It just doesn't install package "man-pages" by default. |
| 00:52 | <@McMartin> | CATERING TO CASUALS |
| 00:52 | < Namegduf> | How big is that package? |
| 00:53 | < Namegduf> | I mean, missing dev stuff by default, sure. |
| 00:53 | < Namegduf> | Missing documentation by default, just seems sutpid. |
| 00:54 | <@McMartin> | It's like 4M |
| 00:54 | <@McMartin> | Basic docs are there, but the C API was not |
| 00:55 | < Namegduf> | Hmm. |
| 00:55 | < Namegduf> | Does Fedora have "suggests"? |
| 00:55 | < Namegduf> | Because that'd be a sane value for GCC to those docs. |
| 01:01 | <@McMartin> | It doesn't seem to. It does, however, have "whatprovides". |
| 01:07 | < celticminstrel> | Yes. Very sutpid. :P Also ver stupid. |
| 01:07 | < celticminstrel> | ^very |
| 01:07 | < celticminstrel> | And now you may point out the irony, if you wish. |
| 01:15 | | shade_of_cpux is now known as cpux |
| 01:20 | | Attilla [Obsolete@Nightstar-30ecd985.threembb.co.uk] has quit [[NS] Quit: ] |
| 01:40 | | Derakon[AFK] is now known as Derakon |
| 02:01 | < ToxicFrog> | McMartin: opensuse, ubuntu and mint have also had this issue forever |
| 02:02 | <@McMartin> | Fedora didn't used to, I don't think, though it may just be that the old DVD system shipped with it by default and the new CD one does not |
| 02:07 | < simon__> | you have anything but a bootstrapping package repo on DVD? |
| 02:08 | < Namegduf> | Looks like in Debian, gcc suggests gcc-doc, which has some stuff, but devices, file formats, and some other stuff is still in manpages |
| 03:17 | | ToxicFrog [ToxicFrog@ServerAdministrator.Nightstar.Net] has quit [Operation timed out] |
| 03:19 | | SmithKurosaki [Smith@Nightstar-b696a095.dsl.teksavvy.com] has quit [Ping timeout: 121 seconds] |
| 03:21 | | SmithKurosaki [Smith@Nightstar-cc72cefe.dsl.teksavvy.com] has joined #code |
| 03:21 | | ToxicFrog [ToxicFrog@ServerAdministrator.Nightstar.Net] has joined #code |
| 03:21 | | mode/#code [+o ToxicFrog] by Reiver |
| 03:22 | <@McMartin> | I made a thing http://hkn.eecs.berkeley.edu/~mcmartin/tlk_parse.c |
| 03:22 | <@McMartin> | This disassembles the NPC dialog files in Ultima IV DOS. |
| 03:22 | < celticminstrel> | Ooh, Berkeley. |
| 03:23 | <@McMartin> | Long time passing, but I've still got the alumnus account. |
| 03:24 | < celticminstrel> | I think UCB Logo was my first ever programming language. |
| 03:24 | <@McMartin> | Not unlikely! |
| 03:38 | | celticminstrel [celticminstre@Nightstar-f8b608eb.cable.rogers.com] has quit [[NS] Quit: And lo! The computer falls into a deep sleep, to awake again some other day!] |
| 03:39 | | celticminstrel [celticminstre@Nightstar-f8b608eb.cable.rogers.com] has joined #code |
| 03:43 | | celticminstrel [celticminstre@Nightstar-f8b608eb.cable.rogers.com] has quit [[NS] Quit: And lo! The computer falls into a deep sleep, to awake again some other day!] |
| 04:38 | | Kazriko [kaz@Nightstar-e09690fa.client.bresnan.net] has quit [Ping timeout: 121 seconds] |
| 04:57 | | Stalker [Z@3A600C.A966FF.5BF32D.8E7ABA] has quit [Ping timeout: 121 seconds] |
| 05:05 | | Thaqui [Thaqui@27B34E.D54D49.F53FA1.6A113C] has joined #code |
| 05:17 | <@Derakon> | I don't suppose anyone knows off the top of their head a good algorithm for generating a random planar graph given a set of starting nodes/edges? |
| 05:17 | <@Derakon> | With the starting nodes/edges not being connected initially, but the final graph being a single connected component. |
| 05:17 | <@Derakon> | I mean, I can hack something together, but if there's prior art I could examine it'd make things go more smoothly. |
| 05:18 | <@Vornicus> | Do you want this graph to be maximal? |
| 05:18 | <@Derakon> | Define maximal? |
| 05:19 | <@Vornicus> | You cannot add any more edges to it without crossing another edge. |
| 05:19 | <@Derakon> | Oh. No. |
| 05:19 | <@Derakon> | This is for generating the Jetblade map -- I need to switch from a directed tree structure to an undirected graph structure. |
| 05:19 | <@Vornicus> | Okay, that's a little harder. |
| 05:19 | <@Derakon> | I don't want any given node to have more than 4 edges. |
| 05:20 | <@Vornicus> | I... guess you could play Brussels Sprouts. |
| 05:20 | <@Derakon> | (Reason for the switch is that the basis for adding roadblocks to the map requires the roadblocks to be placed before the rest of the map is) |
| 05:20 | | * Derakon quirks an eyebrow. |
| 05:20 | | ToxicFrog [ToxicFrog@ServerAdministrator.Nightstar.Net] has quit [Client closed the connection] |
| 05:21 | | SmithKurosaki [Smith@Nightstar-cc72cefe.dsl.teksavvy.com] has quit [Client closed the connection] |
| 05:22 | <@Vornicus> | Well. Okay, let me start by defining "Sprouts" - in Sprouts, you draw a graph. Start with a couple of nodes, connect two of them (or just one), put a new node on that line. No crossings allowed. A node of degree three is dead, you can't add any more edges to it. |
| 05:22 | | ToxicFrog [ToxicFrog@ServerAdministrator.Nightstar.Net] has joined #code |
| 05:22 | | mode/#code [+o ToxicFrog] by Reiver |
| 05:22 | <@Derakon> | This sounds like a fractal generator... |
| 05:23 | <@Vornicus> | It can come up with vaguely fractal things, but actually it's a game. |
| 05:23 | <@Derakon> | If I understand it correctly, you start with a bunch of randomly-placed, unconnected nodes, and you basically choose to connect nodes until all nodes are connected and/or all nodes have degree 3? |
| 05:23 | <@Vornicus> | Yeah. As a game you start with just two nodes. |
| 05:23 | <@Derakon> | Oh, I think I see. |
| 05:23 | <@Vornicus> | And then add a node to the middle of each edge. |
| 05:24 | <@Derakon> | Each player tries to prevent the other from making any more moves by boxing them in with their own graph, or something like that. |
| 05:24 | <@Vornicus> | Yes. |
| 05:25 | <@Vornicus> | Well, Brussels Sprouts is the same sort of thing, except each node is a little plus sign. |
| 05:25 | <@Vornicus> | And the arms of the sign are where you can add edges; each arm gets only one edge. |
| 05:25 | <@Derakon> | I think I get the idea. |
| 05:26 | <@Derakon> | Hmm...I think my main sources of complexity for this are as follows: |
| 05:26 | <@Derakon> | 1) I don't want any two edges to come too close to each other, instead of just wanting to prevent them from crossing. |
| 05:26 | <@Derakon> | 2) I don't want two edges connected to the same node to have too acute of an angle between them. |
| 05:27 | <@Derakon> | 3) I want to base the local shape of the graph on the local map sector (e.g. some sectors have axis-aligned edges, some have long edges, etc.) |
| 05:27 | <@Vornicus> | You can jiggle out a sprouts graph to make it more friendly. |
| 05:27 | <@Derakon> | Hm, generate the graph as an axis-aligned thing with rigid edge lengths, then jitter the edges until they look more organic? |
| 05:28 | <@Vornicus> | That works too. |
| 05:28 | <@Derakon> | That could work. |
| 05:29 | <@Vornicus> | Only problem is that jiggling stuff out like this can be quite computationally expensive. |
| 05:30 | <@Derakon> | The current tree approach generates a set of semi-random angles, then picks random edge lengths and tries to place nodes. If it fails, it tries again until it hits its max retry level for that depth of the tree. |
| 05:30 | <@Derakon> | (So I can enforce really trying hard to grow the tree near the center, without caring so much near the leaves) |
| 05:30 | <@Vornicus> | Hm hm hm |
| 05:30 | <@Vornicus> | Okay okay better plan! |
| 05:30 | <@Derakon> | Do tell. |
| 05:30 | <@Vornicus> | Build your nodes, then /get the Delaunay Triangulation/ of that. |
| 05:31 | <@Vornicus> | (Dual graph of the voronoi diagram) |
| 05:31 | <@Derakon> | I'm reading the Wikipedia page as we speak... |
| 05:31 | <@Vornicus> | THe delaunay triangulation has the interesting property that it maximizes the angles of the triangle. |
| 05:32 | <@Vornicus> | You can then use the delaunay triangulation, apply a Spanning Tree generator to it, and then pick a random number of other edges. |
| 05:32 | <@Derakon> | Ah, so that avoids acute angles to the extent possible. |
| 05:33 | <@Derakon> | Sounds good. |
| 05:33 | <@Derakon> | Though I think what I'd actually do is use the triangulation as the graph and remove unneeded edges. |
| 05:33 | <@Derakon> | Functionally similar, mind. |
| 05:34 | <@Derakon> | Thanks. |
| 05:34 | <@Vornicus> | And because your things will be far apart, it will not be difficult to rejigger a path so it's made entirely of ao directions -- and, at that, your edges are all straight. |
| 05:34 | <@Derakon> | ao? |
| 05:34 | <@Vornicus> | axis-oriented |
| 05:35 | <@Vornicus> | The only difficulty then becomes making sure your edges are all reasonably sized. |
| 05:35 | <@Derakon> | Oh, hm. Actually. |
| 05:36 | <@Derakon> | Won't triangulation be a problem for the axis-aligned section? |
| 05:36 | <@Derakon> | Since I won't want diagonal lines. |
| 05:36 | <@Derakon> | s/lines/edges/ |
| 05:36 | <@Vornicus> | That's what I mean by rejiggering a path - you don't necessarily want to go straight along it. |
| 05:36 | <@Derakon> | Ah. |
| 05:37 | <@Derakon> | So given a triangle in the axis-aligned segment, pop the hypotenuse out to make a rectangle, or ditch the edge entirely, or do a Reimann sum type of deal. |
| 05:38 | <@Vornicus> | Up then left then up for instance would work. |
| 05:39 | <@Derakon> | Depends on how the original triangle is oriented~ |
| 05:39 | <@Derakon> | But yes, that's what I meant by a Reimann sum. |
| 05:39 | <@Derakon> | Approximate a triangle on the grid. |
| 05:51 | | Thaqui [Thaqui@27B34E.D54D49.F53FA1.6A113C] has quit [Connection closed] |
| 05:54 | <@Derakon> | So, something like this? http://paste.ubuntu.com/476765/ |
| 05:58 | <@Vornicus> | Looks it, yes |
| 05:59 | <@Derakon> | Woot, danke. |
| 06:01 | | * Derakon eyes the algorithms Wikipedia lists, tries to decide where he wants to make the cut between implementation time and runtime. |
| 06:18 | <@Vornicus> | I've always liked the hulling method. |
| 06:19 | <@Derakon> | http://www.s-hull.org/ |
| 06:19 | <@Derakon> | Oh, hey... |
| 06:19 | <@Derakon> | There's a Python library~ |
| 06:20 | <@Derakon> | Appears to have a C dependency though. |
| 06:21 | <@Derakon> | Hm, most available solutions are not pure python, looks like. |
| 06:22 | <@Derakon> | I don't really feel that Jetblade needs to pull in all of Scipy. |
| 06:47 | | Derakon is now known as Derakon[AFK] |
| 06:47 | | cpux [cpux@Nightstar-20a84089.dyn.optonline.net] has quit [[NS] Quit: ChatZilla 0.9.86 [Firefox 3.6.8/20100722155716]] |
| 08:21 | | Tarinaky [Tarinaky@Nightstar-f349ca6d.plus.com] has joined #code |
| 08:29 | | AnnoDomini [annodomini@Nightstar-d5315a41.adsl.tpnet.pl] has joined #code |
| 08:29 | | mode/#code [+o AnnoDomini] by Reiver |
| 09:16 | | You're now known as TheWatcher |
| 09:46 | | Thaqui [Thaqui@27B34E.D54D49.F53FA1.6A113C] has joined #code |
| 10:51 | | Stalker [Z@3A600C.A966FF.5BF32D.8E7ABA] has joined #code |
| 11:31 | | Vornicus is now known as Vornicus-Latens |
| 11:38 | | * TheWatcher hrms, is actually starting to really quite like git |
| 12:22 | | Tarinaky [Tarinaky@Nightstar-f349ca6d.plus.com] has quit [Connection closed] |
| 12:23 | | Rhamphoryncus [rhamph@Nightstar-bbc709c4.abhsia.telus.net] has quit [Client exited] |
| 12:39 | | celticminstrel [celticminstre@Nightstar-f8b608eb.cable.rogers.com] has joined #code |
| 12:42 | | You're now known as TheWatcher[afk] |
| 14:32 | | You're now known as TheWatcher |
| 14:53 | | Tarinaky [Tarinaky@Nightstar-f349ca6d.plus.com] has joined #code |
| 17:11 | | Thaqui [Thaqui@27B34E.D54D49.F53FA1.6A113C] has quit [Connection closed] |
| 17:34 | | Attilla [Obsolete@Nightstar-2ec16fe2.threembb.co.uk] has joined #code |
| 17:34 | | mode/#code [+o Attilla] by Reiver |
| 18:02 | | Attilla [Obsolete@Nightstar-2ec16fe2.threembb.co.uk] has quit [Ping timeout: 121 seconds] |
| 18:11 | | Stalker [Z@3A600C.A966FF.5BF32D.8E7ABA] has quit [Ping timeout: 121 seconds] |
| 18:50 | | Attilla [Obsolete@Nightstar-4ecdef8c.threembb.co.uk] has joined #code |
| 18:50 | | mode/#code [+o Attilla] by Reiver |
| 18:54 | | Tarinaky [Tarinaky@Nightstar-f349ca6d.plus.com] has quit [Connection closed] |
| 19:42 | | Stalker [Z@Nightstar-41ddd748.0.fullrate.dk] has joined #code |
| 20:00 | | Vornicus-Latens is now known as Vornicus |
| 21:23 | | aoanla [AndChat@Nightstar-09f38038.range81-156.btcentralplus.com] has joined #code |
| 21:58 | | Derakon [Derakon@Nightstar-1ffd02e6.ucsf.edu] has joined #code |
| 21:58 | | mode/#code [+o Derakon] by Reiver |
| 21:59 | <@Derakon> | ...okay, I joined the channel to ask for help deciphering why my code wasn't working, and then I noticed that I'd mispelled "callback" and was getting an extremely unhelpful error about the required number of non-keyworded arguments. |
| 21:59 | <@Derakon> | So, nevermind~ |
| 22:09 | | Stalker [Z@Nightstar-41ddd748.0.fullrate.dk] has quit [Ping timeout: 121 seconds] |
| 22:14 | <@Vornicus> | pffff |
| 22:18 | < simon__> | TheWatcher, did you use subversion before git? |
| 22:37 | | aoanla [AndChat@Nightstar-09f38038.range81-156.btcentralplus.com] has quit [[NS] Quit: ] |
| 22:51 | | Stalker [Z@3A600C.A966FF.5BF32D.8E7ABA] has joined #code |
| 23:39 | | Derakon [Derakon@Nightstar-1ffd02e6.ucsf.edu] has quit [[NS] Quit: Leaving] |
| 23:44 | <@TheWatcher> | simon__: cvs, then subversion, then bazzar and mercurial, then git |
| 23:44 | <@TheWatcher> | In pretty much that order, with overlaps |
| --- Log closed Fri Aug 13 00:00:33 2010 |